Linköping University Medical Dissertations

No. 951

Nutritional Intervention in Elderly People

Admitted to Resident Homes

Kerstin Wikby

Department of Medicine and Care, Division of Internal Medicine, Faculty

of Health Sciences, Linköping University, SE-58185 Linköping, Sweden

Linköping 2006

Nutritional Intervention in Elderly People Admitted to Resident Homes

© Kerstin Wikby

ISBN 91-85497-87-8

ISSN 0345-0082

Printed in Sweden by Unitryck

Linköping 2006

To my family To my family To my family To my family

Wikby K. 2006. Nutritional Intervention in Elderly People Admitted to Resident Homes.

Linköping University Medical Dissertation No. 951, Department of Medicine and Care,

Division of Nursing Science, Faculty of Health Sciences, Linköping University, SE- 581 85

Linköping, Sweden. ISBN 91-85487-87-8. ISSN 0345-0082

ABSTRACT

The aim was to investigate the effects of an intervention, based on education given to staff

and implementation of an individualized nutritional programme given to the residents, to

compare assessments on admission with a previous study, and to perform diagnostic test and

inter-rater reliability of the Mini Nutritional Assessment (MNA). A further aim was to

identify and describe factors with regard to appetite among the residents.

Upon admission, and after a four month intervention period, residents were

classified as being either protein energy malnourished (PEM), or not, based on anthropometry

and biochemical measurements. On both occasions, the Activity Index and the Mini Mental

State Examination were used. In order to identify individuals in need of nutritional care, the

MNA was performed. Information about medical data was obtained. A total of 127 residents

were consecutively admitted to eight resident homes in a municipality in Sweden. Three

resident homes constituted the experimental unit (n = 68) and five the control unit (n = 59).

Fifteen residents were interviewed using a qualitative method, to investigate what affects their

appetite.

On admission 32 % of the residents were classified as PEM, which was similar to in

the previous study. A higher frequency of residents in the present study had severe medical

diseases and cognitive impairment, compared with the previous study, indicating changed

admission criteria in the present study.

Between the experimental and the control groups, no differences were seen in any

specific anthropometric or biochemical variable. Within the groups, statistically significant

differences were seen, as the number of PEM residents in the experimental group decreased,

and motor activity and overall cognitive function improved. In the control group, motor

activity deteriorated. This indicates that the intervention improved nutritional status and

functional capacity in the residents.

Diagnostic sensitivity was 73 % regarding MNA versus PEM, and 89 %

regarding MNA short form (MNA-SF) versus MNA, which indicates a rather high degree of

sensitivity in both tests. Inter-rater reliability of MNA, carried out by simultaneous

assessments by registered nurses and researcher showed a moderate agreement of 62 %

(kappa 0.41).

The interview study showed that the willingness to eat was what affected the

residents´ appetite. The willingness to eat contains internal factors, dependent on mood and

personal values, as well as external factors, dependent on wholesomeness, food, eating

environment and meal fellowship. When planning and realizing residents´ nutritional care,

factors affecting the residents´ appetite have to be taken into consideration.

In conclusion, the results show that it is important to implement and develop

strategies for individual nutritional care, in order to prevent and treat malnutrition in elderly

people, which is in line with recommendations given by the European Society of Parenteral

and Enteral Nutrition (ESPEN) and with the Swedish goal of nursing actions.

Keywords: Elderly people; resident homes; comparison; nutritional intervention; education;

individualised care; inter-rater reliability; appetite, qualitative method

ORIGINAL PAPERS

This thesis is based on the following papers, which are referred to in the text by the Roman

numerals given below:

I. Wikby K, Ek A-C & Christensson L. Nutritional Status in Elderly People Admitted to

Community Residential Homes: Comparisons between two Cohorts. (2006) Journal of

Nutrition, Health and Aging 10: 232-238.

II. Wikby K, Ek A-C & Christensson L. Implementation of a Nutritional Programme in

Elderly People Admitted to Resident Homes. (Submitted for publication).

III. Wikby K, Ek A-C & Christensson L. The two-step Mini Nutritional Assessment

Procedure in Community Resident Homes. (Submitted for publication).

IV. Wikby K & Fägerskiöld A. The Willingness to eat. An investigation of appetite among

elderly people. (2004) Scandinavian Journal of Caring Sciences 18: 120-127.

Reprints were made with permission from the publishers

CONTENTS

INTRODUCTION 1

BACKGROUND 3

Ageing 3

Circumstances influencing eating 4

Assessment of nutritional status 7

Municipal care 9

Nutritional care 10

Hunger and appetite 12

AIMS OF THE THESIS 13

METHODS 14

Participants 15

Basal characteristics 16

Objective nutritional assessment (I, II) 17

Mini Nutritional Assessment (MNA) (I, III) 18

Assessment of functional capacity (I, II) 19

Assessment of overall cognitive function (I, II) 20

Nutritional programme (II, III) 21

Interview study (IV) 22

Statistics (I-III) 23

Analysis of qualitative data (IV) 24

Ethical considerations 25

RESULTS 26

Characteristics of residents in cohort 2 upon admission 26

Comparison between cohorts 1 and 2 (I) 28

Intervention (II) 30

Test of MNA (I, III) 32

Interview study about appetite (IV) 34

DISCUSSION 35

Basal characteristics in cohort 2 36

Comparison between two cohorts 1 and 2 37

Intervention 38

Assessments 40

Interview study about appetite 43

Clinical implications 44

CONCLUSIONS 46

SUMMARY IN SWEDISH 47

ACKNOWLEDGEMENTS 49

REFERENCES 51

ORIGINAL PAPERS I-IV

1

INTRODUCTION

An important and burning problem is that approximately one third of residents in municipal

resident homes in Sweden suffer from malnutrition (Saletti et al 2000; Christensson et al

2002; Suominen et al 2005). In this context, malnutrition refers to a state of undernutrition, so

called protein- energy malnutrition (PEM) (Keller 1993). It is not only a problem for the

individual but also for the family, community, and the health care system (Visvanathan 2003).

The reserves of elderly people are often reduced with respect to physiological, psychological,

and social factors. This may affect their life and makes them vulnerable. They are at increased

risk of having decreased appetite, insufficient dietary intake and, through this, are also at risk

of developing malnutrition (Morley 2001, 2002). Malnutrition is strongly associated with

declined physical (Unosson et al 1991; Gazzotti et al 2000) and cognitive ability (Ödlund et al

2005), worsening of existing chronic diseases (Akner & Cederholm 2001; Seiler 2001) and

may deeply affect quality of life (Vetta et al 1999; Crogan & Pasvogel 2003). It is associated

with decreased immune reactions (Ek el al 1990; Hudgens et al 2004), presence of pressure

sores (Ek et al 1991; Hudgens et al 2004) and increased mortality rates (Christensson et al

1999; Van Nes et al 2001). Elderly people in general have decreased ability to regulate food

intake, which may contribute to difficulties in compensating for insufficient food intake and

thus to weight loss (Roberts et al 1994; Roberts 2000; Das et al 2001). If the problem is not

observed, they run the risk of their nutritional status deteriorating. Hence, it is important to

use a reliable and valid tool to assess nutritional status (Omran & Morley 2000a; Guigoz et al

1999).

Since a reform in 1992, the municipalities in Sweden have the responsibility for service and

care given to elderly people (The National Board of Health and Welfare 1996), and offer

assistance in elderly people´s own homes, or in resident homes (Sjölenius 1997). The guiding

principles of the reform focus on autonomy, privacy, safety and freedom of choice (The

National Board of Health and Welfare 1996). In Sweden, approximately 17 % of the people

are 65 years of age or older. Five per cent are 80 or older (Statistics Sweden 2002), and

among them 40 % are in need of home service and nursing care. In December 2000, 127 000

residents in Sweden lived in resident homes, which was 10 000 less than in 1996 (The

2

National Board of Health and Welfare 2000). During that period, the number of elderly

people in Sweden remained unchanged (Statistics Sweden 2002).

In order to prevent and treat malnutrition, a set of standards for practical use within the

present health care resources are recommended by the European Society of Parenteral and

Enteral Nutrition (ESPEN). A number of actions are suggested. Screening tools should be

used to identify individuals at risk of malnutrition. Assessments of metabolic, nutritional and

functional variables should be performed, in order to make a diagnosis, which should lead to

appropriate care plans. The effectiveness of the care plans should be monitored, the results

should be disseminated and the process of the care plans should be conducted in a systematic

manner, in order to allow audit of the outcomes (Kondrup et al 2003).

In municipal resident homes actions directed toward malnutrition can be implemented by

giving education to the staff as these actions may have positive effects on the residents´

nutritional status. However, few studies include an examination of these effects of nutritional

education given to the staff and the results are not unambiguous (Christensson et al 2001;

Faxén-Irving et al 2005). Consequently, there is a need for further studies on staff education

within municipal resident homes, in order to examine the effects on the residents´ nutritional

status.

According to Whitney et al (2002) and Morley (2002), eating is a complex process driven by

hunger and appetite. Hunger is described as a strong desire or physiological need for food,

while appetite refers to the enjoyment of eating and to food itself. Since decreased appetite is

seen as a main factor of decreased food intake and malnutrition (Morley 2002), and no studies

have been found where elderly people view their appetite, it is important to let elderly people

describe what affects their appetite in order to gain a deepened understanding of the subject.

3

BACKGROUND

Ageing

During ageing, body composition undergoes changes, from increasing body weight in the first

part of life, which stabilises in late middle age, and decreases in very old people (Dey et al

2001). Seventy five years of age seems to be a turning point, when weight starts to decrease

(Perissinotto et al 2002), and the weight reduction seems to depend mainly upon loss of fat

free mass (Huges et al 2002). According to a review by Bozzetti (2003), there is a gradual

decrease in total body water from adulthood to old age, until it constitutes less than 50 % of

body weight, i.e. below the normal range of from 50 to 65 %. There is also a successive

decrease in the reserve capacity of bodily functions, such as of the capacity of the heart

(Bozzetti 2003) and the immune system (Fulop et al 2005). The individual differences are big,

and there are also variations between different functions (Bozzetti 2003). The reduced

functions decrease the reserves in elderly people (Morley 2001), so that with greater age there

is a higher incidence of chronic illnesses associated with increased metabolic processes, such

as obstructive pulmonary disease, heart failure and rheumatic arthritis (Tracy 2003, Goldspink

2005).

With advanced age, a severe reduction in the flow of saliva is observed (Nagler &

Hershkovich 2005), as well as a reduced number of natural teeth, and decreased stability and

retention of prostheses (Lamy et al 1999). Other problems seen in greater age are difficulties

in swallowing (Wilkingson & de Picciotto 1999; Lamy et al 1999; Locker 2002) and

decreases in the senses of taste (Bozzetti 1993) and smell (Wang et al 2005). Disabilities due

to diseases, such as stroke, are also more common in elderly people than in those who are

younger (Perry 2003).

Mental disorders, such as dementia disease and depression are found in approximately 30 %

of elderly people and have also been seen to increase with ageing (Skoog 2004). During

ageing, the cognitive reserve declines and may result in a reduced capability in functions such

as orientation, memory, abstract thought and perception (Cullum et al 2000). This means that

elderly people need more time to become acclimatised to new circumstances (Couture et al

2005). In later life, there is a continuum from normal cognitive ageing to overt dementia.

4

Changes due to dementia involve reduction in brain capacity, which may result in limitations

in understanding and interpreting the surroundings (Whalley et al 2004). Depression is an

emotional state characterised by inappropriate feelings, such as marked sadness, or a loss of

interest or pleasure in daily activities (Friedlander et al 2003).

Loneliness is common in elderly people and seems to derive from societal life changes, such

as the ability to maintain a social network and from negative life events (Skoog 2004; Savikko

et al 2005). Conditions such as poor health status, poor functional status, poor vision, and loss

of hearing, also increase the prevalence of loneliness (Bickerstaff et al 2003; Savikko et al

2005). Low level of income is associated with limitations, which can lead to isolation and

loneliness, as can societal changes, such as the migration of younger people from the country

to urban areas (Savikko et al 2005).

Because of changed conditions, such as disability or restricted money, elderly people may

have to move from one location to another. This changes their life situation and they have to

adjust to new surroundings. This can involve a painful process, which can produce lasting

trauma (Johnson & Tripp-Reimer 2001). This may, in turn, result in a decline of physical

vitality, depression, cognitive impairment, and malnutrition (Robertson & Montagini 2004).

Circumstances influencing eating

Eating is basic for all human beings, to bring the body nutrients, and to edify and regulate

fundamental bodily functions (Morley 2002). A satisfying nutritional status is a necessary

condition for maintaining health and avoiding illness (Whitney et al 2002). Nutritional status

is, in this thesis, defined as the outcome of the balance between intake and consumption of

nutrients.

Malnutrition is a condition which may adversely affect quality of life in elderly people (Vetta

et al 1999; Crogan & Pasvogel 2003). According to Nordenfelt (1991), quality of life is very

subjective and is, in most cases, based on conceptions such as harmony, satisfaction,

wellbeing and happiness. These conceptions include physical and psychological ability, as

well as social interaction, activity, material status and experienced health status. Sarvimäki &

5

Stenbock-Hult (2000) have defined quality of life in elderly people as a sense of wellbeing,

meaning, and value or self-worth.

The reason why elderly people develop malnutrition is complex according to a review by

Elsner (2002). It can be associated with increased vulnerability and susceptibility to strain,

due to decreased physical, psychological and/or social functions (Morley 1996, 2002). This

may affect eating behavioural, which includes difficulties in eating and in maintaining

sufficient nutritional status (Elsner 2002). Eating difficulties have been defined by Westergren

et al (2001, p 258) as "difficulties that, alone or in combination, negatively interfere with the

preparation and intake of served food and/or beverages."

Poor oral health is associated with eating difficulties (Walls & Steele 2004) and with

malnutrition (Chen et al 2005). Difficulties in chewing, because of poor dental status (Lamy

et al 1999; Andersson et al 2002), as well as dry mouth, red mucus membranes, and oral

sores, can greatly affect eating and nutritional status in elderly people (Andersson et al 2002).

Elderly people who have suffered strokes have often increased difficulties in swallowing,

because of paresis (Terré &Mearin 2006), and are at high risk of insufficient dietary intake

and malnutrition (Kagansky et al 2005). Stroke may also include decreased impaired self-

feeding ability related to e.g. impairment in the arms and visual deficits (Westergren et al

2001, 2002; Perry & McLaren 2003). Changes in the senses of taste and smell may affect

dietary intake in elderly people (Schiffman & Graham 2000).

Because of diseases, many elderly people have prescription medications, which are associated

with loss of appetite and weight loss (Morley 1996, 1997; Omran & Morley 2000a). The

changed proportion of fat and water content in the ageing body may lead to changed effects in

the distribution of medications, thus intensifying their effects. These effects may impede

intellectual and physical abilities (Turnheim 2004) and, as a result, eating. Elderly people

often have a decreased sensation of thirst and reduced fluid intake (Kenney & Chiu 2000).

Dehydration is associated with impaired cognitive ability (Wilson & Morley 2003).

Failure to remember to eat, and how to eat, are common eating difficulties in elderly people

with dementia disease (Morley 1997). Dementia disease has been seen as a risk factor for

malnutrition in elderly people (Kagansky et al 2005). Decreased enjoyment of eating and of

6

food, as well as refusal to eat, are common problems in elderly people with depression, and

are commonly associated with malnutrition (Morley 1996, 1997; Chen et al 2005).

Refusal to eat can indicate an indirect self-destructive behaviour, which has been defined as

"an act of omission or commission that causes self-harm leading indirectly over time to the

patient´s death" (Conwell et al 1996, p 153). It involves self-injurious eating and drinking

patterns, abuse of medication, rejecting medical and nursing advice, and conflict interactions

with other individuals (Conwell et al 1996). Indirect self-destructive behaviour has been seen

in more than half of the residents studied in 25 nursing homes in Australia (Draper et al

2002).

Bereavement is a risk factor for poor nutritional status, since many elderly people are

widowed. It imbues many areas of life functioning related to eating, and is for that reason

likely to lead to a reduction in nutritional intake, weight loss and malnutrition (Browne et al

1997). Loneliness and social isolation have also been seen to be causes of weight loss in

elderly people (Locher et al 2005), and low income has been seen to impact upon nutritional

health in elderly people living in their own homes (Chen et al 2005).

Eating in elderly people involves interaction with others social and cultural manners, which

are based on traditions and norms. When elderly people became disabled, their ability to

behave according to the norms they have learned decreases, e.g. the management of food and

objects on the table. They feel ashamed of themselves and simplify the procedures at the table

or avoid situations they think are embarrassing. When elderly people are served food they do

not like, such as unfamiliar dishes or different cooking, it highly reduces their wellbeing and

their satisfaction at mealtimes (Sidenvall et al 1996). Eating in elderly people is affected by

the meal environment. In a group of patients on a long-term care ward, food intake increased

by 25 % when the dining-room was redecorated, reaching levels similar to what was common

when the patients were young. When the dining-room was arranged back to its original state,

the patients´ food intake decreased again (Elmståhl et al 1987).

When elderly people have to move from one location to another, such as a resident home,

they have to adapt to new circumstances (Johnson & Tripp-Reimer 2001). This can lead to

inactivity, insufficient food intake and weight loss (Robertson & Montagini 2004). The reason

why elderly people have to move into a resident home is because of a non-functioning living

7

situation (Christensson et al 1999), and they have not always been involved in, and thus made

the decision to move (Johnson & Tripp-Reimer 2000). All this, taken together, can be

suggested to involve increased risk factors for eating problems. Saletti et al (2000) have

shown that malnutrition is common among elderly people living in resident homes.

Assessment of nutritional status

Routine nutritional assessment of vulnerable elderly people is the first stage in implementing

a nutritional programme (Todorovic 2001). In order to receive true assessments of nutritional

status, it is important to pay attention to reliability and validity of the tool used (Polit & Beck

2003). The purpose of nutritional assessments is further to establish baseline values, which the

effectiveness of nutritional intervention could be measured against (Omran & Morley 2000b).

Many different methods have been used to assess nutritional status in elderly people, based on

objective nutritional assessments, routine history and physical examination, evaluation of

dietary intake, and assessment tools (Omran & Morley 2000a). At present, there is no gold

standard for nutritional assessment (Berner 2003), which is probably due to the complexity of

the phenomenon of nutritional problems in elderly people.

Objective nutritional assessments, such as anthropometric measurements in combination with

biochemical variables, are often used in nutritional assessments (Larsson et al 1990; Unosson

et al 1995; Christensson et al 1999). Commonly used anthropometric measures are body

weight, Body Mass Index (BMI = weight in kg divided by height in m 2) (Garrow & Webster

1985; Flodinet al 2000; Weigly 1994), weight index in per cent (WI) (Warnold & Lundholm

1984; Bengtsson et al 1981), as well as arm measurements, such as triceps skinfold thickness

(TSF), midarm circumference (MAC) and the calculation of midarm muscle circumference

(AMC) (Symreng 1982).

Serum proteins, such as albumin and transthyretin (also known as prealbumin), are

biochemical variables, which are often used in nutritional assessments, since the serum

protein levels reflect the protein intake (Pirlich & Lochs 2001). Transthyretin shows a much

higher degree of sensitivity to the nutrient intake than albumin does, which is related to its

shorter biological half-life (2 days compared to 20 days) (Ingenbleek & Young 2002). When

protein intake decreases, or during starvation, the synthesis of serum albumin and

8

transthyretin in the liver may decrease to values below normal. Low serum albumin and

transthyretin values are also associated with liver and kidney diseases (Whitney et al 2002), as

well as with chronic inflammation and infections (Ingenbleek & Young 2002).

In Sweden, many studies have used ´objective nutritional assessments´, based on

anthropometric and biochemical measures, to classify an individual as either PEM or non-

PEM (Ek et al 1990; Unosson et al 1991; Christensson et al 1999). An individual is defined as

being PEM if at least one anthopometric and one biochemical value are subnormal. The

reference ranges for subnormal values emerge from a Swedish sample of elderly people

(Bengtsson et al 1981; Symreng 1982; Warnold & Lundholm 1984).

In order to evaluate the risk of malnutrition in frail elderly people and to identify those who

could benefit from early intervention, the Mini Nutritional Assessment (MNA) tool has been

developed (Guigoz et al 1996). The aim of the development of MNA was to develop a valid

and reliable screening tool, to facilitate nutrition intervention in clinical practice. It comprises

18 items, divided into anthropometric, general, dietary, and self assessments. The sum of the

MNA score distinguishes between elderly people with malnutrition, at risk of malnutrition,

and those who are well nourished. In many Swedish studies, the MNA tool has been used to

evaluate nutritional status in elderly people in various municipal resident homes (Saletti et al

2000; Christensson et al 2002; Ödlund et al 2005). The European Society of Parenteral and

Enteral Nutrition (ESPEN) has recommended the MNA tool to detect the presence of

malnutrition and risk of malnutrition among elderly people in home-care programmes,

nursing homes and hospitals (Kondrup et al 2003). The aim of the MNA is to identify those at

risk of developing malnutrition at an early stage, where it can be easily corrected by

nutritional intervention (Vellas et al 1999; Guigoz et al 1999). The tool is easy to administer

and reduces the need for more invasive tests, such as blood sampling (Rubenstein et al 1999,

2001).

It is assumed that the MNA tool can be completed in less than 15 minutes. However, this has

been seen as being too long thus limiting its usefulness in geriatric screening situations

(Guigoz et al 2002). Consequently, the authors developed a short form of the MNA (MNA-

SF), to be used in a two-step procedure. The first step includes MNA-SF, which is compiled

by using six of the total 18 MNA items and takes approximately three minutes to perform.

The score of MNA-SF indicates either well nourished or risk of malnutrition (Rubenstein et al

9

1999). In the latter case the second step is carried out, using the remaining MNA items

(Rubenstein et al 2001).

The MNA-SF has been found to be an appropriate screening tool in elderly patients within

acute care (Ranhoff et al 2005), and in elderly people with pressure ulcers in nursing homes

(Langkamp-Henken et al 2005), but not in middle aged patients undergoing surgery

(Putwatana et al 2005). Several studies have been found where the two-step MNA procedure

has been used, indicating it to be a useful tool to identify elderly patients with malnutrition or

at risk of malnutrition (Cohendy et al 2001; Kuzuya et al 2005; Langkamp-Henken et al

2005).

Municipal care

When elderly people are in need of care, they are offered assistance in their own homes, or in

special types of municipal resident homes, with service at different levels according to

individual requirements (The National Board of Health and Welfare 1996). Accommodation

in special types of resident homes includes service buildings, retirement homes, group living

for people with dementia disease, and nursing homes (Sjölenius 1997). In service buildings,

the residents mostly have mild dys-functions and receive home help service and home

medical care as required. In retirement homes, the residents need relatively extensive support,

such as assistance with eating. Group living is co-housing accommodation for approximately

8 to 10 residents with dementia disease. In nursing homes, the residents need extensive

special care, since common diagnoses are stroke, malignancy, dementia and cardiovascular

diseases. In order to decide the right level of resident home, the decision is made in

accordance with the resident´s wishes and ability to manage his or her daily life, and is

managed by a home help administer (The National Board of Health and Welfare 2000).

In the late 1990s, the care of elderly people gradually changed. Short-term stay was

established, as well as palliative care given to residents in the final stage of life being

introduced. The aim of short-term stay is to offer elderly people support, so that they can

continue to live in their own homes as long as possible. In order to achieve this, temporary

care is given to residents in need of further rehabilitation after hospital care, and to residents

whose next of kin need to receive temporary support. When elderly people enter special types

10

of housings, they are mostly in great need of comprehensive care day and night, and the

assistance offered in their own homes does not fulfill their needs any longer. Between 1996

and 2000, the number of accommodation places in special resident homes decreased by 10000

in Sweden (The National Board of Health and Welfare 2000).

In resident homes, registered nurses (RN) are responsible for giving safe and appropriate

nursing care of good quality (SOSFS 1997:10). The RN´s working tasks include planning,

realising or delegating, leading, documenting, and evaluating nursing care. They include

responsibility for the residents´ nutrition, and especially for those residents who have

difficulties in meeting their own nutritional needs. Medical needs are supplied by the

resident´s general practitioner, employed by the County Council (The National Board of

Health and Welfare 2000).

Nutritional care

As the reasons for decreased appetite, insufficient food intake and malnutrition in elderly

people are multi factorial, elderly people with nutritional problems constitute a very

heterogeneous group (Morley 2002). In every individual identified as at risk of malnutrition,

an appropriate care plan should be carried out, which should be based on individual´s desires,

needs and resources (Kondrup et al 2003, SOSFS 1993:17).

In studies designed to improve nutritional status in elderly people, the focus has often been on

standardised actions, such as oral supplements in addition to regular meals. In a Cochrane

database systematic review of randomised, controlled trials (RCT) by Milne et al (2005), the

evidence from trials for improvement in nutritional status and clinical outcomes were

examined. The review included 49 trials involving 4790 randomised elderly people, where

extra protein and energy were provided, usually in the form of commercial ´sip-feeds´. The

results showed that oral supplementation in general produced a small but consistent weight

gain in elderly people. In a majority of the trials, a reduced mortality rate was also observed in

the supplemented group, as compared with the control group. However, there was no

evidence of improvement in functional benefit or reduction in length of hospital stay in the

supplemented group. This review points out that the results of standard actions towards

nutritional problems in elderly people are not unambiguous (Milne et al 2005).

11

When following the recommendations given by ESPEN (Kondrup et 2003), staff education is

suggested to be a basic pre-requisite for giving the residents appropriate nutritional care.

In two Swedish studies (Faxén-Irving et al 2005, Christensson et al 2001), education was

given to the staff in resident homes, in order to improve the nutritional status of the residents.

The education involved identification of underlying causes of eating difficulties and

malnutrition, to fulfil individual nutritional requirements, as well as instructions, such as what

to serve. The intention was that the staff should apply what they had learned to nursing

practice. One study (Faxén-Irving et al 2005) was designed as a RCT study, where the

intervention group included 37 residents and was based on various consistencies of food and

drinks, as well as of enriched drinks. After five months, no objective changes were seen,

either in the intervention or in the control group. Another study (Christensson et al 2001) had

single-case design and involved 11 malnourished residents, and was based on individually

adjusted meals, using ordinary food. After three months, the residents´ nutritional status and

functional capacity improved in general. The results of these studies, point out a need to

further investigate whether education given to staff, and individual nutritional care based on

ordinary food, have effects on residents´ nutritional status.

Present health care resources (such as ordinary food), are recommended when preventing and

treating malnutrition (Kondrup et al 2003). Serving ordinary food instead of commercial ´sip-

feeds´ should be a key component within nutritional care. The importance of ordinary food for

elderly people in resident homes is shown in a study by Lengyel et al (2004), where 205

elderly residents in long-term care expressed concern over food variety, quality, taste, and

appearance, as well as with the menus. The residents in that study were less satisfied with

areas related to their autonomy, such as food choice and snack availability. Another study by

Evans et al (2005), including 20 nursing home residents showed that food, which reflects the

family background of the individual, is a source of comfort in nursing home residents. Food

can play an important part in recovery from illness or adaptation to nursing homes. That study

states that individualised nutritional care, based on food can promote nutritional status and

quality of life in nursing home residents (Evans et al 2005). When implementing of

individualised nutritional care in resident homes, the RNs have to take an active part, like the

researcher did in the study by Christensson et al (2001). The RNs have to be responsible for

identifying residents in need of nutritional attention and for carrying out individualised

nutritional care.

12

Hunger and appetite

Since decreased appetite is seen as a main factor of decreased food intake and malnutrition in

elderly people (Morley 2002), it is important to gain a deeper understanding about appetite.

Hunger and appetite initiate eating, and can be seen as parallel and interacting functions

(Whitney et al 2002). Complicated feed back systems, transmitted by the nervous system

regulate hunger and satiety and the nutrient intake (MacIntosh et al 2000, Morley 2001).

Eating is also initiated by appetite, which can be good or poor. It is based on a psychological

phenomenon that makes it possible to experience good appetite without hunger, such as when

someone offers an unexpected treat like a delicious dessert after a meal. In other situations,

the opposite may occur, such as when an individual is faced with a stressful situation, thus

hunger can exist but appetite is poor (Whitney et al 2002). Beliefs may also overcome hunger,

such as when persons are starving themselves to death in order to make a political point

(Morley 2002). A reduced food intake causing weight loss may then be due to physiological,

psychological or social factors, or a combination of all of these (Morley 2001, 2002; Donini et

al 2003).

Many research projects focus on eating strictly from a physiological point of view, and in that

context, appetite is described as a physiological drive to eat and to regulate energy balance (le

Roux & Bloom 2005; Mattes et al 2005; Erlanson-Albertsson 2005). In other studies, hunger

and appetite are described as interacting functions, but with focus on the physiological drive

to eat (Rogers 1999; Kristensen 2000; Morley 2002). No study has been found focusing

mainly on appetite from a psychosocial point of view. This makes it of importance to carry

out studies which elucidate the psychosocial influences on appetite, in order to gain a deeper

understanding of how elderly people view their appetite and what affects their appetite.

13

AIMS OF THE THESIS

The general aim of this thesis was to investigate the effects of an intervention, based on

education given to the staff and implementation of an individualized nutritional programme

given to the residents.

The specific aims were:

- to examine nutritional status and socio-demographic and medical data, in people who

were newly admitted to municipal resident homes. A further aim was to compare the

results with a previous study performed in the same municipality (I).

- to test the hypothesis that education given to the staff will improve nutritional status, as

measured by anthropometric and biochemical variables, as well as functional capacity in

elderly people newly admitted to municipal homes, as compared with a control group (III)

- to test internal consistency and inter-rater reliability of Mini Nutritional Assessment

(MNA) during implementation of MNA, and to test sensitivity, specificity and diagnostic

predictivity of MNA-short form (MNA-SF) in relation to MNA (II)

- to identify and describe factors of importance with regard to appetite among elderly

people (IV)

14

METHODS

This thesis consists of four studies of residents, newly admitted to resident homes in a

municipality in the southern part of Sweden. Two approaches are used, based on paradigms

with roots in two different scientific traditions i.e. quantitative and qualitative. The

quantitative approach focuses on the universality in a group of residents and data is collected

using deductive methods. One study is on nutritional status and functional capacity in newly

admitted residents, and includes comparisons between two cohorts (I). In one another study,

an individualized nutritional programme in newly admitted residents is tested (II). In the third

study, the two-step MNA procedure is tested in municipal resident homes (III).

A qualitative approach focuses on the individual and unique phenomenon and the data is

collected inductively in the residents´ natural context. One study is on how residents view

their appetite and how they describe factors affecting their appetite from a qualitative

perspective (IV).

Internal validity refers to the extent the experimental treatment can be inferred to be

responsible for the observed effect, and is obtained by control mechanism. Due to ethical and

practical concerns regarding the residents, a true experimental design was not possible to

carry out in study II, and a quasi-experimental design was used. Apart from random

assignment, the design was the same as with an experimental design (Polit & Beck 2003).

This thesis is part of a research project with the aim of assessing and describing nutritional

status, of identifying factors associated with nutritional issues and of investigating the effects

of individualised nutritional care, in elderly people newly admitted to resident homes in

Sweden. The project has been ongoing since 1996 in a municipality in the southern part of

Sweden (Christensson et al 1999), with mainly rural work and industries.

When the research project started in 1996, approximately 27,500 people lived in the

municipality. Four years later, the number was 26 600. Approximately half of the inhabitants

in the municipality lived in the urban district. The municipality comprised a total of 14

resident homes, spread out over an area of 1500 km2, where eight resident homes were

involved in the project. In 1996, those eight resident homes comprised 334 beds, while the

15

number of beds in 2000 was 277. During the time period between the two cohort studies,

palliative care given to residents in the final stage of life was established, as well as, short-

term stays being introduced.

Participants

Residents, 65 years or older, newly admitted to eight resident homes in the municipality

described above were consecutively included. Exclusion criteria were terminal stage,

malignant diseases and kidney and liver diseases. From October 2000 to April 2002, 134

residents entered the resident homes involved. Of these residents, six did not wish to

participate and one died unexpectedly before the examination was performed. One hundred

and twenty-seven residents were included (I, II, III). Fifteen residents included in the

interview study were recruited from this group (IV).

Study I. The participants were recruited from 208 residents included in a previous study

carried out from October 1996 to October 1997 (cohort 1), and from the group of 127

residents included from October 2000 to April 2002 (cohort 2). Both cohorts were obtained

from the same municipality, with the same resident homes and the same definitions of

malnutrition. The same inclusion and exclusion criteria were used in both cohorts. The

previous cohort study, however, had also included residents admitted to service buildings.

When comparing the two cohorts in this thesis, these residents were excluded in order to

make the cohorts comparable. In cohort 2, fewer residents were admitted to the resident

homes during the 12 months, and consequently fewer were included as compared with cohort

1. To increase the number of residents, the study continued for another six months.

Study II. Three and five resident homes involved constituted the experimental and control

units, respectively. The experimental and the control units included 135 and 142

accommodation places, respectively, and were similar regarding workload, type of nursing

care, nursing competence and number of staff. Sixty-eight residents entered the experimental

and 59 the control unit. Each of the residents was in the study for a four month period. During

the study period, five residents died and one dropped out because of newly diagnosed

malignant disease, in each group. In the experimental group, 62 residents (20 men, 42

women) completed the study. The corresponding number of residents in the control group was

16

53 (14 men, 39 women). In order to test the hypothesis adequately, a power calculation based

on body weight was performed. The calculation was based on the results from an earlier

study, where the same inclusion and exclusion criteria were used to assess nutritional status.

The calculation showed that 20 residents would be sufficient in each group for showing

statistically significant differences regarding body weight. Since the aim also was to

investigate functional capacity, the number was increased to 50 in each group.

Study III. In three of the resident homes involved, called A, B and C, parallel identical

nutritional assessments were performed by the researcher and registered nurses (RN),

employed in those resident homes. Two RNs were employed in resident home A, one in B

and one in C. The RNs´ burden of work, was judged to be equal, since the RNs in resident

home A also had responsibility for accommodation places not included in this study. Resident

homes A, B and C contained in the indicated order, 36, 49 and 50 accommodations. Fourteen

residents were admitted to resident home A, and 22 and 32 to resident homes B and C,

respectively.

Study IV. Sixteen residents were recruited from the group of 127 newly admitted residents.

They were asked to take part in the study after they had finished the four month study period

in study III. Fifteen accepted and one refused, due to lack of interest. The residents were also

selected regarding gender, body shape, earlier work, marital status, dental status and ability to

walk. The interviews were carried out from February 2002 to July 2002 and took part in the

residents´ normal settings, at the resident homes.

Basal characteristics

During the two first weeks after admission, the residents or a next of kin, answered questions

regarding the type of dwellings the residents moved from (I), and details of medical diagnosis

and prescription medications were obtained from the residents´ medical records (I,II). Medical

diseases were classified according to the International Classification of Diseases (ICD-9), and

prescriptions according to the Anatomical Therapeutic Chemical Classification system

(ATC).

17

All residents were assessed on three occasions: Upon admission (I - III), after two (II), and

after four months (II). Upon admission, residents were assessed during their two first weeks

and the assessment included the following procedure: It started with anthropometric measures

(I, II), and with MNA (I, III). Functional capacity (II) and overall cognitive function (I, II)

were assessed. Finally biochemical measurements were collected (I, II). Two months after

admission, anthropometric and biochemical measurements were carried out (II). Four months

after admission, the same assessment procedure as on admission was carried out (II). When

the assessment procedure had been completed after four months, 15 residents were

interviewed (IV).

Objective nutritional assessments (I, II)

In order to classify the residents as either PEM or non-PEM, body weight, height, mid-arm

circumference (MAC), triceps skinfold thickness (TSF), arm muscle circumference (AMC),

serum albumin and transthyretin, were used (Table 1).

Table 1. Criteria used to determine protein- energy malnourishment (PEM). A resident was classified as being PEM if two or more of the nutritional variables were subnormal, including at least one anthropometric and one biochemical measurement.

______________________________________________________________________________ Men Women

______________________________________________________________________________ Anthropometry WI < 80 % < 80 % TSF ≤ 6 mm ≤ 12 mm AMC ≤ 79 years ≤ 23 cm ≤ 19 cm > 79 years ≤ 21 cm ≤ 18 cm Serum proteins Albumin < 36 g/l < 36 g/l Transthyretin < 0.23 g/l < 0.23 g/l _____________________________________________________________________________ WI = Weight Index; TSF = triceps skinfold thickness; AMC = arm muscle circumference

18

Height was estimated with the resident in a supine position on a flat bed. It was measured to

the nearest cm, using a measuring instrument with a fixed foot plate and an adjustable head

plate. Each resident´s body weight was measured to the nearest 0.1 kg by a mechanical

balance chair. Weight index (WI), in percent, was calculated from actual weight on admission

divided by the reference weight and multiplied by 100. The reference weight was calculated

to be 0.80 x height (cm) - 62.0 (kg) for men and for women 0.65 x height (cm) - 40.4 (kg)

(Bengtsson et al 1981, Warnold & Lundgren 1984). According to Pearson´s correlation

analysis, the correlation coefficient between WI and BMI was 0.99, and a WI of 80 % was

equivalent to a BMI of approximately 20.0.

Mid-arm circumference was measured to the nearest 0.1 cm with a measuring-tape and TSF,

to the nearest mm, with a Harpenden skinfold calliper at the midpoint of the upper arm

between the process of acromion and olecranon (Symreng 1982). The mean of three

measurements was used and the non-dominant arm was measured, unless the arm was

paralysed or otherwise injured. Arm muscle circumference was calculated: AMC (cm) =

MAC (cm) - 0.1 [π x TSF (mm)]. Serum albumin and transthyretin were measured, and local

reference values were used. A resident was classified as having PEM if two or more of the

nutritional variables were subnormal, including at least one anthropometric and one

biochemical measurement (Symreng 1982, Unosson et al 1995, Christensson et al 1999).

In order to obtain information about possible infections, which have been seen to decrease

serum albumin (Kalender et al 2002) and transthyretin levels (Ingenbleek & Young 2002), C-

reactive protein (CRP) was measured. C-reactive values below 10 were defined as normal,

according to local reference values, and were not specified.

Mini Nutritional Assessment (MNA) (I- III)

The anthropometric area of MNA consists of BMI, MAC, calf circumference, and weight loss

during the last three months. Information about weight loss was collected from the residents,

and from the residents´ record. The general area assessments (questions related to living,

medical use, physical and mental status, and mobility) and the dietary area assessments

(questions related to dietary intake and eating problems) were collected from the residents,

and from the residents´ records. If necessary, a next of kin, or a member of the staff who knew

19

the resident well, answered the questions. The self assessment questions (one question on

nutritional status and one on health) were collected from the residents. If the resident was not

able to answer the questions, he/she received the score for ´is uncertain of nutritional state´

and ´does not know´ about his/her health.

To assist in judgement to complete the MNA form accurately and consistency, a user guide

has been developed (MNA 2006). In the user guide, each question in the MNA is explained in

turn and the scoring described. The six MNA items mobility, having suffered psychological

stress or acute disease in the past three months, having neuropsychological problems, having

pressure sores or skin ulcers, the amount of fluid consumed per day, and mode of feeding,

were dichotomised into yes or no (I). In mobility, the resident was assessed as having inability

to walk if the resident not could get out of bed/chair. Psychological stress could arise in

connection with losing a spouse, or when one of the couple was in need of acute hospital care,

leaving the partner in municipal care. Cardiac and cerebrovascular diseases, recovering from

surgery and pneumonia, were examples of acute diseases. The following conditions led to the

residents being assessed as having neuropsychological problems: Mild to severe dementia,

state of depression or confusion. The resident was scored as having pressure sores or skin

ulceration after verifying. The amount of fluid consumed per day was assessed as insufficient

if it was less than six cups. In mode of feeding, the resident was assessed as in need of help

during meals if the resident was not able to manage eating independently. Help during meals

ranged from e.g. help with buttering a slice of bread, to complete assisted feeding, where the

resident did not participate at all during meal. The maximum MNA score is 30 points and the

sum classifies the residents in the following manner: Well-nourished ≥ 24 points (MNA 1); at

risk of malnutrition, 23.5 to 17 points (MNA 2); malnourished, < 17 points (MNA 3) (Guigoz

et al 1999, 2002).

Assessment of functional capacity (I, II)

In order to assess the residents´ functional capacity, the Activity Index (Hamrin & Wohlin

1982) was used. The Activity Index was developed to evaluate functional capacity in stroke

patients and consists of 16 variables. The total activity index score ranges from 16, reflecting

poor functional capacity, to 92 reflecting normal functional capacity. It is divided into the

three parts mental capacity, motor activity and ADL (activity of daily living) function. Part

20

one, mental capacity includes four items and comprises degree of consciousness, orientation,

communication, and psychological activity (4-32 points). The second part is motor activity,

which includes six items and comprises motor activity in left and right arm and leg (6-24

points). Part three, ADL function includes six items comprising ambulation, personal hygiene,

dressing, feeding, and emptying/function of bladder and bowel (6-36 points). The aim of the

Activity Index is to score what the resident can do, which means his/her maximum

performance. The tool has been validated in Sweden by Lindmark & Hamrin (1988), and in

the United States by Chong (1995). It has been used in several Swedish studies (Lindmark

1988, Elmståhl et al 1996, Christensson et al 2001).

Assessment of overall cognitive function (I, II)

In order to screen for overall cognitive function, the Mini Mental State Examination (MMSE)

was used. It is a brief screening tool developed by Folstein et al (1975), and attempts to

quantify the individual´s orientation, short-term memory, attention, language, comprehension,

writing and a visual task. Mini mental state examination has often been used, both

internationally (deSilva & Gunatilake 2002, Tognoni et al 2005) and nationally (Aevarsson &

Skoog 2000). The test takes approximately 5-10 min, but can take longer with those who are

mildly impaired or who have auditory and/or communication difficulties.

The maximum and best score of the tool is 30 points. In the original paper (Folstein et al

1975), 20 points or less were the cut off score in order to be classified as having cognitive

impairment. The authors have subsequently recommended different cut off scores, since

cognitive ability varies in the population due to age and educational level (Crum et al 1993).

Different cut off scores have also been used for classifying individuals as either cognitively

intact, moderately impaired, or demented (Beck et al 2001). In studies I and II, 20 points or

less were considered as representing cognitive impairment. Seven residents, out of 127, were

not tested according to MMSE, because of aphasia in six residents and unwillingness to

participate in one resident.

21

Nutritional programme (II, III)

The intervention comprised education and instructions regarding an individualised nutritional

programme, and was carried out in resident homes A, B and C (II). Registered nurses (RNs)

and nurse aids (NAs) attended an educational programme, especially developed for this study

and which was divided into three steps. First, all staff received information about the study

and the individualised nutritional programme. Every single staff member received the book

"Food and dietary management in elderly people - Problems and opportunities" in Swedish

(The Swedish National Food Administration 1998) (II, III).

Second, all RNs (n = 4) and NAs, with tasks of acting as dietary ombudsmen (NAO) (n = 3),

were trained how to manage study circles. Both RNs and NAOs were taught how to identify

individual needs and underlying causes of nutritional problems, how medications might affect

appetite and eating, and how to structure nutritional documentation (II). The RNs were taught

how to assess nutritional status by using the two-step MNA and to use the guide of the MNA

form (Guigoz et al 1999, 2002) (III). They were also taught how to calculate energy

requirements, based on estimated metabolic rate (BMR), using the equation given by the

Nordic Nutrition Recommendation (1996) (II).

Third, education was given to the NAs. This was organized as study circles, with a RN or a

NAO as leaders. Each study circle included 8-10 members of staff. On five afternoons during

three months, each study circle had meetings directed by a study guide prepared by

Christensson et al (2001). The study guide focused on issues regarding nutritional problems in

elderly people, with the aim of stimulating the NAs to scrutinize how they could improve the

nutritional care in their daily nutritional arrangements, as well as improving their knowledge.

The NAs were taught how to use a ´dietary plan´, which had previously been developed and

used by Christensson et al (2001) and how to document in a food record. The meals in the

´dietary plan´ covered every meal every day during a 10-week period, and were presented at

three base levels: 5500, 7200 and 9000 kJ. Giving service at the right level was made possible

by the meals being described in domestic terms, e.g. a small or normal slice of bread, a small

or middle-sized potato, deciliter of soup or peas. The three base levels were based on the

residents´ calculated energy requirements. No upper limit was set. The aim was to use the

´dietary plan´ in residents with nutritional problems. A simplified food record, based on the

22

´dietary plan´, was used in order to evaluate the residents´ daily intakes. In the control group,

all residents received routine care (II).

The individualized nutritional programme was conducted on the basis of the residents´ habits,

desires, resources and problems, as well as how the residents managed their meals before

admission. Information was obtained from the residents, or their next of kin, as well as from

the residents´ records, concerning diseases and medical treatments that might influence food

intake. The staff scrutinized how they could improve the residents´ food intake, and

investigated how they could avoid phenomena interfering with the residents eating, such as

disturbing events in the eating environment. In some cases, an occupational therapist, or a

physiotherapist, or the resident´s general practitioner or dentist was consulted. During the

intervention period, successful and unsuccessful events during meals were recorded, including

aids used and the techniques used for assisted feeding, as well as complications, such as

nausea and diarrhoea. When the residents did not reach calculated daily energy intake,

according to the food record, supplements were added, such as sandwiches, soups, snacks,

and/or enriched drinks.

Education was given to the staff and, in cases where residents were identified as at risk of

malnutrition, including that measured by the MNA consisting of two parts, implementation of

individualized nutritional care was also implemented. Those residents who were not identified

as at risk of malnutrition, or malnutrition by the MNA, received no individualised nutritional

care, but it is noteworthy that they received benefit of the education given to the staff.

Interview study (IV)

In order to investigate how the residents viewed their appetite and what affects their appetite,

an interview study was carried out, which had to be performed in a trustworthy way. It

included paying attention to credibility by respecting the residents´ desires, conditions and

what they really wanted to say. It also included assuring dependability by strict adherence to

grounded theory methodology. Transferability was attained through making comparisons with

the result of other research studies (Lincoln & Guba 1985).

23

Data were collected by interviews according to grounded theory, since the aim was to identify

and describe factors of importance with regard to appetite among the residents. The primary

purpose of the grounded theory approach is to generate comprehensive explanations of a

phenomenon, such as appetite, that are grounded in reality. It is an inductive research method

developed by two sociologists, Glaser and Strauss (1967), in the 1960s and has its roots in

symbolic interactionism. This means that individuals believe and interact based on how they

give meaning to and interpret specific symbols in their lives, such as food and eating or verbal

and non-verbal expressions (Glaser & Strauss 1967). The authors have then developed the

method in two different ways. The method according to Glaser is the traditional method,

while the method according to Strauss has been developed to involve conceptual description

(Stern 1994). In this interview study, the grounded theory according to Glaser has been used

(Glaser 1978, 1998), since the aim of the study required paying attention to data and to

allowing the data to tell its own story (Stern 1994).

The participants in study IV were chosen, one by one, when they were needed for their

theoretical relevance for the area studied, so called theoretical sampling. Residents with 23

points or more according to MMSE were considered to have the cognitive ability to

understand what it would mean to participate in the interview study. The sampling process

continued until saturation was attained, which occurred when new data did not yield any new

information. Consequently, the sample size could not be known until saturation was reached

and the sample process was ended (Glaser 1978, 1998). Saturation was reached after 10

interviews, but 15 interviews were performed. The five last interviews were valuable because

they confirmed the previous interviews.

The interviews were performed in the form of a conversation with two guiding questions: a)

How do you view your appetite? b) What affects your appetite? The interviews varied in

length from 10 to 45 minutes, because of what raised during the conversation. All interviews

were tape-recorded and transcribed verbatim.

Statistics (I-III)

The results are described using frequencies or percentages, arithmetic means and standard

deviations (I-III), and as medians and inter quartile ranges (II, III). Differences between

24

independent groups were determined by using student´s t-test (I-III), one way ANOVA (II),

chi-square test (I-III), Mann Whitney U-test (III), and Kruskal Wallis test (II). In order to

analyse differences within groups, paired t-test (III) and ANOVA for repeated measures (III),

Wilcoxon´s signet rank test (III), and McNemar´s test (III), were used. When determining

means between subgroups for repeated measures, MANOVA was employed (III).

Multiple logistic regression forward stepwise analysis (WALD) was used to estimate the

MNA items´ association with being classified as PEM, where all MNA items and their

weighted scores constituted independent variables (I). Internal consistency of the MNA items

was evaluated by using Cronbach´s alpha (III). The MNA classes were dichotomised into well

nourished (MNA 1) or at risk of malnutrition and malnourished (MNA 2 and 3) (I, III).

Sensitivity, specificity and diagnostic predictivity, regarding the dichotomised MNA

classification towards PEM/non-PEM (I), as well as MNA-SF towards the dichotomised

MNA classification, were estimated (III). Inter-rater reliability of the MNA tool was tested

regarding parallel assessments performed by the researcher and the RNs by using kappa

agreement test (strength of agreement: < 0.20, poor; 0.21-0.40, fair; 0.41-0.60, moderate;

0.61-0.80, good; 0.81-1.00, very good) (Altman 1991) (III).

Spearman´s rank correlation analysis was used in order to test the correlation coefficient

between the dichotomised MMSE and the item neuropsychological problems in MNA (I), and

the correlation between each of the MNA items toward the item total of MNA (item

calculated excluded) (III). Differences were considered statistical significant at a p-value

below 0.05 (I-III). The statistical programme SPSS ®, versions 11.5 (I, II) and 13.0 (III) were

used.

Analysis of qualitative data (IV)

The data analysis of the interviews in study IV was carried out on three levels. At the first

level, all data was fractured into different incidents, with the intention of finding as many

substantive codes as possible (e.g. happiness, integrity, tiredness, favourite dish, and

arguments). These codes described factors affecting the appetite positively or negatively. The

second level involved comparisons between the codes, in order to find similarities between

them and to create categories. Codes reflecting the same thing, but from different directions,

25

were compiled into the same category (e.g. independence and resignation in the category

“personal values”). Six different categories were compiled by this procedure. At the third

level, the intention was to find general patterns and features, which could link categories and

codes together and explain what was going on in data (Glaser 1978, 1998). Through this

method of analysis, the core category “the willingness to eat” emerged from the categories.

Levels two and three involved an increasing level of the abstract levels of the data analysis

(Hutchinson 1986). During the data collection procedure, newly collected data were compared

in an ongoing process with data obtained earlier, in order to refine theoretically relevant

categories. Data were read several times in order to verify their fit, work and relevance

(Glaser 1978, 1998). During the whole analysis process, theoretical ideas emerged about the

codes, categories and their relationship. These ideas were written down as memos, and were

included in the analysis. In that way, data analysis involved interplay between deductive

memos and inductive data components (Hutchinson 1986). The analysis was performed

mainly by the researcher, but all steps in the analysis process were discussed with the co-

author of the study.

Ethical considerations

The ethical principles of this thesis are based on the principles of autonomy (the residents´

right to self-determination), beneficence (doing only those things that are of benefit to the

residents), non-malfeasance (avoiding harm) and justice (equal care for all) (Swedish

Research Council 2003). Autonomy was achieved by obtaining informed consent from the

residents, after they had received verbal and written information about the aims of the studies.

They were informed that participation was voluntary, that they could discontinue participation

whenever they wanted without giving any explanation, and how data would be used. For

those residents with cognitive dysfunctions, the next of kin were also informed, and their

informed consent was obtained. The principle of beneficence was obtained through

knowledge of how to improve nutritional and functional status in the frail and vulnerable

residents by using the nutritional intervention programme, by obtaining knowledge about

nutritional assessments, as well as receiving a deeper understanding what affects the

residents´ appetite. Non-malfeasance was reached by respecting the residents´ wishes and by

26

minimising alteration of their life situation as much as possible, as well as not exposing them

to unnecessary stress. Therefore, residents were at first hand admitted to resident homes in the

neighbourhood. Justice was obtained by not excluding residents with cognitive dysfunctions,

since it is important to receive knowledge about the residents as a whole. The thesis was

approved by the Research Ethics Committee, Faculty of Health Sciences, Linköping

University (Registration numbers 96-205, 00-243).

RESULTS

Characteristics of residents in cohort 2 upon admission

The sample comprised 127 residents admitted from October 2000 to April 2002, to resident

homes. The age of the residents varied from 70 to 98 years, with a median age of 85 years.

Eighty-seven residents (68 %) were women. The most common diseases among the residents

were dementia, cerebrovascular disease and symptomatic heart failure. Each of these three

diseases was present in 33, 33, and 32 %, respectively, of the residents. In men, symptomatic

heart failure was statistically significant more common than it was in women (Table 2). Other

diseases among the residents were diabetes mellitus (16 %) and ischaemic heart failure (13

%). Some residents had more than one of the diseases.

The median number of prescription medications was, for both men and women, six, with a

variation in women from zero to 18, and in men from one to nine. The majority of those on

medications, among both men and women, were taking them for the nervous system, the

circulatory system, and the blood and blood-producing organs. Less than half of the residents

were on medication affecting digestion. The residents were admitted from another resident

home, their own homes and from short-term stay. Fifty-three per cent of the women and 47 %

of the men were admitted from their own homes (Table 2). In the interview study, three

residents were men and 12 were women. The median age of these residents was 89 years,

with the youngest being 79, and the oldest 95.

27

Table 2. Characteristics of residents in cohort 2 (n = 127) at the time of entering resident homes. Disease and prescription are mainly more than one in each resident. __________________________________________________________________________ Variable Women Men n = 87 n = 40 __________________________________________________________________________ Admitted from Another resident home: % 30 25 Own home: % 53 47 Short-term stay: % 17 27 Disease Dementia: % 34 30 Cerebrovascular: % 29 42 Symptomatic heart failure: % 24 50 ** Diabetes mellitus: % 18 12 Ischaemic heart failure: % 11 17 Prescription medications Nervous system: % 85 75 Circulatory system: % 64 67 Blood and blood-producing organs: % 62 72 Digestive organs and metabolism: % 42 30 ________________________________________________________________________ ** p < 0.01; Mann Whitney U-test

According to the objective assessment of nutritional status, 32 % of the residents were

classified as being PEM (11 men, 30 women) (I, II). The highest frequency was seen among

residents admitted from short-term stay, where 10 residents (38 %) were classified as PEM. In

residents admitted from another resident home or from their own homes, the number was 12

(33 %) and 19 (29 %), respectively (I). In PEM-residents, the CRP median value was 0.0 (0.0

- 16.5), and it was 0.0 (0.0 - 19.2) in non-PEM residents (ns) (II).

According to MNA, 38 of the residents (30 %) were classified as well nourished (MNA 1), 64

(50 %) were as at risk of malnutrition (MNA 2) and 25 (20 %) were classified as being

malnourished (MNA 3) (I, III). No difference was seen regarding residents admitted from

another resident home, their own homes or short-term stay, according to chi-square test

(p = 0.9) (I).

The overall cognitive function according to MMSE in the residents (n = 120) showed a

median value of 16 (0-22). Two residents scored at most 30 points. Eighty-three residents

scored 20 points or less, according to MMSE and were thus assessed to be cognitively

28

impaired (I, II). Residents admitted from their own homes had statistically significant better

overall cognitive function, as well as mental capacity, ADL function, and total activity score,

as compared with those who were admitted from another resident home and from short-term

stay (Table 3). No differences were seen between men and women.

Table 3. Characteristics of overall cognitive function and functional capacity in residents in cohort 2, admitted from another resident home (n = 65), own home (n = 36) and short-term stay (n = 26) at the time of entering resident homes. Each value is given as the median (inter quartile range). __________________________________________________________________________________ Admitted from Variable Another Own home Short-term Resident Home Stay _________________________________________________________________________________ Overall cognitive function MMS 13 (0-19) 1 20 (6-24) 2 * 17 (5-22) 3 Functional capacity Mental capacity 27 (19-29) 29 (27-32) * 27 (23-32) Motor activity 22 (18-24) 24 (20-24) 20 (19-24) ADL function 24 (17-32) 28 (22-26) * 24 (18-28) Total activity score 71 (55-82) 81 (70-88) * 74 (60-80) __________________________________________________________________________________ * p < 0.05; Kruskal-Wallis test; 1 = three missing values, 2 = one missing value, 3 = three missing values Comparison between cohorts 1 and 2 (I)

In cohort 1, 208 residents were included during a 12 month period, while in cohort 2, 127

residents were included during an 18 month period. During the interval between the two

cohort studies, the number of accommodation places in the resident homes involved decreased

by 57. In cohort 1, 38 % of newly admitted residents were classified as PEM, as compared

with 32 % in cohort 2. The median ages in cohorts 1 and 2, were 84 and 85 years,

respectively.

Forty-seven per cent of the residents in cohort 1 were admitted from their own homes. In

cohort 2, 51 % of the residents entered resident homes from another resident home. None of

29

the residents in cohort 1 was admitted from short-term stay, and none of the residents in

cohort 2 was admitted from hospital care (Table 4). The highest frequency of PEM in cohorts

1 and 2 was seen in residents admitted from hospital care (42 %) and from short-term stay (38

%), respectively.

In cohort 1, dementia disease, symptomatic heart failure, cerebrovascular disease, diabetes

mellitus, and ischaemic heart failure were present in 22 %, 21 %, 20 %, 19 % and 19 % of the

residents (n = 208), respectively, as compared to cohort 2, where the above mentioned

diseases were present in 33% (p < 0.05), 32 % (p < 0.01), 33 % (p < 0.05), 16 % (ns) and 13

% (ns), respectively, of the residents (n = 127). Some of the residents in both cohorts had

more than one disease.

According to MNA items, the presence of neuropsychological problems among newly

admitted residents, increased statistically significant from cohort 1 to cohort 2. Other

differences seen were that less residents in cohort 2 had been exposed to stress or acute

disease in the previous three months, and more were able to walk, as compared with the

residents in cohort 1 (Table 4).

Table 4. Comparison of type of dwelling/stay before admission and Mini Nutritional Assessment (MNA) items in residents in cohorts 1 (n = 208) and 2 (n = 127). _____________________________________________________________________ Variable Cohort 1 Cohort 2 % % _____________________________________________________________________ Type of dwelling/stay before admission (1)

Another resident home: % 28 51 *** Own home: % 47 28 *** Short-term stay: % - 20 Hospital care: % 25 - MNA-items (2)

Drinks less than 6 cups of fluid/day: % 21 13 Need of help during meals: % 44 36 Neuropsychological problems: % 45 65 *** Inability to walk % 35 23 * Stress or acute disease % 41 23 ** Skin ulceration % 25 24 _____________________________________________________________________ (1) = Chi-square test between cohorts 1 and 2; (2) = Mann Whitney U-test between cohorts 1 and 2; * p < 0.05, ** p < 0.01, *** p < 0.001

30

Intervention (II)

Upon admission, after two, and after four months, the anthropometric mean values within

those residents (n = 62) in the experimental group, who completed the study period were: WI,

94.8 ± 15.5, 96.1 ± 15.4, and 97.2 ± 15.2 (p < 0.01); TSF, 12.4 ± 5.0, 13.0 ± 5.4, and 13.1 ±

5.3 (p < 0.01); and AMC, 23.2 ± 2.9, 23.3 ± 2.6, and 23.6 ± 2.8 (ns). The corresponding

figures within the residents (n = 53) in the control group, who completed the study period

were: WI, 92.0 ± 14.8, 93.0 ± 14.6, and 93.9 ± 14.7 (p < 0.05); TSF, 11.2 ± 5.4, 11.9 ± 5.3,

and 12.1 ± 5.6 (ns); AMC, 22.6 ± 2.5, 22.7 ± 2.4, and 22.9 ± 2.3 (p < 0.05). Between group

analysis showed no statistically significant differences between the experimental and the

control groups, for any of the anthropometric variables.

The biochemical mean values on admission, after two, and after four months, within the

experimental group (n = 62) were: Albumin/s, 37.1 ± 5.0, 37.6 ± 3.9, and 38.1 ± 3.9 (ns);

transthyretin/s, 0.22 ± 0.07, 0.23 ± 0.07, and 0.23 ± 0.06 (ns). Within the control group (n =

53) the corresponding figures were: Albumin/s, 37.2 ± 4.0, 37.3 ± 3.9, and 37.4 ± 3.8 (ns);

transthyretin/s, 0.23 ± 0.05, 0.23 ± 0.04, and 0.24 ± 0.06 (ns). Between group analysis showed

no statistically significant differences between the experimental and the control groups, for

any of the biochemical variables.

Upon admission, 32 % of the residents in the experimental (n = 62) and in the control group

(n = 53) were classified as being PEM. During the four month study period, the presence of

PEM decreased statistically significant in the experimental, but not in the control group

(Table 5).

Table 5. Residents in cohort 2 (n = 127) assessed as PEM on admission to municipal resident homes and after four months, in an experimental (n = 62) and a control group (n = 53). _________________________________________________________________________ On admission After four months n (%) n (%) p-value _________________________________________________________________________ Experimental group 20 (32) 7 (11) 0.004 Control group 17 (32) 11 (21) 0.143 _________________________________________________________________________ McNemar´s test

31

Within the experimental group, the mean body weight (kg) on admission, after two, and after

four months, was 61.4 ± 10.3, 62.1 ± 10.2, and 62.8 ± 9.7 (p < 0.001), respectively. The

corresponding figures within the control group were 59.5 ± 11.4, 60.0 ± 11.5, and 60.7 ± 11.7

(p < 0.001), respectively.

Among those 62 residents in the experimental unit, who completed the four months study

period, the RNs carried out the two-step MNA procedure in 42 residents. Reasons for the RNs

not carrying out the two-step MNA procedure in all the residents were forgetfulness and

absence due to illness, according to the RNs´ verbal explanations. Out of those 42 residents,

who were assessed by the RNs, 17 were included in the individualized nutritional programme.

In these residents, energy requirement was predicted, individualised actions toward eating

problems were implemented and registration of food intake was carried out. Within the group

of 17 residents, the anthropometric and the biochemical values on admission, after two, and

after four months, were: WI, 84.8 ± 16.6, 86.7 ± 17.0, and 89.0 ± 16.6 (p < 0.01); TSF, 10.8 ±

4.7, 11.7 ± 5.2, and 11.5 ± 4.9 (ns); and AMC, 21.9 ± 2.5, 21.5 ± 1.6, and 21.7 ± 1.9 (ns);

albumin/s, 35.7 ± 5.4, 36.5 ± 4.2, and 37.9 ± 4.5 (ns); transthyretin/s, 0.21 ± 0.07, 0.24 ± 0.07,

and 0.23 ± 0.02 (ns). In these 17 residents, mean body weight (kg) on admission, after two,

and after four months was 55.5 ± 9.0, 56.6 ± 9.5 and 57.8 ± 8.8 (ns), respectively.

From admission to four month later, the residents in the experimental group (n = 62)

improved statistically significant regarding motor activity and MMSE, according to within

group tests. During the same period, motor activity decreased in the control group (n = 53)

(Table 6). Between the experimental and the control groups, statistically significant

differences were seen on admission in mental capacity, ADL function, and total activity score.

After four months, statistically significant differences were seen between the two groups in

mental capacity, motor activity, ADL function, total activity score, and MMSE (Table 6).

In those 17 residents in the experimental unit, who were included in the individualized

nutritional programme, the median values regarding activity capacity, on admission and after

four months, respectively, were: mental capacity, 27 (21-28) and 24 (21-28); motor activity,

19 (14-24) and 22 (16-24); ADL, 26 (16- 29) and 21 (14-30); total activity score 71 (51-82)

and 71 (51-80); and MMSE 14 (0-18) and 11 (0-20). Within group tests showed no

statistically significant differences in any of the variables.

32

Table 6. Total activity score (Activity Index, AI) and overall cognitive function (Mini Mental State Examination, MMSE), on admission and after four months, in residents in cohort 2 (n = 127), newly admitted to municipal resident homes, in an experimental and a control group. A higher score indicates a higher activity capacity. __________________________________________________________________________________ Variable Admission After four months Wilcoxon rank

Md (IQR) Md (IQR) p-value __________________________________________________________________________________ Experimental group (n = 62) Functional capacity (AI) Mental capacity (4-32) 28 (24-30) 28 (25-30) 0.219 Motor activity (6-24) 23 (18-24) 24 (19-24) 0.006 ADL (6-36) 27 (22-34) 29 (22-36) 0.583 Total activity score (16-92) 79 (68-86) 79 (69-88) 0.052 Overall cognitive function MMSE (0-30) 1 17 (6-22) 19 (6-24) 0.021 Control group (n = 53) Functional capacity (AI) Mental capacity (4-32) 27 (18-28) * 27 (18-30) * 0.113 Motor activity (6-24) 22 (19-24) 20 (17-24) * 0.025 ADL (6-36) 23 (16-28) ** 24 (16-32) * 0.264 Total activity score (16-92) 72 (55-79) * 70 (53-83) * 0.607 Overall cognitive function MMSE (0-30) 2 12 (0-21) 11 (0-21) * 0.990 __________________________________________________________________________________ * p < 0.05, ** p < 0.01; Mann Whitney U-test, between the experimental and the control groups on admission and after four months; 1 = four missing values, 2 = two missing values

Test of MNA (I, III)

In those 41 residents classified as PEM, 30 were simultaneously classified as MNA 2 or MNA

3, implying a sensitivity of 73 % (a/[a+c]). Eighty-six residents were classified as non-PEM,

of whom 27 were classified as MNA 1. Specificity was 31 % (d/[b+d]) and diagnostic

predictivity was 33 % (a/[a+b]) (I) (Table 7).

33

Table 7. The relation between PEM/non-PEM and MNA classification after dichotomisation (well nourished: MNA 1, at risk of malnutrition or malnourished: MNA 2 and 3), in the residents in cohort 2 (n = 127). _________________________________________________________________________ MNA classification PEM non-PEM Total n n n _________________________________________________________________________ MNA 2 and 3 30 (a) 59 (b) 89 (a+b) MNA 1 11 (c) 27 (d) 38 (c+d) Total 41(a+c) 86 (b+d) 127 (a+b+c+d) _________________________________________________________________________ PEM = protein energy malnutrition; MNA = Mini Nutritional Assessment

The internal consistency of MNA in the residents was 0.68 according to Cronbach´s alpha

(III). With regard to the two-step MNA procedure, 79 residents were simultaneously

classified at risk of malnutrition by MNA-SF and at risk of malnutrition/malnourished by

MNA. Thirty-one residents were classified as well nourished, in both assessment steps. This

gives an agreement of 87 % between MNA-SF and MNA.

In 89 residents classified as at risk of malnutrition or malnourished by MNA, 79 were

simultaneously classified as at risk of malnutrition by MNA-SF, implying a sensitivity of 89

%. Thirty-eight residents were classified as well nourished by MNA, of whom 31 were

classified in the same way by MNA-SF, giving a specificity of 82 %. Diagnostic predictivity

was 92 % (III).

After receiving education and training concerning how to use the two-step MNA procedure,

the RNs in resident homes A, B and C carried out the assessment procedure on admission in

45 residents, out of 68. The agreement level between the RNs and the researcher MNA

classification was 62 %, since 12 residents were simultaneously classified as well nourished,

ten at risk of malnutrition, and six as malnourished, implying a kappa agreement level of 0.41.

The RNs´ median MNA value in those 45 residents was, 24.0 (18.5 - 25.5), while the

researcher´s corresponding figure in those residents was 22.5 (19.0 - 25.0). In those 23

residents, not assessed by the RNs, the researcher´s median value was 22.5 (19.5 - 26.0).

34

According to the researcher´s assessment, 33 residents were classified as at risk of

malnutrition, while the RNs classified 13 residents in the same way. In resident home C, the

RN classified four residents as at risk of malnutrition, as compared with 16 by the researcher.

Out of 22 residents in resident home B, 13 were not assessed by the RN (Table 8).

Table 8. Classification by Mini Nutritional Assessment (MNA) in 68 residents in cohort 2, newly admitted to three municipal resident homes (A, B and C), performed by registered nurse (RN) and researcher (R). The registered nurse carried out MNA in 45 residents _______________________________________________________________________ Resident home A B C Total RN/R RN/R RN/R RN/R n/n n/n n/n n/n _______________________________________________________________________ MNA 1 0/0 2/10 22/15 24/25 MNA 2 3/8 6/9 4/16 13/33 MNA 3 6/6 1/3 1/1 8/10 Total 9/14 9/22 27/32 45/68 ________________________________________________________________________ MNA 1 = well nourished; MNA 2 = risk of malnutrition; MNA 3 = malnourished

Interview study about appetite (IV)

The residents in this study described their appetite on a sliding scale, from good to poor, as

well as how their appetite expressed itself through concrete descriptions of their eating

patterns and how much they ate. They also viewed their appetite as their experiences of

eating, such as enjoying eating or not liking food or eating. All the residents in this study

described eating to be, to a great extent, an act of will, based on desire or uneasiness to eat.

Sometimes they did not want to eat but forced themselves to do so and at other times they ate

with restraint because they did not wish to eat too much. They strived to control their eating

using their sense and conduct it with their will.

The willingness to eat is the theory as to what affects the appetite among the residents in this

study. It contains internal as well as external factors. The internal factors are dependent on

35

mood and personal values. Mood was described as being dependent on inner feelings and

associated with how the residents looked upon their life situation, where the character of

every individual was of importance. They also described how emotion affects appetite, where

joy and delight affected it positively, while anxiety and grief affected it negatively.

Independence and integrity were personal values the residents believed had importance for

their appetite. The ability to do things without assistance was important to them, as was

pursuing their interests and maintaining their usual life style. They also said that it was

important to be satisfied with living at the resident home and to have influence over their own

lives.

The external factors are dependent on wholesomeness, food, eating environment and meal

fellowship. Wholesomeness involves good health, which seems to benefit the appetite of the

residents, in contrast to diseases and unpleasant symptoms. The residents described healthy

diet and physical ability as being of great importance, as well as the desire to maintain their

health. Their knowledge about what was healthy gave them the motivation to eat even when

they had poor appetite. The residents described having an increased appetite when the food

was well prepared and in line with their expectations, i.e. the right consistency and served

fresh. They said their appetite increased when the food tasted good and it was a pleasure to

eat. Eating environment, including external conditions and different sight-and sound-

impressions, was a factor of importance for the appetite of the residents. They said they

wanted to eat in dignified circumstances, to appreciate the food in the right way. They said

that they liked eating in peace and quiet, which included eating without stress and in smaller

groups. Meal fellowship included occasions when the residents met and socialised. All of

them said with emphasis that it was of importance for them to eat together. Their appetites

increased when they ate together and gladly with people they knew and most of all with

relatives or close friends.

DISCUSSION

Newly admitted residents in this thesis formed a fragile group, many of them with a

deteriorated nutritional status and functional capacity, which makes them to a relevant group

for nutritional intervention. During the four years between cohorts 1 and 2, the number of

36

newly admitted residents classified as PEM did not decrease, despite the awareness of the

problem.

After implementation of the nutritional programme, no statistically significant differences

were seen between the experimental and the control groups for any specific nutritional

variable. However, in the experimental group, the number of residents classified as PEM

decreased statistically significant, but not in the control group. In the experimental group,

statistically significant improvements were seen in motor activity and MMSE function during

the intervention period, while in the control group, motor activity deteriorated statistically

significantly. Despite education given to the RNs, they did not fulfill the instructions and

assessed nutritional status in all residents, and consequently not all residents in need of

nutritional attention received individual nutritional care.

When the nutritional programme was implemented, the course of actions suggested by

ESPEN, was followed (Kondrup et al 2003) as were the Swedish guidelines for nursing care

(SOSFS 1993:17). Due to the fact that nutritional problems in the residents are multi factorial,

individual nutritional care is an appropriate strategy to fulfill their nutritional needs. This is

confirmed by what emerged during the interview study, that the willingness to eat, which

comprises mood, personal values, wholesomeness, food, eating environment and meal

fellowship, is what affects appetite and eating in the residents.

Basal characteristics in cohort 2

Newly admitted residents were in this thesis very old, and the majority of them were women.

Women in Sweden have a higher average length of life than men do (Statistics Sweden 2002),

which could be the reason of a higher proportion of elderly women in need of help. The most

common diseases and prescription medications were equal, except for symptomatic heart

failure, which was twice as common in men as compared with women. Akner & Cederholm

(2001) have shown that malnutrition in residents is often seen in connection with a variety of

chronic diseases and in residents with several prescriptions. But even previously ´healthy´

elderly people, e.g. those affected by hip fractures, have been observed to be malnourished

(Bachrach-Lindström & Unosson 2000). In a study by Saletti et al (2005), approximately half

37

of home-living elderly people with support from the municipality were at risk of malnutrition

or were malnourished.

Half of the residents in this study were admitted from another residential home and 28 % from

short-term stay. The reason why the residents moved from one resident home to another

suggests a non-functioning living situation, such as changed conditions of the resident, or that

the residents or their next of kin were dissatisfied with the residents´ previous living

arrangements. This indicates that when some of the residents had to move into a resident

home, their living arrangements were settled by the vacant place situation, and this was not

always in accordance with the residents´ conditions and wishes.

Moving from one location to another has been described as a painful process for residents

(Johnson-Tripp-Reimer 2001), resulting in decreased activity, insufficient food intake, and

weight loss (Robertsson & Montagini 2004). The residents who had previously moved from

one resident home to another had statistically significantly worse MMSE scores, mental

capacity, ADL function, as well as total activity score, as compared with those who entered

resident homes from their own homes. It was in the residents admitted from short-term stay

and from another resident home that the highest (38 %), and the second highest (33 %),

frequencies of PEM were seen. It is notable that one-third of residents admitted from another

resident home were classified as being PEM, and there may be several explanations for this.

Maybe the residents were already PEM when they moved into the previous resident home,

and their nutritional status had not improved because of too short a time having elapsed.

Another explanation might be that the residents had had to move because they were in need of

more advanced care due to deteriorated health status, accompanied by deteriorated nutritional

status. It might also have been because the residents´ nutritional needs had not been observed

and fulfilled by the staff in the previous resident home.

Comparison between cohorts 1 and 2

In cohorts 1 and 2, approximately one-third of the residents were classified as PEM, showing

a similar prevalence as in other nursing homes studies during the last years (Lauque et al

2000, Beck & Ovesen 2002, Kofod & Birkemose 2004). During the inclusion year in cohort 1

(I), 82 more residents were admitted to resident homes as compared with cohort 2, even

38

though the cohort 2 study continued for another six months. Half of the residents in cohort 1,

entered resident homes from their own home, and 25 % from hospital care, while in cohort 2,

half of the residents were admitted from another resident home, and 20 % from short-term

stay. The highest frequency of PEM in cohorts 1 and 2, were seen in residents admitted from

hospital care and from short-term stay, respectively. This probably reflects the changes in the

organization of elderly care in the municipality during the time between cohorts 1 and 2. The

expansion of the home care service, fewer accommodation places, and the introduction of

short-term stay are changes performed during the period. In short-term stay, the residents

received further rehabilitation after hospital care, and consequently no residents in cohort 2

were admitted directly from hospital care.

The changes in the organization, seems also to have lead to changed patterns of admission

criteria to resident homes. More residents in cohort 2 suffered from severe medical diseases

and neuropsychological problems. At the same time fewer residents had suffered

psychological stress or acute disease in the previous three months, and more were able to

walk, as compared to cohort 1. It indicates that residents supported by home care service are

able to manage physically, while when neuropsychological problems occur, they require

institutional care. A shifting balance of long-term care is seen in Sweden according to

Sundström et al (2002) to support elderly people within their own homes with help from

family members, rather then in institutional settings. This trend is seen over the whole western

world (McCormack 1997). If this trend continues in Sweden, there is a risk of insufficient

number of accommodation places in the future. More people in Sweden are ageing

(Aijanseppa et al 2005), simultaneously as health problems in old age have increased (Parker

et al 2004). Many elderly people are also living alone (Carpenter et al 2004) with no support

from family members (Sundström et al 2002). All together indicates an increased need of

nursing care among elderly people in the future.

Intervention

In the experimental unit, actions toward malnutrition were implemented by given education to

the staff, so they could identify residents in need of nutritional attention, make diagnosis, and

give appropriate individually nutritional care based on ordinary food. To use ordinary food,

with individually planned nutritional care, in order to improve nutritional status and functional

39

capacity, has not earlier been seen neither in nursing homes studies (Lauque et al 2000,

Fiatarone Sing et al 2000, Faxen-Irving et al 2002) and nor in hospital settings (Bachrach-

Lindsröm et al 2001, Potter et al 2001, Wouters-Wesseling et al 2002, Gazzotti et al 2003).

The intervention followed the recommendation given by ESPEN (Kondrup et al 2003) and the

Swedish guidelines of nursing care (SOSFS 1993:17).

Even if there was no statistical significant difference seen in anthropometric and biochemical

variables between the experimental and the control groups, there is a tendency seen that the

intervention had effects. This is confirmed by a statistically significantly decreased number of

residents classified as PEM in the experimental group, and statistically significant improved

motor activity and overall cognitive function according to MMSE. In the control group, no

improvements were seen, but motor activity deteriorated statistically significant, as well as the

MMSE median score declined from 12 to 11 during the study period. Increased mobility in

the residents in the experimental group, may explain why increased energy intake had

moderately effect on the residents´ weight gain, which also has been seen in a study by

Elmståhl et al (1987). According to the power calculation, based on body weight, 50 residents

in each group were estimated to be sufficient for receiving statistical differences between the

groups. Despite this, no statistically significant differences were seen when the two groups

were compared, which may be explained by difficulties during the study period, insufficient

number of residents in the study, or that the intervention period maybe was too short.

In those 23 residents, who were not assessed by the RNs, and in those 45 who were assessed,

the researcher´s MNA assessment was the same (median value 22.5), which indicate that they

did not differ with regard to the MNA assessments. Due to that, it is reasonable to suggest that

if the RNs should have carried out the two-step MNA procedure in the remaining 23

residents, the assessments should have been similar as in those 45 residents, who were

assessed by the RNs. Consequently more residents should have been included in the

individually nutritional program, which might have affected the outcome of the intervention.

Despite the education given to the RNs about nutritional care and how to use the MNA, the

RNs performed the two-step MNA assessment procedure in two-third of the residents. In

resident home B, the RN carried out MNA in only nine residents out of 22 (41 %). The reason

why the RNs not assessed all the residents was because of forgetfulness and absence due to

illness, according to oral explanations by the RNs. It has been observed in another study that

40

the staffs motivation and changes in the staff affected the study outcome (Wikby et al 2003).

The reason why the RNs forget to follow the instructions given may be related to their

attitudes toward nutritional problems in elderly people. Malnutrition has been observed as a

neglected problem in elderly people (Elmståhl et al 1997) and to have low priority among

nurses and physicians (Rasmussen et al 1999). Also time-consuming actions and lack of time

have been seen to be barriers to nutritional care (Crogan et al 2001). To influence staff

attitudes towards nutritional nursing care is difficult, which Christensson et al (2003) have

shown in a study of staff participating in an educational program.

Twelve residents in resident home C, were classified as at risk of malnutrition by the

researcher, but as well nourished by the RN. As the RN classified theses residents as well

nourished, they consequently did not receive individualized nutritional care. The differences

between the researcher and the RNs MNA classification in resident home C was maybe

because of the RN´s less experience to perform the two-step MNA procedure, as compared

with the researcher. In a hospital study, differences were seen in nutritional assessments

between an experienced and a less experienced observer (Ek et al 1996). That study points out

that it is necessary to be well trained in order to manage assessment of nutritional status.

Other circumstances may also have affected the outcome of the intervention, such as

assimilating knowledge about the intervention in the control group, as it was not possible to

keep the resident homes isolated. Also the novelty effects and the knowledge of participating

in a study may have affected the behaviour of the staff, as well as, different nursing care in the

various resident homes could exert an important influence on the homogeneity of the result. It

may also depends on other circumstances affected the outcome of the intervention, such as the

complexity in elderly people.

Assessments

Since physical, psychological or social problems in elderly people usually tend to make them

vulnerable, this increases the risk of loss of appetite, insufficient dietary intake and by that

risk for developing malnutrition (Morley 2001, 2002). Malnutrition in turn has negative

influences on the total life situation by decreased functional status and psychosocial wellbeing

(Crogan & Pasvogel 2003). Nutritional status in elderly people may thus be seen as a marker,

41

pointing out the overall condition of the individual, and as important to assess by using

standard geriatric assessment. In resident homes, a comprehensive geriatric assessment

(Devons 2002) should be used on admission, and repeatedly when there are signs indicating

problems.

In order to make comparison with the cohort 1 study (Christensson et al 1999), performed

four years earlier in the same municipality, objective nutritional assessment classifying

residents as either PEM or non-PEM, was used. C-reactive protein was measured to give

information about possible infections and inflammations (Ingenbleek & Young 2002;

Kalender et al 2002). Similar CRP values in PEM and non-PEM residents, indicates that the

residents not were classified as being PEM because of ongoing infections or inflammations,

but because of energy intake below that required.

To use anthropometric and biochemical variables as standard geriatric assessment in resident

homes involves difficulties, as anthropometric measures, such as TSF and AMC, demand to

use a Harpender skinfold calliper which is not a commonly used tool. Biochemical variables,

such as serum albumin and transthyretin, demands blood-sampling and laboratory analysis,

which are not always possible to carry out in resident homes.

Mini Nutritional Assessment was used because it is regarded as a valid tool specially

developed for detection of elderly people at risk of malnutrition (Guigoz et al 1996), but also

as it is a rapid tool without requiring specialised instruments or a specialised nutrition team

(Guigoz et al 2002). Selected items in MNA have earlier been used in the previous cohort 1

study (Christensson et al 1999). In order to make comparisons with that study it was

appropriate to use the MNA.

When testing the agreement between the dichotomised MNA classes in relation to PEM/non-

PEM (I), 30 PEM residents, out of 41, were simultaneously classified as at risk of

malnutrition/malnourished according to MNA, which was regarded as a satisfying degree of

agreement. The fact that not all residents classified as being PEM, also were classified as at

risk of malnutrition or malnourished by MNA, is in line with a study by Chumlea et al (1999),

where limitations in sensitivity of MNA versus measured body composition was shown. On

the other hand, not all residents categorised by MNA as at risk of malnutrition were classified

as PEM.

42

When diagnostic test was performed using MNA-SF versus the dichotomised two-step MNA

procedure, 110 residents out of 127 were classified in the same way in both assessments,

implying a high agreement level. The sensitivity was 89 %, which indicates that MNA-SF

itself might be an alternative to the two-step MNA procedure, since it has the practical

advantages of being quick and easy to use. The purpose of a screening tool, such as MNA-SF,

is to identify malnourished individuals or those at risk of becoming malnourished (Dougherty

et al 1995, Kondrup et al 2003), so that more extensive nutrition assessment can be performed

and intervention implemented (Berner 2003).

Activity Index was chosen to assess functional capacity in the residents despite it was

developed for stroke patients. Functional capacity vary in elderly people due to a high

frequency of diseases and impairments and as the activity index has ability to assess great

variations in mental capacity, motor activity and ADL function, it seemed to be appropriate to

use (Hamrin & Wohlin 1982). The tools relevance for clinical use (Chong 1995) was good

since it was easy and quick to use in all residents. Unfortunately is it not a wide spread

geriatric assessment, and has as so far most been used in Swedish studies (Lindmark 1988,

Elmståhl et al 1996, Christensson et al 2001).

Mini mental state examination was used to screen for overall cognitive function of the

residents, since it is a well validated and used tool and is able to separate individuals with

cognitive impairment from those without. According to Kurlowicz & Wallace (1999) the tool

is able to make comparisons between different studies and when it is used repeatedly, the tool

is able to assess changes in cognitive function. Different cut off score is recommended for

MMSE, since there is an inverse relationship between MMSE score, age and educational level

(Crum et al 1993). As the residents were old and admitted from an area with mainly rural

work and industries, a cut off score of 20 or less considered as representing cognitive

impairment (Folstein et al 1975) in study I and II, which is in line with the recommendation

given by Crum et al (1993). In the interview study (IV), the cut off score was increased to 23,

to make sure that the residents had ability to understand what participation would mean for

them. According to Kurlowicz & Wallace (1999), the MMSE relies heavily to physical

ability. Therefore residents with physical impairments may perform poorly even when they

are cognitively intact (Kurlowicz & Wallace 1999). In this thesis six residents were not tested

according to MMSE, due to aphasia. Three residents who participated in the interview study

43

with score 23, were not able to carry out MMSE because of physical disabilities, like impaired

sight and hand tremor.

Interview study about appetite

In the interviews the residents viewed their appetite by given concrete descriptions of what

they liked and disliked regarding different circumstances, e.g. food and eating patterns, rather

than described their appetite in more abstract terms, i.e. in that way appetite has been

described in the literature (Whitney et al 2002). Each of the categories, mood, personal

values, food, wholesomeness, eating environment and meal fellowship, were described by the

residents as affecting the appetite either positively or negatively, like on a sliding scale.

The willingness to eat is the theory as to what affects the appetite among the residents in this

study. It imbues the interviews, links the codes and categories together, and explains what is

going on in data. The willingness to eat is in line with a theory of human motivation, which

belongs to a certain conception of the human being as a subject in goal-directed actions (Pörn

1977). According to Pörn (1981,1984), motivation might be seeing as the willingness to

decrease the discrepancy between what the subject wants to be and what he/she really is. The

subject acts due to the relation between his/her abilities (repertoire) and environment, where

the environment involves internal (psychological) and external (socio- cultural) factors

(Gustafsson & Pörn 1994). Mood and personal values can in the perspective of Gustafsson &

Pörn (1994) being seeing as internal factors, and wholesomeness, food, eating environment

and meal fellowship as external factors. Out of the residents´ presuppositions, based on

abilities and environment, he/she can be motivated either to act or not act towards a specific

goal, such as eating a meal.

A connection may be seen between the willingness to eat and the residents´ quality of life.

Mood and personal values are in agreement with what Sarvimäki & Stenbock-Hult (2000) call

intra-individual conditions. Wholesomeness, food, eating environment and meal fellowship

are in line with what they call external conditions. All the conditions influence each other and

constitute conditions for quality of life (Sarvimäki & Stenbock-Hult 2000). Mood, personal

values, food, wholesomeness, eating environment and meal fellowship may thus be seeing as

affecting appetite and quality of life to a varying extent. When the residents in this study had

44

ability to maintain their internal and external conditions, it affected positively their appetite

and quality of life. The opposite was also seen, such as when the residents had difficulties to

maintain their internal or external conditions, it negatively affected their appetite and quality

of life. Eating problems in residents may therefore not be seeing as isolated problems, but

integrated parts of the residents and are not possible to separate from the rest of the residents´

life, as these things constitute a whole. This has also been seen in other studies (Gastman

1998; Vetta et al 1999; Crogan & Pasvogel 2003).

In this study consideration was taken to the residents´ conditions, and what they really wanted

to say. As they were old and frail and thus became easily tired and confused in stressful

situations, it was important to not force them in any way. If they were stressed to answer

unexpected questions, they easily became confused and did not know what to answer. It was

therefore important to not interrupt their train of thought and let the interviews proceed on

their own terms. The residents were also informed that the participation was voluntary and

they could discontinue participation whenever they wanted. Due to that, one resident

discontinued participation after 10 minutes.

The residents in this study had cognitive ability to reflect upon their situation and express

their thoughts and experiences. Not all residents have that ability. Even if every individual is

unique, all people as a whole are alike. In that way the residents in this study may be seeing as

spokespersons for residents in similar situations who do not have the ability to reflect upon

their life situation and express their thoughts and experiences. By that the result of this study

can be transferred to elderly people in similar situations.

Clinical implications

To identify residents in need of nutritional intervention should be a key component within

nursing care, where MNA is a useful screening tool for early detection of those at risk. Due to

a high agreement regarding the sensitivity of MNA-SF as compared with MNA, MNA-SF

itself can be a sufficient screening tool in resident homes, since it has practical advantages. To

overcome the problem among the RNs to not carry out nutritional assessments in all residents,

due to forgetfulness and absence due to illness, more resources are needed. Suggested actions

45

are to give the RNs more education about nutritional issues and about the two-step MNA

procedure, but also to give the RNs continuous supervision to motivate and support them to

include all residents in need of nutritional attention. Maybe also directions within the

organization of elderly care are needed, with instructions to the staff to priority nutritional

assessments and issues, as well as adequate number of staff.

In residents identified as at risk of malnutrition or malnourished, individualised actions have

to be carried out. This is in line with European recommendations and with the Swedish goal

of nursing actions. The individualised actions ought to be based on ordinary food, and

nutritional supplementation would be considered only if it is not possible to meet the

residents´ nutritional requirements with food, or when additional food has been ineffective.

It is of importance that the RNs at the retirement home identifies underlying causes of

insufficient nutrient intake among the residents, as well as find out the residents problems,

desires and resources. This demands that the RNs have sufficient knowledge about every one

of the residents, with regard to their mental, social and physical needs and conditions. When

nursing actions are planned and realised towards eating, the willingness to eat has to be

considered. Every individuals´ desire, which includes factors, such as mood, personal values,

wholesomeness, food, eating environment and meal fellowship, have to be taken into

account, not only every single factor, but also with regard to how the factors affect each

other. Planning, implementation and evaluation of nursing actions ought to be realised as

teamwork and have to involve the residents, their relatives, the RN and other staff. The

residents and their relatives should be seen as important resources in this process.

Even if the intention is to identify all residents in need of nutritional attention, it is probably

not realistic to have a nil vision regarding malnutrition, as the population includes elderly

people in the last stage of life. Before life ends in elderly people, illnesses are common

problems, which mainly can affect appetite and eating. But, even if some residents are in the

very end of their lives, and have difficulties to eat and to improve their nutritional status, they

should still receive individualised nutritional care. In that situation, individualised nutritional

care based on factors affecting the appetite is of great importance, as those factors contain

qualitative dimensions which also affect the quality of life.

46

CONCLUSIONS

Elderly people newly admitted to resident homes are a fragile group, many of them with a

deteriorated nutritional status and functional capacity.

The nutritional intervention seemed to have effect, interpreted from decrease in malnutrition

frequency and improvement in functional capacities

Despite education given to the staff, regarding nutritional care, it was not evident for the RNs

to implement nutritional assessments and measures.

Due to high agreement between MNA-SF and MNA total the MNA-SF itself can be a

sufficient screening tool for practice

The willingness to eat, based on internal and external factors, is supposed to affects the

appetite in elderly people.

47

SUMMARY IN SWEDISH

Ett angeläget och aktuellt problem inom äldrevården är att många äldre riskerar att bli

undernärda. Med stigande ålder riskerar människan att drabbas av förluster av olika slag, som

gör henne mer sårbar. En del av deras problem består av minskad aptit, otillräckligt

näringsintag och som i förlängningen kan resultera i undernäring. Undernäring försämrar

ytterligare den åldrande människans liv genom sänkt livskvalitet, försämrad fysisk och mental

förmåga, samt ökad risk att dö i förtid.

Det övergripande syftet med studierna var att studera om en intervention, baserad på

utbildning av personal och införandet av ett individuellt anpassat nutritionsprogram, kunde

förbättra nyinflyttade vårdtagares näringstillstånd och minska risken för utveckling av

undernäring. Vidare var syftet att studera effekterna av förändringar i den kommunala

äldreomsorgen över en fyraårsperiod, att utföra diagnostiska test och inter-bedömarreliabilitet

av Mini Nutritional Assessment (MNA) och att identifiera och beskriva faktorer av betydelse

för vårdtagarnas aptit.

Vid inflyttningen och efter en fyra månaders interventionsperiod, blev vårdtagarnas

näringstillstånd (protein- energi malnutrition, PEM) bedömt med hjälp av kroppsmått och

serumproteiner. Vid båda tillfällena bedömdes också fysisk förmåga med Aktivitetsidex

(psykiskt status, motorik och ADL) och kognitiv förmåga med Mini Mental Test. För att

identifiera vårdtagare i behov av individuellt anpassade nutritionsåtgärder använde tränade

sjuksköterskor MNA och beslutade om behov av åtgärder eller inte. Vid inflyttningen

samlades information in om vårdtagarnas sjukdomar och mediciner. Totalt 127 vårdtagare

inkluderades konsekutivt till åtta äldreboenden i en kommun i södra Sverige. Tre av

äldreboendena utgjorde experimentenhet (n = 68) och fem kontrollenhet (n = 59). Femton

vårdtagare intervjuades med en kvalitativ metod om vad som påverkar deras aptit.

Resultatet visar att vid inflyttningen bedömdes 32 % av vårdtagarna som undernärda, vilket

var jämförbart med resultatet i en studie genomförd fyra år tidigare. En högre andel av

vårdtagarna i den nuvarande studien hade allvarliga sjukdomar och nedsatt psykiskt status,

vilket indikerar att inflyttningskriterierna har förändrats under de fyra år som har förflutit

mellan studierna.

48

Vid jämförelser mellan experiment- och kontrollgruppen fanns inga statistiskt signifikanta

skillnader beträffande kroppsmått eller serumproteiner. Däremot fanns statistiskt signifikanta

skillnader inom grupperna, där antalet vårdtagare bedömda som PEM minskade i

experimentgruppen och deras motoriska förmåga och psykiska status förbättrades. I

kontrollgruppen försämrades vårdtagarnas motoriska förmåga. Detta tyder på att trots brist på

skillnader mellan experiment- och kontrollgrupp så hade interventionen förbättrat

näringstillståndet, motorisk förmågan och psykiskts status hos vårdtagarna i

experimentgruppen. En orsak till att det inte fanns skillnader mellan experiment och

kontrollgrupp kan vara att sjuksköterskorna inte bedömde alla vårdtagare på

experimentenheten och därmed missade vårdtagare som behövde individuella åtgärder.

Diagnostisk sensitivitet var 73 % hos MNA gentemot PEM, och 89 % hos kortversionen av

MNA (MNA-SF) gentemot totala MNA, vilket visar på en relativt hög diagnostisk sensitivitet

i båda fallen. Inter-bedömarreliabiliteten mellan samtidiga MNA bedömningar utförda av

sjuksköterskor och forskare visade en moderat överensstämmelse på 62 % (kappa 0.41).

I intervjustudien framkom att viljan att äta är vad som påverkar vårdtagarnas aptit. Viljan att

äta inrymmer interna faktorer, som sinnesstämning och personliga värden, och yttre faktorer,

som nyttigheter, mat, ätmiljö och måltidsgemenskap. När omvårdnadshandlingar planeras och

genomförs bör faktorer som påverkar vårdtagarnas aptit beaktas.

Sammanfattningsvis framkom att det är viktigt att införa och utveckla strategier för att

genomföra individuellt anpassad omvårdnad i syfte att förebygga och behandla undernäring

hos äldre. Detta är i linje med rekommendationer givna av European Society of Parenteral and

Enteral Nutrition (ESPEN) och med Socialstyrelsens allmänna råd om omvårdnad inom

hälso- och sjukvården (SOSFS 193:17).

49

ACKNOWLEDGEMENTS

I would like to express my deep gratitude to everyone who has any part in the work of this

thesis. I am especially grateful to:

All residents, their relatives and staff members, who participated in this study and without

whom this work would not been accomplished.

Anna-Christina Ek, Professor, my main supervisor, for excellent scientific guidance,

constructive criticism and for continuous encouragement and for accepted me as a doctoral

student.

Lennart Christensson, PhD, my co-supervisor, for guiding me into the world of research

within nutrition, for constructive criticism, and for inspiring me becoming a researcher.

Astrid Fägerskiöld, PhD, my co-author in study IV, for stimulating discussions and for

guiding me into the world of qualitative research.

All doctoral student, colleges and friends at the Department of Medicine and Care, the

Division of Nursing Science, Faculty of Health Sciences, Linköping University for your

support, helpful criticism during seminars, stimulating discussions and friendship during this

period.

Susan Barclay Öhman, Urban Morén, thanks for your excellent revision on the English

language.

Eva Borg and Erika Lindén, former and current secretaries at the Department of Medicine and

Care, the Division of Nursing Science, Faculty of Health Sciences for all practical help and

support.

All colleagues and friends at the Highland Hospital, Eksjö, thank you for your support and

friendship.

50

The colleagues and friends at the University College of Health Sciences, Jönköping, thank

you for given me support and encouragement during my doctoral studies.

The staff at the library at the University College of Health Sciences, Jönköping, thanks for

valuable assistance of references.

The studies were supported by grants from Vårdalstiftelsen no. 98/359, 03/042, the Faculty of

Health and Sciences, Linköping University, Linköping, Sweden, and Signe Thorfinn´s

foundation.

1

REFERENCES Aevarsson O & Skoog I (2000) A longitudinal population study of the mini-mental state examination in the very old: relation to dementia and education. Dementia and Geriatric Cognitive Disorders 11, 166-175. Aijanseppa S, Notkola IL, Tijhuis M, van Staveren W, Kromhout D & Nissinen A (2005) Physical functioning in elderly Europeans: 10 year changes in the north and south: the HALE project. Journal of Epidemiological Community Health 59, 413-419. Akner G & Cederholm T (2001) Treatment of protein-energy malnutrition in chronic nonmalignant disorders. American Journal of Clinical Nutrition 74, 6-24. Altman DG (1991) Practical statistics for medical research. Chapman & Hall/CRC, Florida. Andersson P, Westergren A, Karlsson S, Rahm Hallberg I & Renvert S (2002) Oral health and nutritional status in a group of geriatric rehabilitation patients. Scandinavian Journal of Caring Sciences 16, 311-318. Bachrach-Lindstrom MA, Ek AC & Unosson M (2000) Nutritional state and functional capacity among elderly Swedish people with acute hip fracture. Scandinavian Journal of Caring Sciences 14, 268-274. Bachrach-Lindstrom M, Unosson M, Ek AC &Arnqvist HJ (2001) Assessment of nutritional status using biochemical and anthropometric variables in a nutritional intervention study of women with hip fracture. Clinical Nutrition 20, 217-223. Beck AM, Ovesen L & Schroll M (2001) A six months' prospective follow-up of 65+-y-old patients from general practice classified according to nutritional risk by the Mini Nutritional Assessment. European Journal of Clinical Nutrition 55, 1028-1033. Beck AM & Ovesen L (2002) Body mass index, weight loss and energy intake of old Danish nursing home residents and home-care clients. Scandinavian Journal of Caring Sciences 16, 86-90. Bengtsson C, Hulten B, Larsson B, Noppa H, Steen B & Warnold I (1981) Nya längd - vikttabeller för medelålders och äldre män och kvinnor, (New weight-height tables in Swedish for middle-aged and elderly men and women). Läkartidningen 78, 3152-3154. Berner YN (2003) Assessment tools for nutritional status in the elderly. Israel Medical Association Journal 5, 365-367. Bickerstaff KA, Grasser CM & McCabe B (2003) How elderly nursing home residents transcend losses of later life. Holistic Nursing Practice 17, 159-165. Bozzetti F (2003) Nutritional issues in the care of the elderly patient. Critical Review in Oncological Hematology 48, 113-121.

2

Browne JP, O'Doherty VA, McGee HM, McLaughlin B, O'Boyle CA & Fuller R (1997) General practitioner and public health nurse views of nutritional risk factors in the elderly. Ireland Journal of Medical Sciences 166, 23-25. Carpenter I, Gambassi G, Topinkova E, Schroll M, Finne-Soveri H, Henrard JC, Garms-Homolova V, Jonsson P, Frijters D, Ljunggren G, Sorbye LW, Wagner C, Onder G, Pedone C & Bernabei R (2004) Community care in Europe. The Aged in Home Care project (AdHOC). Aging: Clinical and Experimental Research 16, 259-269. Chen CC, Chang CK, Chyun DA & McCorkle R (2005) Dynamics of nutritional health in a community sample of american elders: a multidimensional approach using roy adaptation model. Advances in Nursing Science 28, 376-389. Chong DK (1995) Measurement of instrumental activities of daily living in stroke. Stroke 26, 1119-1122. Christensson L, Unosson M & Ek A-C (1999) Malnutrition in elderly people newly admitted to a community resident home. Journal of Nutrition, Health and Aging 3, 133-139. Christensson L, Ek AC & Unosson M (2001) Individually adjusted meals for older people with protein-energy malnutrition: a single case study. Journal of Clinical Nursing 10, 491-502. Christensson L, Unosson M & Ek AC (2002) Evaluation of nutritional assessment techniques in elderly people newly admitted to municipal care. European Journal of Clinical Nutrition 56, 810-818. Christensson L, Unosson M, Bachrach-Lindstrom M & Ek AC (2003) Attitudes of nursing staff towards nutritional nursing care. Scandinavian Journal of Caring Sciences 17, 223-231. Chumlea WC, Hall G, Lilly F, Siervogel RM & Guo SS (1999) The Mini Nutritional Assessment and body composition in healthy adults. Nestlé Nutrition Workshop Series Clinical & Performance Programme 1, 13-21. Cohendy R, Rubenstein LZ & Eledjam JJ (2001) The Mini Nutritional Assessment-Short Form for preoperative nutritional evaluation of elderly patients. Aging 13, 293-297. Conwell Y, Pearson J & DeRenzo EG (1996) Indirect self-destructive behavior among elderly patients in nursing homes. The American Journal of Geriatric Psychiatry, 4, 152-163. Couture M, Lariviere N &Lefrancois R (2005) Psychological distress in older adults with low functional independence: a multidimensional perspective. Archives of Gerontology and Geriatrics 41, 101-111. Crogan NL & Pasvogel A (2003) The influence of protein-calorie malnutrition on quality of life in nursing homes. Journals of Gerontology B: Biological Sciences and Medical Sciences 58, 159-164. Crogan NL, Shultz JA, Adams CE & Massey LK (2001) Barriers to nutrition care for nursing home residents. Journal of Gerontological Nursing 27, 25-31.

3

Crum RM, Anthony JC, Bassett SS & Folstein MF (1993) Population-based norms for the Mini-Mental State Examination by age and educational level. Journal of the American Medical Association (JAMA) 12, 2386-2391. Cullum S, Huppert FA, McGee M, Dening T, Ahmed A, Paykel ES & Brayne C (2000) Decline across different domains of cognitive function in normal ageing: results of a longitudinal population-based study using CAMCOG. International Journal of Geriatric Psychiatry 15, 853-862. Das SK, Moriguti JC, McCrory MA, Saltzman E, Mosunic C, Greenberg AS & Roberts SB (2001) An underfeeding study in healthy men and women provides further evidence of impaired regulation of energy expenditure in old age. Journal of Nutrition 131, 1833-1838. Devons CA (2002) Comprehensive geriatric assessment: making the most of the aging years. Current Opinion of Clinical Nutritional Metabolic Care 5, 19-24. Dey DK, Rothenberg E, Sundh V, Bosaeus I & Steen B (2001) Height and body weight in elderly adults: a 21-year population study on secular trends and related factors in 70-year-olds. Journals of Gerontology A: Biological Sciences and Medical Sciences 56, 780-784. Donini LM, Savina C &Cannella C (2003) Eating habits and appetite control in the elderly: the anorexia of aging. International Psychogeriatrics 15, 73-87. Dougherty D, Bankhead R, Kushner R, Mirtallo J & Winkler M (1995) Nutrition care given new importance in JCAHO standards. Nutritional Clinical Practice 10, 26-31. Draper B, Brodaty H, Low LF, Richards V, Paton H & Lie D (2002) Self-destructive behaviors in nursing home residents. Journal of American Geriatric Society 50, 354-358. Ek A-C, Larsson J, von Schenck H, Thorslund S, Unosson M & Bjurulf P (1990) The correlation between anergy, malnutrition and clinical outcome in an elderly hospital population. Clinical Nutrition 9, 185-189. Ek A-C, Unosson M, Larsson J, von Schenck H & Bjurulf P (1991) The development and healing of pressure sores related to the nutritional state. Clinical Nutrition 10: 245-250. Ek AC, Unosson M, Larsson J, Ganowiak W & Bjurulf P (1996) Interrater variability and validity in subjective nutritional assessment of elderly patients. Scandinavian Journal of Caring Sciences 10, 163-168. Elmståhl S, Blabolil V, Fex G, Kuller R & Steen B (1987) Hospital nutrition in geriatric long-term care medicine. I. Effects of a changed meal environment. Comprehensive Gerontology 1, 29-33. Elmståhl S, Sommer M & Hagberg B (1996) A 3-year follow-up of stroke patients: relationships between activities of daily living and personality characteristics. Archives of Gerontology and Geriatrics 22, 233-244.

4

Elmståhl S, Persson M, Andren M & Blabolil V (1997) Malnutrition in geriatric patients: a neglected problem? Journal of Advanced Nursing 26, 851-855. Elsner RJ (2002) Changes in eating behavior during the aging process. Eating Behaviours 3, 15-43. Erlanson-Albertsson C (2005) How palatable food disrupts appetite regulation. Basic Clinical Pharmacology and Toxicology 97, 61-73. Evans BC, Crogan NL & Shultz JA (2005) The meaning of mealtimes: connection to the social world of the nursing home. Journal of Gerontological Nursing 31, 11-17. Faxen-Irving G, Andren-Olsson B, af Geijerstam A, Basun H & Cederholm T (2002) The effect of nutritional intervention in elderly subjects residing in group-living for the demented. European Journal of Clinical Nutrition 56, 221-227. Fiatarone Singh MA, Bernstein MA, Ryan AD, O'Neill EF, Clements KM & Evans WJ (2000) The effect of oral nutritional supplements on habitual dietary quality and quantity in frail elders. Journal of Nutrition, Health and Aging 4, 5-12. Flodin L, Svensson S & Cederholm T (2000) Body mass index as a predictor of 1 year mortality in geriatric patients. Clinical Nutrition 19, 121-125. Folstein MF, Folstein SE & McHugh PR (1975) ”Mini-Mental State”, a practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research 12: 189-198. Friedlander AH, Friedlander IK, Gallas M & Velasco E (2003) Late-life depression: its oral health significance. International Dental Journal 53, 41-50. Fulop T, Larbi A, Wikby A, Mocchegiani E, Hirokawa K & Pawelec G (2005) Dysregulation of T-cell function in the elderly: scientific basis and clinical implications. Drugs and Aging 22, 589-603. Garrow JS & Webster J (1985) Quetelet´s Index (W/H2) as a Measure of Fatness. International Journal of Obesity 9, 147-153 Gastman C (1998) Meals in nursing homes: an ethical appraisal. Scandinavian Journal of Caring Sciences 12: 231-237. Gazzotti C, Arnaud-Battandier F, Parello M, Farine S, Seidel L, Albert A & Petermans J (2003) Prevention of malnutrition in older people during and after hospitalisation: results from a randomised controlled clinical trial. Age and Ageing 32, 321-325. Glaser BG & Strauss AL (1967) The Discovery of grounded theory. Strategies for qualitative research. New York: Aldine Publishing Company.

Glaser BG (1978) Theoretical Sensitivity: Advances in the methodology of grounded theory. Mill Valley, California: Sociology Press.

5

Glaser BG (1998) Doing grounded theory: issues and discussions. Mill Valley, California: Sociology Press. Goldspink DF (2005) Ageing and activity: their effects on the functional reserve capacities of the heart and vascular smooth and skeletal muscles. Ergonomics 48, 1334-1351. Guigoz Y, Vellas B & Garry PJ (1996) Assessing the nutritional status of the elderly: The Mini Nutritional Assessment as part of the geriatric evaluation. Nutrition Reviews 54 59-65. Guigoz Y & Vellas B (1999) The Mini Nutritional Assessment (MNA) for grading the nutritional state of elderly patients: presentation of the MNA, history and validation. Nestlé Nutrition Workshop Series Clinical & Performance Programme 1, 3-11 Guigoz Y, Lauque S & Vellas B (2002) Identifying the elderly at risk for malnutrition, the mini nutritional assessment. Clinical Geriatric Medicine 18, 737-757. Gustafsson B & Pörn I (1994) A motivational approach to confirmation: an interpretation of dysphagic patients' experiences. Theoretical Medicine 15, 409-430. Hamrin E & Wohlin E (1982) Evaluation of the functional capacity of stroke patients through an activity index. Scandinavian Journal of Rehabilitation Medicine 14: 93-100. Hudgens J, Langkamp-Henken B, Stechmiller JK, Herrlinger-Garcia KA & Nieves C Jr (2004) Immune function is impaired with a mini nutritional assessment score indicative of malnutrition in nursing home elders with pressure ulcers. Journal of Parenteral and Enteral Nutrition 28, 416-422. Hughes VA, Frontera WR, Roubenoff R, Evans WJ & Singh MA (2002) Longitudinal changes in body composition in older men and women: role of body weight change and physical activity. American Journal of Clinical Nutrition 76, 473-781. Hutchinson SA(1986) Grounded theory: the method. In: Nursing research, a qualitative perspective. (Munhall P, Oiler C. eds) Norwalk: Appleton-Century Crofts, 111-130. Ingenbleek Y & Young VR (2002) Significance of transthyretin in protein metabolism. Clinical Chemistry and Laboratory Medicine 40, 1281-1291. ICD-9. Swedish version of international classification of diseases, 9:th revision. Johnson RA & Tripp-Reimer T (2001) Relocation among ethnic elders. A review--part 2. Journal of Gerontological Nursing 27, 22-27. Kagansky N, Berner Y, Koren-Morag N, Perelman L, Knobler H & Levy S (2005) Poor nutritional habits are predictors of poor outcome in very old hospitalized patients. American Journal of Clinical Nutrition 82, 784-791. Kalender B, Mutlu B, Ersoz M, Kalkan A & Yilmaz A (2002) The effects of acute phase proteins on serum albumin, transferrin and haemoglobin in haemodialysis patients. International Journal of Clinical Practice 56, 505-508.

6

Keller HH (1993) Malnutrition in institutionalized elderly: how and why? Journal of American Geriatric Society 41, 1212-1218. Kenney WL & Chiu P (2002) Influence of age on thirst and fluid intake. Medicine and Science in Sports and Exercise 33, 1524-1532. Kofod J & Birkemose A (2004) Meals in nursing homes. Scandinavian Journal of Caring Sciences 18, 128-134. Kondrup J, Allison SP, Elia M, Vellas B & Plauth M; Educational and Clinical Practice Committee, European Society of Parenteral and Enteral Nutrition (ESPEN) (2003) ESPEN guidelines for nutrition screening 2002. Clinical Nutrition 22, 415-421. Kristensen ST (2000) Social and cultural perspectives on hunger, appetite and satiety. European Journal of Clinical Nutrition 54, 473-478. Kurlowicz L& Wallace M (1999) The Mini-Mental State Examination (MMSE). Journal of Gerontological Nursing 25, 8-9. Kuzuya M, Kanda S, Koike T, Suzuki Y, Satake S & Iguchi A (2005) Evaluation of Mini-Nutritional Assessment for Japanese frail elderly. Nutrition 21, 498-503. Langkamp-Henken B, Hudgens J, Stechmiller JK & Herrlinger-Garcia KA (2005) Mini nutritional assessment and screening scores are associated with nutritional indicators in elderly people with pressure ulcers. Journal of the American Diet Association 105, 1590-1596. Lamy M, Mojon P, Kalykakis G, Legrand R & Butz-Jorgensen E (1999) Oral status and nutrition in the institutionalized elderly. Journal of Dentistry 27, 443-448. Larsson J, Unosson M, Ek A-C, Nilsson L, Thorslund S & Bjurulf P (1990) Effect of dietary supplement on nutritional status and clinical outcome in 501 geriatric patients - a randomised study. Clinical Nutrition 9, 179-184. Lauque S, Arnaud-Battandier F, Mansourian R, Guigoz Y, Paintin M, Nourhashemi F & Vellas B (2000) Protein-energy oral supplementation in malnourished nursing-home residents. A controlled trial. Age and Ageing 29, 51-56. Lengyel CO, Smith JT, Whiting SJ &Zello GA (2004) A questionnaire to examine food service satisfaction of elderly residents in long-term care facilities. Journal of Nutrition for the Elderly 24, 5-18. Lincoln YS, Guba EG. Naturalistic inquiry. 1985, Newbury Park, California: Sage Publications. Lindmark B (1988) Evaluation of functional capacity after stroke with special emphasis on motor function and activities of daily living. Scandinavian Journal of Rehabilitation Medicine Supplement 21, 1-40.

7

Lindmark B & Hamrin E (1988) Evaluation of functional capacity after stroke as a basis for active intervention. Validation of a modified chart for motor capacity assessment. Scandinavian Journal of Rehabilitation Medicine 20, 111-115. Locher JL, Ritchie CS, Roth DL, Baker PS, Bodner EV & Allman RM (2005) Social isolation, support, and capital and nutritional risk in an older sample: ethnic and gender differences. Social Science and Medicine 60, 747-761. Locker D (2002) Changes in chewing ability with ageing: a 7-year study of older adults. Journal of Oral Rehabilitation 29, 1021-1029. MacIntosh C, Morley JE & Chapman IM (2000) The anorexia of aging. Nutrition 16, 983-995. McCormack P (1997) Undernutrition in the elderly population living at home in the community: a review of the literature. Journal of Advanced Nursing 26, 856-863. Mattes RD, Hollis J, Hayes D & Stunkard AJ (2005) Appetite: measurement and manipulation misgivings. Journal of American Dietarian Association 105, 87-97. Milne AC, Potter J & Avenell A (2005) Protein and energy supplementation in elderly people at risk from malnutrition. Cochrane Database Systematic Review 18, CD003288. MNA (2006) User´s Guide to Completing the Mini Nutritional Assessment http://www.mna-elderly.com/clinical-practice.htm (2006-02-06) Morley JE (1996) Anorexia in Older Persons: Epidemiology and optimal treatment. Drugs and Aging 8, 134-155. Morley JE (1997) Anorexia of aging: physiologic and pathologic. American Journal of Clinical Nutrition 66, 760-777. Morley JE (2001) Decreased food intake with aging. Journals of Gerontology A: Biology Sciences and Medical Sciences 56, 81-88. Morley JE (2002) Pathophysiology of anorexia. Clinical Geriatric Medicine 18, 661-673. Mowé M & Bohmer T (2002) Reduced appetite. A predictor for undernutrition in aged people. Journal of Nutrition, Health and Aging 6, 81-83. Nagler RM & Hershkovich O (2005) Age-related changes in unstimulated salivary function and composition and its relations to medications and oral sensorial complaints. Aging: Clinical and Experimental Research 17, 358-366. Nordenfelt L (1991) Quality of life and health: theory & critique. Falköping: Almqvist & Wiksell Förlag AB. (In Swedish). Nordic nutrition recommendations. Köpenhamn: Nordiska Ministerrådet, 1996. (In Swedish).

8

Omran ML & Morley JE (2000a). Assessment of protein energy malnutrition in older persons, Part I: History, examination, body composition, and screening tools. Nutrition 16, 50-63. Omran ML & Morley JE (2000b) Assessment of protein energy malnutrition in older persons, Part II: Laboratory evaluation. Nutrition 16, 131-140. Parker MG, Ahacic K & Thorslund M (2005) Health changes among Swedish oldest old: prevalence rates from 1992 and 2002 show increasing health problems. Journals of Gerontology A: Biological Sciences and Medical Sciences 60, 1351-1355. Perry L & McLaren S (2003) Eating difficulties after stroke. Journal of Advanced Nursing 43, 360-369. Perissinotto E, Pisent C, Sergi G & Grigoletto F; ILSA Working Group (Italian Longitudinal Study on Ageing) (2002) Anthropometric measurements in the elderly: age and gender differences. British Journal of Nutrition 87, 177-186. Pirlich M & Lochs H (2001) Nutrition in the elderly. Best Practice and Research. Clinical Gastroenterology 15, 869-884. Polit D & Beck C (2003) Nursing research, principles and methods. New York: Lippincott. Potter JM, Roberts MA, McColl JH & Reilly JJ (2001). Protein energy supplements in unwell elderly patients - a randomized controlled trial. Journal of Parenteral Enteral Nutrition 25, 323-329. Putwatana P, Reodecha P, Sirapo-ngam Y, Lertsithichai P & Sumboonnanonda K (2005) Nutrition screening tools and the prediction of postoperative infectious and wound complications: comparison of methods in presence of risk adjustment. Nutrition 21, 691-697. Pörn I (1977) Social dynamics. In: Action Theory and Social Science. Dordrecht-Holland: D Riedel, Publishing Company, 83-104. Pörn I (1981) On the dialectic of the soul: An essay on Kirkegaard. Acta Philosophica Fennica 32, 198-210. Pörn I (1984) Kirkegaard and the study of the self. Inquiry 27, 199-205. Ranhoff AH, Gjoen AU & Mowe M (2005) Screening for malnutrition in elderly acute medical patients: the usefulness of MNA-SF. Journal of Nutritions, Health and Aging 9, 221-225. Rasmussen HH, Kondrup J, Ladefoged K & Staun M (1999) Clinical nutrition in Danish hospitals: a questionnaire-based investigation among doctors and nurses. Clinical Nutrition 18, 153-158. Roberts SB, Fuss P, Heyman MB, Evans WJ, Tsay R, Rasmussen H, Fiatarone M, Cortiella J, Dallal GE &Young VR (1994) Control of food intake in older men. Journal of the American Medical Association (JAMA) 272, 1601-1606.

9

Roberts SB (2000) Regulation of energy intake in older adults: recent findings and implications. J Nutr Health Aging 2000; 4: 170-171. Robertson RG & Montagnini M (2004) Geriatric failure to thrive. American Family Physicians 15, 343-350. Rogers PJ (1999) Eating habits and appetite control: a psychobiological perspective. Proceedings of the Nutrition Society 58, 59-67. le Roux CW & Bloom SR (2005) Peptide YY, appetite and food intake. Proceedings of the Nutrition Society 64, 213-216. Rubenstein LZ, Harker J, Guigoz Y & Vellas B (1999) Comprehensive geriatric assessment (CGA) and the MNA: an overview of CGA, nutritional assessment, and development of a shortened version of the MNA. Nestlé Nutrition Workshop Series Clinical & Performance Programme 1, 101-115. Rubenstein LZ, Harker JO, Salva A, Guigoz Y & Vellas B (2001) Screening for undernutrition in geriatric practice: developing the short-form mini-nutritional assessment (MNA-SF). Journal of Gerontology: Medical Sciences 56, 366-372. Saletti A, Lindgren EY, Johansson L & Cederholm T (2000) Nutritional status according to Mini Nutritional Assessment in an institutionalized elderly population in Sweden. Gerontology 46, 139-145. Saletti A, Johansson L, Yifter-Lindgren E, Wissing U, Osterberg K & Cederholm T (2005) Nutritional status and a 3-year follow-up in elderly receiving support at home. Gerontology 51, 192-198. Sarvimäki A & Stenbock-Hult B (2000) Quality of life in old age described as a sense of well-being, meaning and value. Journal of Advanced nursing 32, 1025-1033. Savikko N, Routasalo P, Tilvis RS, Strandberg TE & Pitkala KH (2005) Predictors and subjective causes of loneliness in an aged population. Archives of Gerontology and Geriatric 41, 223-233. Schiffman SS & Graham BG (2000) Taste and smell perception affect appetite and immunity in the elderly. European Journal of Clinical Nutrition 54, 54-63. Seiler WO (2001) Clinical pictures of malnutrition in ill elderly subjects. Nutrition 17, 496-498. Sidenvall B, Fjellstrom C & Ek AC (1996) Cultural perspectives of meals expressed by patients in geriatric care. International Journal of Nursing Studies 33, 212-222. de Silva HA & Gunatilake SB (2002) Mini Mental State Examination in Sinhalese: a sensitive test to screen for dementia in Sri Lanka. International Journey of Geriatric Psychiatry 17, 134-139.

10

Sjölenius B., Hälso- och sjukvård i kommunerna. Inför 2000-talet, (Health care in the municipalities. At the dawn of the 21st Century). Stockholm: Kommentus Förlag, 1997. Skoog I (2004) Psychiatric epidemiology of old age: the H70 study--the NAPE lecture 2003. Acta Psychiatrica Scandinavica 109, 4-18. SOSFS 1993:17. The National Board of Health and Welfare. General guidelines for nursing care. (Socialstyrelsen. Socialstyrelsens allmänna råd (SOSFS 1993:17) om omvårdnad inom hälso- och sjukvården) (In Swedish). SOSFS: 1997:10. The National Board of Health and Welfare. General guidelines for the medical responsible nurse within health and care in the municipalities. (Socialstyrelsen. Socialstyrelsens allmänna råd (SOSFS 1997:10) om medicinskt ansvarig sjuksköterska i kommunernas hälso- och sjukvård) (In Swedish). Statistics Sweden (2002) Statistical year book 2000. (Statistiska centralbyrån. Statistisk årsbok 2000), Örebro (In Swedish). Stern PN (1994) Eroding Grounded Theory. In: Critical issues in qualitative research methods. (Morse JM. ed) Thousands Oaks, California: Sage Publications, 212-223. Strömberg A & Mårtensson J (2003) Gender differences in patients with heart failure. European Journal of Cardiovascular Nursing 2, 7-18. Sundström G, Johansson L & Hassing LB (2002) The shifting balance of long-term care in Sweden. Gerontologist 42, 350-355. Suominen M, Muurinen S, Routasalo P, Soini H, Suur-Uski I, Peiponen A, Finne-Soveri H & Pitkala KH (2005) Malnutrition and associated factors among aged residents in all nursing homes in Helsinki. European Journal of Clinical Nutrition 59, 578-583. Swedish Research Council (2003) Guidelines for ethical assessment of medical research on humans. Policy and organisation of research ethics in Sweden. Stockholm. (Vetenskapsrådet. Riktlinjer för etisk värdering av medicinsk humanforskning. Forskningsetisk policy och organisation i Sverige.) (In Swedish). Symreng T (1982) Arm anthropometry in a large reference population and in surgical patients. Clinical Nutrition 1, 211-219. Terre R & Mearin F (2006) Oropharyngeal dysphagia after the acute phase of stroke: predictors of aspiration. Neurogastroenterology and Motility 18, 200-205. The National Board of Health and Welfare (1996) The Ädel-reform - final report, 1996:2 (Socialstyrelsen. Ädelreformen - Slutrapport, 1996:2) Stockholm (In Swedish) The National Board of Health and Welfare (2000) The Ädel-reform - final report, 2000:2 (Socialstyrelsen. Ädelreformen - Slutrapport, 2000:2) Stockholm (In Swedish)

11

The Swedish National Food Administration (1998) Food and Dietary Management in Elderly People. Problems and Opportunities. (Livsmedelsverket. Mat och kostbehandling för äldre. Problem och möjligheter.) Ed. Andersen M, Uppsala. (In Swedish). Todorovic V (2001) Detecting and managing undernutrition of older people in the community. British Journal of Community Nursing 6, 54-60. Tognoni G, Ceravolo R, Nucciarone B, Bianchi F, Dell'Agnello G, Ghicopulos I, Siciliano G & Murri L (2005) From mild cognitive impairment to dementia: a prevalence study in a district of Tuscany, Italy. Acta Neurological Scandinavia 112, 65-71. Tracy RP (2003) Emerging relationships of inflammation, cardiovascular disease and chronic diseases of aging. International Journal of Obese Related Metabolic Disorders 27, 29-34. Turnheim K (2004) Drug therapy in the elderly. Experimental Gerontology 39, 1731-1738. Unosson M, Ek AC, Bjurulf P & Larsson J (1991) Demographical, sociomedical and physical characteristics in relation to malnutrition in geriatric patients. Journal of Advanced Nursing 16, 1406-1412. Unosson M, Ek A-C, Bjurulf P, von Schenck H & Larsson J (1995) Influence of macro-nutrient status on recovery after hip fracture. Journal of Nutritional Environmental Medicine 5, 23-34. Walls AW & Steele JG (2004) The relationship between oral health and nutrition in older people. Mechanisms of Ageing and Development 125, 853-857. Wang J, Eslinger PJ, Smith MB &Yang QX (2005) Functional magnetic resonance imaging study of human olfaction and normal aging. Journals of Gerontology A: Biological Sciences and Medical Sciences 60, 510-514. Van Nes MC, Herrmann FR, Gold G, Michel JP & Rizzoli R (2001) Does the mini nutritional assessment predict hospitalization outcomes in older people? Age and Ageing 30, 221-226. Warnold I & Lundholm K (1984) Clinical significance of preoperative nutritional status in 215 non cancer patients. Annals of Surgery 199, 299-305. Weigley E (1994) The Body Mass Index in clinical practice. Top Clinical Nutrition 9, 70-75. Vellas B, Guigoz Y, Garry PJ, Nourhashemi F, Bennahum D, Lauque S & Albarede JL (1999) The Mini Nutritional Assessment (MNA) and its use in grading the nutritional state of elderly patients. Nutrition 15, 116-122. Westergren A, Karlsson S, Andersson P, Ohlsson O & Hallberg IR (2001) Eating difficulties, need for assisted eating, nutritional status and pressure ulcers in patients admitted for stroke rehabilitation. Journal of Clinical Nursing 10, 257-269. Westergren A, Ohlsson O & Hallberg IR (2002) Eating difficulties in relation to gender, length of stay, and discharge to institutional care, among patients in stroke rehabilitation. Disability of Rehabilitation 10, 523-533.

12

Vetta F, Ronzoni S, Taglieri G & Bollea MR (1999) The impact of malnutrition on the quality of life in the elderly. Clinical Nutrition 18, 259-267. Whalley LJ, Deary IJ, Appleton CL & Starr JM (2004) Cognitive reserve and the neurobiology of cognitive aging. Ageing Research Review 3, 369-382. Whitney EN, Cataldo CB & Rolfes SR (2002) Understanding normal & clinical nutrition. 6th Ed. St Paul: West Publishing Company. Wilkinson T & de Picciotto J (1999) Swallowing problems in the normal ageing population. South African Journal of Community Disorders 46, 55-64. Wilson MM & Morley JE (2003) Impaired cognitive function and mental performance in mild dehydration. European Journal of Clinical Nutrition 57, 24-29. Wikby A, Johansson B & Fergusson F (2003) The OCTO and NONA immune longitudinal studies: a review of 11 years studies of Swedish very old humans. Advances in Cell Aging and Gerontology 13, 1-16. Visvanathan R (2003) Under-nutrition in older people: a serious and growing global problem! Journal of Postgradual Medicine 49, 352-360. Wouters-Wesseling W, Wouters AE, Kleijer CN, Bindels JG, de Groot CP & van Staveren WA (2002) Study of the effect of a liquid nutrition supplement on the nutritional status of psycho-geriatric nursing home patients. European Journal of Clinical Nutrition 56, 245-251. Ödlund Olin A, Koochek A, Ljungqvist O & Cederholm T (2005) Nutritional status, well-being and functional ability in frail elderly service flat residents. European Journal of Clinical Nutrition 59, 263-270.