Download - GREAT Lecture1 Epid Study Design Handout
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 1/83
GREAT Network Training Program
2013
Lecture 1Epidemiology and Study Designs
Instructor: Dr. Jason Pole
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 2/83
GREAT Network Epidemiology and Surveillance Program
Epidemiology and Study Designs Lecture
Presented by Jason D. Pole, PhD
Objectives:
By the end of this session, participants will be able to define and distinguish between the
following study designs: cohort, randomized control trial, case control, prospective, retrospectiveand cross-sectional. Participants will also be able to define the following terms in relation to
these study designs: population perspective, clinical perspective, incidence, prevalence,confounding, causation, risk, bias, odds ratio, relative risk and person-years.
Background:
Before the lecture please read the following:
(1) Chapter 4 - Epidemiology An Introduction by Kenneth J. Rothman, 2002, OUP
(2) Chapter 11 - Epidemiology An Introduction by Kenneth J. Rothman, 2002, OUP
(3) Priftis KN, D. Mermiri, A Papadopoulou, M. Papadopoulos, A Fretzayas and E Lagona.Asthma Symptoms and Bronchial Reactivity in School Children Sensitized to Food Allergens in
Infancy. Journal of Asthma, 2008, 45:590-595.
(4) Kiechl-Kohlendorfer U, E Horak, W Mueller, R Strobi, C Haberland, F-M Fink, MSchwaiger, K-H Gutenberger, H Reich, D Meraner and S Kiechl. Neonatal characteristics andrisk of atopic asthma in schoolchildren: results from a large prospective birth-cohort study. Acta
Paediatrica, 2007, 96:1606-1610.
The information in the two chapters will largely be covered as part of the lecture, but youwill be able to follow the lecture more easily if you have read the chapters a head of time. You
only need to read the abstract, introduction and methods sections of the two papers. We will
discuss these papers during the lecture.
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 3/83
LECTURE POWERPOINT
PRESENTATION
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 4/83
GREAT Network:
Epidemiology and Surveillance Program
Epidemiology Methods Lecture
Jason D. Pole, PhD
Pediatric Oncology Group of Ontario
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 5/83
Outline
• Epidemiology Perspectives
• Random Pieces (well, not exactly random)
• Study Designs
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 6/83
Epidemiology:
Population versus Clinical• Population Epidemiology
– study of distribution and determinants of disease in groups of
people
– Historical focus on biological and social determinants of health
– Now includes health service and health policy
• Clinical Epidemiology
– epidemiological methods and biostatistics applied to research in
health care practice – focus on
• evaluating diagnostic value of new tests
• evaluating value of treatments and procedures
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 7/83
Distribution
• Person
– age, gender, race, socio-economic status etc.
• Place
– census tract, county, province, country, urban / rural etc.
• Time
– year, season etc.
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 8/83
Example: Mystery Epidemic
Socio-Economic
Class N
Death
Rate (%) N
Death
Rate (%) N
Death
Rate (%) N
Death
Rate (%)
High 173 66.5 144 3.5 5 0 322 37.3Middle 160 91.9 93 16.1 24 0 277 58.5
Low 454 87.9 179 45.3 76 71.1 709 75.3
Unknown 875 78.4 23 8.7 0 -- 898 76.6
Total 1662 81 439 23.5 105 51.4 2206 68.2
Adult Males Adult Females Children Total
Note: N refers to the total numbers of persons in each strata
Questions:
1. Describe the epidemiological features of this episode.2. Based on the descriptive characteristics:
formulate a hypothesis concerning the etiology of this episode.
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 9/83
Random Pieces
• Incidence Rate
• Prevalence• Person-Time
• Causation
• Relative Risk• Odds Ratio
• Bias
• Confounding
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 10/83
Incidence Rate
• Number of new cases of an outcome that occurs during
a specified period of time in a defined population at risk
for developing the outcome divided by the number of
persons in the population at risk
• Time (required for a rate)
• New cases (numerator)
• Specified population at risk (denominator)
– those in the denominator must have potential to be counted in numerator
• Incidence of Asthma in Ontario:
– 10 per 1000 population aged 0-39 years in 2001
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 11/83
Prevalence
• Proportion of persons in a defined population who had
the outcome at some point during the time period
• Proportion and not a rate!
• Number of existing cases (numerator)
• Number of persons in specified population (denominator)
• Prevalence of Asthma in Ontario:
– 58 per 1000 population aged 0-39 years in 2001 (5.8 %)
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 12/83
Person-Time
• Often rates are described not per person but for some
unit of person time
• Useful when people are followed for different periods of
time
• Implicit assumption that observing 10 persons for 10
years is equivalent to observing 100 persons for 1 year
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 13/83
Causation
• Strength
• Consistency• Specificity
• Temporality
• Biological Gradient (dose response)• Plausibility
• Coherence
• Experimental Evidence
• Analogy (other possible explanations)
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 14/83
Relative Risk
• RR = incidence in group A / incidence in group B
• RR is a ratio so dimensionless (units cancel)
• Required to make clear which groups are beingcompared (it is relative)
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 15/83
Relative Risk Calculation
Outcome Develops Outcome Didn’t Develop
Group A (often exposed) a b
Group B (often nonexposed) c d
Iexp = a/(a+b)
Inonexp = c/(c+d)
RR = Iexp / Inonexp
Those who are exposed are *.* times as likely to develop the outcome
compared to those who are not exposed.
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 16/83
Relative Risk Interpretation
• RR = 1
– risk in group A is same as in group B
– no association
• RR > 1
– risk in group A is larger compared to group B
– positive association
– larger the RR the stronger the association
– does not infer causation
• RR < 1
– risk in group A is smaller compared to group B
– negative association
– smaller the RR the stronger the negative association
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 17/83
Odds Ratio
• Approximates the RR if the disease is rare in the
unexposed population (probability of event is < 5%)
• OR = odds that those with outcome were exposed / odds
that those without the outcome were exposed
• OR is a ratio so dimensionless (units cancel)
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 18/83
Odds Ratio Calculation
Outcome Develops
Cases
Outcome Didn’t Develop
Controls
Exposed a b
Nonexposed c d
Odds of Exposurecase = a/(a+c) = a/c
Odds of Exposurecontrols = b/(b+d) = b/d
OR = Odds of Expcase / Odds of Expcontrol = (a/c)(b/d) = ad/bc
Those who develop the outcome are *.* times as likely to be exposed
compared to those who did not develop the outcome.
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 19/83
Bias
• Systematic error in design or conduct that results in a
distortion of the estimate of association
• Three main categories:
– selection
• design stage
• systematic inclusion or exclusion of participants by any mechanism
– information
• study execution stage
• systematic difference in measurement
– confounding
See: Choi BC and AL Noseworthy. 1992. Classification, direction, and prevention of bias in
epidemiological research. JOM 34:3 pg 265-271
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 20/83
Confounding
• Specific type of bias
• Distortion of the association of interest by a third variable
• Confounder – can not be in causal pathway
– must be associated with both the exposure and outcome
OutcomeExposure
Confounder
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 21/83
Pause…
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 22/83
Epidemiology Study Designs
• Experimental Studies
– Randomized Control Trials
– Community Trials
• Observational Studies
– Cohort Study
– Case-Control Study – Cross-Sectional Study
– Ecological Study
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 23/83
Randomized Controlled Trials
• Features
– random assignment of subjects to 2 or more groups
– one group acts as a ‘control’ – other groups are provided ‘new’ treatment
– groups followed over time for rates of outcome
– the groups only differ by the treatment
– temporality established
• Regarded as most scientifically rigorous method
• RCTs generally test efficacy not effectiveness!
– analyze as randomized!
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 24/83
RCT Design
Eligible
Subjects
Treatment GroupOutcome Present
Outcome Not Present
Control Group
Outcome Present
Outcome Not Present
Randomization
Time
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 25/83
RCT Advantages & Disadvantages
• Advantages
– randomization
– concurrent control group
– blind assessment
• subject, outcome assessor, treatment administrator
• Disadvantages
– expensive (time, money, people)
– potential ethical issues – issues with representativeness
– can’t be used to study harmful exposures
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 26/83
Pause…
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 27/83
Cohort Design
• Features
– starts with a subject pool that does not have outcome
– subjects self-select into exposure group
– entire subject pool followed over time for rates of outcome
– temporality established
• Researcher has no control over changes in exposure
• Open versus Closed Cohort
• Among observational studies, they are the gold standard
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 28/83
Cohort Design
Eligible
Subjects
Exposed GroupOutcome Present
Outcome Not Present
Unexposed Group
Outcome Present
Outcome Not Present
Self Selection
Time
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 29/83
Cohort Advantages & Disadvantages
• Advantages – minimizes bias
• recall and information – prospective - temporality
– limited ethical issues (you don’t have to withhold treatment)
– can study several outcomes in same study
• Disadvantages – expensive (time, money, people)
– potential for confounding – loss of follow-up
– not good for rare outcomes
– sample size of exposed and unexposed group is not controlled
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 30/83
Paper Discussion: Kiechl-Khlendorfer et al.
• What is the purpose / objective of the study?
• What type of study design has been used?
• Does it make sense to do this study given the literature / past studies (i.e. isthere a clear rationale, or gap in the literature, to justify this study?)
• How is the study population defined?• What criteria were used to define the study sample?
• Did the authors address the size of the sample needed to show the desiredresults?
• Did the authors obtain the consent of participants? If so, how?
• Was it clear how the participants were followed and how long they neededto be followed?
• What was the nature of the exposed and unexposed groups?
• What was the health outcome and how was it defined / measured?
• Did the authors take into account any potential cofounders? If so, how?• Did the authors approach the analysis in a way that would answer the
research question?
• What types of bias might be present in the study?
• What are the advantages of using this study design?
• Are there alternative study designs to address the question?
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 31/83
Pause…
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 32/83
Case-Control Study
• Features
– start with all cases in population of interest
– identify a set of controls
• from the same population that gave rise to the cases
– retrospectively assess exposure
• Selection of control group can be problematic
• Potential to introduce bias
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 33/83
Case-Control Design
EligibleSubjects
Cases(those with outcome)
Exposed
Not Exposed
Controls
(those without outcome)
Exposed
Not Exposed
Time
Present
Case Control
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 34/83
Case-Control
Advantages & Disadvantages• Advantages
– relatively cheap to conduct
– concurrent control group
• the investigator controls the size of the two groups
– good for rare outcomes
• Disadvantages
– difficult to obtain retrospective exposure information
• potential to introduce bias
– difficult to select appropriate control group
• potential to introduce bias
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 35/83
Paper Discussion: Priftis et al.• What is the purpose / objective of the study?
• What type of study design has been used?
• Does it make sense to do this study given the literature / past studies (i.e. isthere a clear rationale, or gap in the literature, to justify this study?)
• How is the study population defined?• What criteria were used to define the study sample?
• Did the authors address the size of the sample needed to show the desiredresults?
• Did the authors obtain the consent of participants? If so, how?
• What were the outcome groups and how was the outcomedefined/measured?
• What was the nature of the exposed and unexposed groups?
• Did the authors take into account any potential cofounders? If so, how?
• Did the authors approach the analysis in a way that would answer theresearch question?
• What types of bias might be present in the study?
• What are the advantages of using this study design?
• Are there alternative study designs to address the question?
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 36/83
READING MATERIALS
This article was downloaded by: [University of Toronto]
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 37/83
PLEASE SCROLL DOWN FOR ARTICLE
This article was downloaded by: [University of Toronto] On: 15 October 2008 Access details: Access Details: [subscription number 793222078] Publisher Informa Healthcare Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK
Journal of AsthmaPublication details, including instructions for authors and subscription information:http://www.informaworld.com/smpp/title~content=t713597262
Asthma Symptoms and Bronchial Reactivity in School Children Sensitized toFood Allergens in Infancy
Kostas N. Priftisa
; Despina Mermiria
; Athina Papadopouloua
; Marios Papadopoulosa
; Andrew Fretzayasb
;Evagelia Lagona c
a Department of Allergy-Pneumonology, Penteli Children's Hospital, Athens, Greece b Third Department ofPediatrics, Attikon Hospital, Athens University Medical School, Athens, Greece c First Department ofPediatrics, “Aghia Sophia” Children's Hospital, Athens University Medical School, Athens, Greece
Online Publication Date: 01 September 2008
To cite this Article Priftis, Kostas N., Mermiri, Despina, Papadopoulou, Athina, Papadopoulos, Marios, Fretzayas, Andrew and Lagona,Evagelia(2008)'Asthma Symptoms and Bronchial Reactivity in School Children Sensitized to Food Allergens in Infancy',Journal ofAsthma,45:7,590 — 595
To link to this Article DOI 10.1080/02770900802032941
URL http://dx.doi.org/10.1080/02770900802032941
Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf
This article may be used for research, teaching and private study purposes. Any substantial orsystematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply ordistribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation that the contentswill be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses
should be independently verified with primary sources. The publisher shall not be liable for any loss,actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directlyor indirectly in connection with or arising out of the use of this material.
Journal of Asthma, 45:590–595, 2008Copyright C 2008 Informa Healthcare USA, Inc.
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 38/83
Copyright 2008 Informa Healthcare USA, Inc.ISSN: 0277-0903 print / 1532-4303 onlineDOI: 10.1080/02770900802032941
ORIGINAL ARTICLE
Asthma Symptoms and Bronchial Reactivity in School Children Sensitizedto Food Allergens in Infancy
KOSTAS N. PRIFTIS,1 DESPINA MERMIRI,1 ATHINA PAPADOPOULOU,1 MARIOS PAPADOPOULOS,1
ANDREW FRETZAYAS,2AND EVAGELIA LAGONA
3
1 Department of Allergy-Pneumonology, Penteli Children’s Hospital, Athens, Greece2Third Department of Pediatrics, Attikon Hospital, Athens University Medical School, Athens, Greece
3First Department of Pediatrics, “Aghia Sophia” Children’s Hospital, Athens University Medical School, Athens, Greece
Food allergy in infancy usually disappears but is followed primarily by respiratory allergy. We hypothesized that children allergic to common food
allergens in infancy are at increased risk of wheezing illness and bronchial hyperresponsiveness during school age. In a c ase-control study 69 children
7.2 to 13.3 years of age allergic to egg (N = 60) and/or fish (N = 29) in early life (first 3 years) who attended our allergy outpatient clinic were
recruited. They received follow-up for 1 year and were evaluated by parental questionnaire, skin prick testing, spirometry, and metacholine bronchial
challenge. Another 154 children (70 sensitized to inhaled allergens) recruited selectively from a general population sample with no history of food
allergy during their first 3 years served as control subjects. Twenty-three children (38.3%) maintained their sensitization to egg and 19 (65.5%) to fish;
the prevalence of sensitization to ≥1 inhaled allergen(s) increased from 59.4% to 71% during childhood. Current asthma symptoms were reported
more frequently in the study group than in either control groups, sensitized to inhaled allergens and non-sensitized. Children of the study group
showed a significantly increased frequency of positive response to metacholine bronchial challenge compared to the control group as a whole; thedifference was statistically indicative when study groups separately were compared to the sensitized control subjects. Multivariate logistic regression
analysis showed that bronchial hyperresponsiveness, as well as reported current asthma symptoms were associated with early wheezing and early
sensitization to inhaled allergensbut not with atopicdermatitis in infancy or persistence of eggor fishallergy. Childrenallergicto egg or fishin infancy
are at increased risk for wheezing illness and hyperactive airways in school age; asthma and bronchial hyperresponsiveness development is mostly
determined by wheezing and senzitization to inhaled allergens in early life regardless of atopic dermatitis in infancy or retention of food allergy.
Keywords allergic march, asthma, atopic dermatitis, bronchial hyperresponsiveness, food allergy, inhaled allergens
INTRODUCTION
During the last few decades, a worldwide phenomenonof increase in the prevalence of asthma in parallel to otherallergic conditions such as food allergy, rhinitis, and eczemahas been consistently observed (1, 2). Atopic dermatitis andfood allergy often begin in infancy and there appears to bean onward trend that leads to asthma and allergic rhinitisin school age or early adulthood (3). Prevention strategieshave mainly targeted the interruption of the allergic trend (4);therefore, knowledge of specific predictors in early infancy isa prerequisite if preventive intervention is being considered.
Sensitization to hen’s egg in early life has been proposed asa predictor of subsequent sensitization to inhalant allergensthat in itself constitutes a strong determinant of asthma (5, 6).Early onset of immunoglobulin (IgE)-mediated food allergy,especially if associated with atopic dermatitis, indicates a Th2
dominant lymphocytic response pattern and a cytokine pro-
of life demonstrating a clear temporal relationship with the
introduction of these foods intothe children’s diet. Over time,most food allergy is lost, although the possibility of suchloss depends on the individual child and the specific foodallergen. In contrast to cow’s milkand egg, allergies to fishareusually not outgrown (8). It is not clear if infants who presentwith food allergies with different natural history would tenddifferently to develop asthma at school age.
The mechanisms that predispose children with food allergyto develop asthma at a later age remain poorly understood.It is speculated that the immunologic basis of specific or-
gan syndromessuch as allergic rhinitis and asthma actuallyresults from a systemic dysregulationof immunity (9, 10);eczema and at a later age hyperreactive airways could bemanifestations of different target organs within the frame of the same systemic disease. Most studies that address this is-sue are observational and focus on the so-called “allergic
h ” i h li i l i f h b i i h
D o w nl
o ad
ed
B y : [
U ni
v e r si
t y
of
T o r o n t o]
A t : 21
:20 1
5
O c t ob
e r 2008
FOOD ALLERGY, ASTHMA AND BRONCHIAL REACTIVITY 591
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 39/83
TABLE 1.—General characteristics of the study population (study group and two
control groups).
Study group
N = 69
Sensitisedcontrols
N = 70
Non-sesnsitisedcontrols
N = 84
Male (%) 52 (75.4) 44 (62.8) 50 (59.5)Median age, years (range) 9.9 (7.2–13.3) 10.3 (8.0–13.5) 10.0 (7.5–13.1)Atopic family history (%) 41 (59.4)∗ 35 (50.0)¶ 21 (25.0)Passive smoking (%) 45 (65.2) 46 (65.7) 56 (66.7)Cent ral house heat ing (%) 59 (85.5) 56 (80.0) 67 (79.8)Lifetime asthma symptoms (%) 38 (55.1)∗# 20 (28.5)¶ 9 (10.7)Current asthma symptoms (%) 30 (43.5)∗# 17 (24.3)¶ 5 (5.9)
Lifetime chronic rhinitis (%) 27 (39.1)§ 23 (32.8) 19 (22.6)
Current chronic rhinitis (%) 20 (29.0)§ 21 (30.0)§ 13 (15.5)Lifetime atopic dermatitis (%) 38 (55.1)∗# 20 (28.5) 13 (15.5)Current atopic dermatitis (%) 7 (10.1)# 19 (27.1) 8 (9.5)
∗In comparison to non-sensitised controls: p < 0.001.#In comparison to sensitised controls: p < 0.01.¶In comparison to non-sensitised controls: p < 0.01.§In comparison to non-sensitised controls: p < 0.05.
PATIENTS AND METHODS
Study population
The present case-control study evaluated three groups of children (Table 1). The study group consisted of 69 school
age children who presented to our outpatient allergy clinicduring the first 3 years of life for a variety of symptoms fromJanuary 1995 until the end of 1998. The diagnosis of foodallergy was based on a positive skin prick test (SPT) resultor a positive serum-specific IgE test to hen’s egg or/and fish(>0.35 IU/mL), a well-documented history of reaction torelevant food(s), and, in 13 cases, on immediate symptomsafter open challenge with suspected food. Sixty children wereallergic to hen’s egg (egg white and yolk) and 29 to fish (20were sensitized to both). Open food challenges were also
performed when appropriate to investigate development of tolerance (14, 15).
Study inclusion criteria were (1) symptomatic sensitiza-tion to hen’s egg and/or fish during the first 3 years of life,(2) age on recruitment over 7 years, and (3) ability to per-form reproduciblespirometry according to the standards of the American Thoracic Society (16). Children with diag-nosed or suspected cystic fibrosis, primary ciliary dyskinesia,myoskeletal abnormalities, immunodeficiency, and history of bronchopulmonary dysplasia were excluded.
Study Design
Follow-up and review of patients and controls were per-formed between February 2003 and September 2005. A de-tailed medical and socialhistory, physical examination, SPTs,spirometry, methacholine challenge test (MCT) when feasi-
expiratory volume in one second (FEV1), and forced expi-ratory flow at 50% FVC (FEF50) were recorded. Spirometrywas deferred if the child had a respiratory tract infectionwithin 3 weeks or had used a bronchodilator within 12 hoursof assessment for study purposes. Values were expressed aspercentage of predicted for gender and height (17).
Assessment of BHR was performed only when FEV1 val-ues were more than 70% of the predicted value. Childrenunderwent MCT according to the American Thoracic Soci-ety guidelines (18). The short protocol of five tidal breathswas performed. The procedure was discontinued if FEV1
decreased by more than 20% of baseline values or whena 16 mg/mL concentration had been administered. Asthmamedication such as short- and long-acting beta2-agonists,
leukotriene modifiers, or chromones were withheld for 24to 48 hours before testing; anti-histamines were also discon-tinued for an appropriate period depending on the pharma-cokinetic characteristics of the various medications. Inhaledcorticosteroids were continued as prescribed. Spirometry andMCT were performed during asymptomatic periods (a min-imum of 6 weeks after resolution of any acute respiratorysymptomatology), between 9 A.M. and 1 P.M.
Atopic status was assessed by skin-prick testing (SPT).They were performed on the volar aspects of the forearm
with eight common standardized inhaled allergen extractssuch as house dust mites ( Dermatophagoides pteronyssinus,
Dermatophagoides farinae), Alternaria alternata, olive tree,mixed grasses and Parietaria officinalis pollens, cat and dogdander; another battery of five common food standardized al-lergenextracts,i.e., cow’s milk, hen’s egg, mixed cereals, fish,and mixed nuts, was tested (SoluprickTM, ALK, Hørsholm,Denmark). Histamine dihydrochloride 10 mg.mL−1 was usedas the positive and 50% glycerol as the negative control. Theprick tests were performed according to the instructions of
the European Academy of Allergy and Clinical Immunology(EAACI) (19). Children were considered to be sensitized if a palpable wheal reaction to any allergen, calculated as thesum of the longest and the midpoint orthogonal diametersdivided by two, was ≥3 mm larger than the negative control.
Children that belonged to the study group were followedprospectively by clinical assessment every 4 months for ayear to establish their respiratory and skin manifestations;at the third visit all children were evaluated by SPTs andspirometry, whereas 57 out of 69 underwent MCT to assess
BHR; 4 refused and 8 were unable to cooperate.Subjects, matched for age and sex with the study group,
with a negative history of food allergy in the first 3 yearsof life were selected from a general population sample of Athenian schoolchildren and assigned to two control groups.One consisted of 70 children sensitized to inhaled allergens
D o w nl
o ad
ed
B y : [
U ni
v e r si
t y
of
T o r o n t o]
A t : 21
:20 1
5
O c t ob
e r 2008
592 K. N. PRIFTIS ET AL.
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 40/83
TABLE 2.—Clinical characteristics of study group in the first 3 years of life; the OR (95% CI) refers to sensitized only to egg and only to fish subjects.
Total
N = 69∗Sensitised only
to egg N = 40
Sensitised only
to fish N = 9 OR (95% CI)
Mean age at diagnosis of sensitisation (mo) 20.6± 10.3 19.5 ± 10.5 21.2 ± 3.6 0.97 (0.89–1.05)Mean age at first clinical presentation (mo) 10.7± 4.3 10.1 ± 5.2 12.4 ± 3.6 1.15 (0.97–1.37)Presenting symptoms (%)
Urticaria/anaphylaxis 46 (66.6) 29 (74.4) 7 (77.8) 0.77 (0.14–4.07)Atopic dermatitis 30 (43.5) 16 (40.0) 2 (22.2) 0.48 (0.07–3.32)Gastrointestinal symptoms 12 (17.3) 5 (12.5) 4 (44.4) 5.83 (1.0–33.77)¶
Respiratory distress 7 (10.3) 3 (7.5) 2 (22.2) 1.52 (0.11–20.76)Sensitisation to other food allergens (%) 43 (62.3) 15 (37.5) 8 (88.9)¶ 13.9 (1.48–131.17)
Milk 11 (15.9) 6 (15.0) 1 (11.1) 0.45 (0.02–8.34)Nuts 16 (23.2) 4 (10.0) 6 (66.5)# 66 (3.89–1146.47)Cereals 12 (17.4) 4 (10.0) 1 (11.1) 1.07 (0.08–13.41)Shellfish 8 (11.6) 1 (2.5) 4 (44.4)# 43 (2.84–674.90)Fruits 9 (13.0) 2 (5.0) 2 (22.2) 4 (0.32–42.83)Vegetables 6 (8.7) 3 (7.5) 1 (11.1) 1.08 (0.07–15.65)
Sensitisation to inhaled allergens (%) 41(59.4) 20 (50.0) 4 (44.4) 0.81 (0.13–4.94)
Mites 18 (26.1) 8 (20.0) 0 NAMolds 10 (14.5) 5 (12.5) 0 NAPollen 28 (40.5) 13 (32.5) 4 (44.4) 1.9 (0.26–14.82)Pets 13 (18.8) 12 (30.0) 1 (11.1) 0.22 (0.02–2.74)
∗20 cases were sensitized to both egg and fish protein; NA=non applicable.¶ p < 0.05; # p < 0.01.
exact test was performed for categorical variables; Student’st or Mann-Whitney U tests were suitable for continuous para-metric and non-parametric variable. The characteristics of
children of the primary study group, i.e., sensitized to hen’segg versus fish protein at presentation, were compared sepa-rately and odds ratios with 95% confidence interval (CI) werecalculated.Logistic regression analysis wasused to assess theindependent effects of various risk factors (maternal smoking,atopic family history, wheezing in infancy, sensitization to in-haled or food allergens allergens, current or lifetime asthmasymptoms, chronic rhinitis, and atopic dermatitis) on BHR.Significance was two tailed and set at p ≤ 0.05.
RESULTS
The main characteristics of the study population (studygroup, controls) are presented in Table 1. Atopic familyhistory, lifetime asthma symptoms, lifetime-current chronicrhinitis and lifetime atopic dermatitis reported by studygroup differed significantly from non-sensitized controls;current asthma symptoms were reported significantly morefrequently in the study group than in either control groups.Current atopic dermatitis was found to be more frequentin sensitized versus study group and non-sensitized con-
TABLE 3.—Sensitization to inhaled and/or food allergens at school-age in study group and sensitized controls.
Study group
Sensitised controls
Sensitized only to egg N = 40 Sensitized only to fish N = 9 OR (95% CI) Total N = 69* N = 70 OR (95% CI)
trols. Seventeen (56.7%) current asthmatics of study groupand 4 (23.5%) of sensitized controls were on mainte-nance anti-asthma treatment (OR 0.23, 95%CI 0.06-0.89,
p = 0.04).The presenting symptoms were reactions to relevant food
in 45 (65.2%) children of thestudygroup. Theclinical charac-teristics ofthe 69children in the first 3 years of life are shownin Table 2. Sixty-two (89.8%) presented with skin manifesta-tions (atopic dermatitis and/or urticaria) and 29 (42.0%) re-ported wheezing before 3 years of age, which was associatedto sensitization to inhaled allergens by that time ( p = 0.001).
At school age, 38 (63.3%) cases had developed tolerance toegg and were able to consume it freely; the one child that re-
tained positive SPT for egg was also able to consume it freelyas well. The median age of egg tolerance was 2.1 (1.0-7.4)years. Ten outof 29 (34.5%) children lost their sensitizationtofish but only 5 (17.2%) of these children developed tolerance;however, tolerance to fish occurred at a significantly older ageas compared to egg, i.e., 5.0 (2.5–8.2) years ( p = 0.004).
The prevalence of sensitization to inhaled and food aller-gens during school age in the study group and sensitized con-trols is shown in Table 3. Sensitization to molds was detected
D o w nl
o ad
ed
B y : [
U ni
v
e r si
t y
of
T o r o n t o]
A t : 21
:20 1
5
O c t ob
e r 2008
FOOD ALLERGY, ASTHMA AND BRONCHIAL REACTIVITY 593
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 41/83
TABLE 4.—Spirometry (mean ± SD) and BHR in the three groups of the study.
Study group
N = 69*
Sensitised controls
N = 70
Non-sensitised
controls N = 84
FVC (% predicted) 94.5± 8.2 96.2 ± 9.1 99 ± 6.6FEV1 (% predicted) 98.4± 8.3 99.3 ± 7.2 97.7 ± 7.1FEF50 (% predicted) 87.7± 16.9 75.9 ± 16.5 83 ± 10.6Positive metacholine 31 (54.4)# 28 (40.0)# 8 (9.5)challenge test (%)
∗Methacholine bronchial challenge test performed in 57 out of 69 children.¶ In comparison to either study group or non-sensitized controls: p < 0.01.#In comparison to non-sensitized controls: p < 0.001.
more frequently in the study group than among sensitizedcontrols.
All children of the study group exhibited normal pul-monary function tests (Table 4). A significant decrease of MEF50 values was detected among controls sensitized to in-haled allergens as compared to the study group. Multivariateanalysis of spirometric indices and specific sensitization re-vealed an independent association of sensitisation to miteswith MEF50 values (OR: 0.97, 95%CI: 0.95-1.0, p = 0.04).
Study group showed a significantly increased frequency of positive response to MCT when compared with the controlsas a whole (OR: 0.26, 95%CI: 0.13-0.49, p < 0.001); the
difference was statistically insignificant when study groupsseparatelywere compared with the sensitizedcontrol subjects(OR: 0.56, 95%CI: 0.28–1.12, p = 0.07).
The results of multivariate logistic regression analysis be-tween BHR and current asthma symptoms (dependent vari-ables) in the index case group and relevant clinical charac-teristics of children during infancy and school age (indepen-dent variables) are presented in Table 5. BHR was associated
TABLE 5.—Multivariate logistic regression analysis (95% CI) between BHR andcurrent asthma (dependent variables) and various characteristics of study group
(independent variables).
Current
Variables
BHR OR
(95% CI)
asthma
OR (95% CI)
Atopy family history 2.52 (0.34–21.47) 3.05 (0.57–15.89)Pa ssive smoking 0.21 (0.02–2.06) 2.86 (0.50–16.12)Urticaria/anaphylaxis in
infancy0.65 (0.19–2.25) 0.40 (0.07–2.37)
Atopic dermatitis in infancy 1.57 (0.50–4.88) 2.10 (0.43–10.6)
Gastrointestinalmanifestations in infancy 2.2 (0.50–0.56) 1.2 (0.28–5.7)
Wheezing in infancy 12 (1.40–103.63)∗ 15.18 (2.62–94.23)#
Current eczema 2.23 (0.11–44.37) 1.47 (0.11–19.35)Current rhinitis 0.45 (0.04–4.33) 2.09 (0.44–9.83)Lifetime asthma symptoms 0.41 (0.01–9.92) 23.27 (2.35–241.62)#
Current asthma symptoms 51.42 (2.34–1105.31)∗ NASensitization to perennial in
infancy13.3 (2.9–59.8)# 12.5 (2.6–60.2)#
with wheezing andsensitization to perennial allergens (mites,molds) in infancy as well as in school age; it was also associ-ated with early sensitization to inhaled allergens (OR: 13.24,95%CI: 1.76-85.20, p < 0.01). There was no association toanaphylaxis at presentation, persistence of sensitization toegg/fish or development of tolerance.
DISCUSSION
The present case control study of infants allergic to eggand/or fish followed to school age demonstrates that they areat increased risk of wheezing illness and BHR at school age.Diagnosis of childhood asthma and airway hyperreactivity isassociated with early sensitization to inhaled allergens and
late sensitization to mites and molds; conversely, they are notassociated with atopic dermatitis of infancy or retention of food allergy into school age.
Our results confirm the transient character of egg allergyand are in agreement with the findings of others that sen-sitization to hen’s egg in infancy is a marker of increasedrisk of sensitization to inhaled allergens and asthma symp-toms later in life (6, 20). In the Isle of Wight birth cohortstudy, allergy to egg, whether during infancy or cumulative,correlated with respiratory allergic symptoms (asthma and/or
rhinitis) and sensitization to aero-allergens at 4 years of age(6). In the GermanMulticenter Allergy Study (MAS) childrenwith a long-lasting sensitization to food allergens developedallergic rhinitis or allergic asthma more often than childrenonly transiently or never sensitized to food (20). In contrastto MAS we found that asthma symptoms as well as BHRwere not associated with retention of egg or fish allergy. Thisdifference could be due to methodological issues; the MASstudy investigated sensitized children regardless of clinicalexpression and did not differentiate in its analysis between
the various food allergens; thus its findings do not take intoaccount differences in the natural history of allergic diseasecaused by different food allergens. However, these reports(6, 20) did not include follow-up into school age. It is of interest that when the MAS cohort was reviewed at 7 yearsof age, children with early atopic dermatitis and concomi-tant wheeze carried an almost threefold higher risk of havingcurrent wheeze; this effect was independent of concurrentsensitization to egg or milk (21).
According to our findings the prevalence of current asthma
symptoms as well as BHR was higher in the index case groupas compared to the control group; the difference in BHRdid not reach significance when compared to controls sensi-tized to inhaled allergens, but there was an apparent trendtowards higher prevalence in study group. However, con-trol group as a whole is a general population sample with
D o w nl
o ad
ed
B y : [
U ni
v
e r si
t y
of
T o r o n t o]
A t : 21
:20 1
5
O c t ob
e r 2008
594 K. N. PRIFTIS ET AL.
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 42/83
exposures to environmental factors are more likely to havea greater impact on the immature immune system and air-ways and the subsequent development of disease (22). Ini-tial assault may be sensitization to food allergens, clinicalmanifestation of atopic dermatitis, wheezing illness, or evensensitization to inhaled allergens.
The earliest serologic marker for atopic immunoreactiv-ity in infancy is the presence of IgE antibodies to egg, fol-lowed by milk (8), with an estimated point prevalence of allergy to egg of 1.6% in children 2.5 years of age (23). In aBritish population-based birth cohort study (ALSPAC) 0.4%of children skin prick tested at approximately 7.5 years werepositive to egg (24). A strong association between sensitiza-tion to egg and sensitization to aeroallergens, such as grass
or tree pollen and mites, was found in the ALSPAC study;unfortunately, no information on the clinical presentation of sensitized children was reported (24).
We found that development of BHR in symptomatic infantsallergic to egg and/or fish in infancy is not associated withatopic dermatitis either in infancy or in school age. Most stud-ies use the outcome of atopic dermatitis as the “entry point”of subsequent allergic disease that is often accompanied byfood allergy (3, 11, 21, 25). However, there are data that sup-port the hypothesis of two or more variants of eczema that
carry different prognoses and disease associations (26–29).Our finding of an association between BHR and sensitizationto inhaled allergens but not to atopic dermatitis supports thishypothesis. It seems that a genetic background that facili-tates a specific immune response constitutes the main drivethat determines the manifestations of the allergic march (11);this genetic background may not be present in all cases of the“atopic eczema/dermatitis syndrome.”
We found different patterns in the natural history of egg ascompared to fish allergy; however, sensitization to hen’s egg
or fishat schoolage didnot constituteindependentrisk factorsof BHR or current asthma symptoms in school age (Table 5).Specific sensitization to food depends on the specific allergencharacteristics, degree of utilize,and host characteristics (30).Conversely, while the development of BHR is also related toexposure, adjuvant factors, and host characteristics (31, 32),ourresults suggest that it most likely is unrelated to the naturalhistory of food allergy.
Lung function measurements revealed lower air flows atmid-FVC of controls sensitized to inhaled allergens as com-
pared to the study group and non-sensitized controls; thiscould be the result of inhaledmaintenance treatment of symp-tomatic children. Sensitization to mites was associated withlower FEF50 values. A negative association between FEV1 orFEF50 and sensitization to house dust mite has been reportedpreviously (33).
(14, 15). A negative history of food allergy by 3 years of age in control groups may be subject to recall bias. To mini-mize this possibility parents were interviewed personally byone of the study’s physicians (A.P.); despite this limitation,the low prevalence of food allergy in the general population(34) makes it highly likely that such bias only minimally af-fected our results. Lastly, we recognize that a small numberof children exclusively allergic to fish represent a limitationof the study but we were not able to obtain more cases mono-sensitized to fish up to 3 years of age.
In conclusion, the findings of our study show that childrenwho had egg or fish allergy in infancy have an increased risk for asthma symptoms and hyperreactive airways in schoolage as compared to their non-allergic peers. Late wheezing
or BHR is mostly determined by wheezing and sensitizationto inhaled allergens in early life and is not associated with themanifestation of atopic dermatitis in infancy or persistenceof egg or fish allergy until school age.
REFERENCES
1. Eder W, Ege MJ, von Mutius E. The asthma epidemic. N Engl J Med 2006;
355:2226–2235.
2. AsherMI, MontefortS, BjorkstenB, LaiC, StrachanD, Weiland S, Williams
H; ISAAC Phases Three Study Group. Worldwide time trends in the preva-
lence of symptoms of asthma, allergic rhinoconjunctivitis, and eczema
in childhood: ISAAC Phases One and Three repeat multicountry cross-
sectional surveys. Lancet 2006; 368:733–743.
3. Gustafsson D, Sjoberg O, Foucard T. Development of allergies and asthma
in infantsand young childrenwith atopic dermatitis – a prospective followup
to 7 years of age. Allergy 2000; 55:240–245.
4. HalkenS. Prevention of allergicdisease in childhood: clinicaland epidemio-
logical aspects of primary and secondaryallergyprevention. Pediatr Allergy
Immunol 2004; (15 suppl )16:9–32.
5. Nickel R, Kulig M, Forster J, Bergmann R, Bauer C, Lau S, Guggenmoos-
Holzmann I, WahnU. Sensitisation to hen’segg at theage of twelve monthsis predictive for allergic sensitization to common indoor and outdoor aller-
gens at the age of three years. J Allergy Clin Immunol 1997; 99:613–617.
6. TariqSM, Matthews SM,Hakim EA,Arshad SH.Egg allergy in infancypre-
dicts respiratory allergic disease by 4 yearsof age. Pediatr AllergyImmunol
2000; 11:162–167.
7. Kawamoto N, Kaneko H, Takemura M, Seishima M, Sakurai S, Fukao
T, Kasahara K, Iwasa S, Kondo N. Age-related changes in intracellular
cytokine profiles and Th2 dominance in allergic children. Pediatr Allergy
Immunol 2006; 17:125–133.
8. Kulig M, Bergmann R, Klettke U, Wahn V, Tacke U, Wahn U. Natural
course of sensitization to food and inhalant allergens during the first 6 yearsof life. J Allergy Clin Immunol 1999; 103:1173–1179.
9. Togias A. Systemic effects of local allergic disease. J Allergy Clin Immunol
2004; 113(1 suppl):S8–S14.
10. Peters S. The impact of comorbid atopic disease on asthma: clinical expres-
sion and treatment. J Asthma 2007; 44:149–161.
11. Wahn U. What drives the allergic march? Allergy 2000; 55:591–599.
D o w nl
o ad
ed
B y : [
U ni
v
e r si
t y
of
T o r o n t o]
A t : 21
:20 1
5
O c t ob
e r 2008
FOOD ALLERGY, ASTHMA AND BRONCHIAL REACTIVITY 595
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 43/83
paper from the European Academy of Allergology and Clinical Immunol-
ogy. Allergy 2004; 59:690–697.
16. Standardization of spirometry: 1987 update. Statement of the American
Thoracic Society. Am Rev Respir Dis 1987; 136:1285–1298.
17. Quanjer PhH. Standardized lung function testing. Bull Eur Physiopathol
Respir 1983; 19(suppl 5):45–51.18. Crapo RO, Casaburi R, Coates AL, Enright PL, Hankinson JL, Irvin CG,
MacIntyre NR, McKay RT, Wanger JS, Anderson SD, Cockcroft DW, Fish
JE, Sterk PJ. Guidelines for methacholine and exercise challenge testing-
1999. Thisofficialstatementof the American Thoracic Society was adopted
by the ATS Boardof Directors, July 1999. Am J RespirCritCareMed 2000;
161:309–329.
19. Dreborg S, Backman A, Basomba A, BousquetJ, DiegesP,Malling H. Skin
tests used in type I allergy testing. Position paper, prepared by the subcom-
mittee on skin prick tests within the European Academy of Allergology and
Clinical Immunology. Allergy 1984; 44(suppl 10):1–59.
20. Kulig M, BergmannR, Tacke U, Wahn U, Guggenmoos-HolzmannI. Long-lasting sensitization to food during the first two years precedes allergic
airway disease. The MASStudy Group, Germany. Pediatr AllergyImmunol
1998; 9:61–67.
21. IlliS, von MutiusE, Lau S,Nickel R,Gruber C, NiggemannB, WahnU, the
Multicenter Allergy Study Group. The natural course of atopic dermatitis
from birth to age 7 years and the association with asthma. J Allergy Clin
Immunol 2004; 113:925–931.
22. Huang SW. Follow-up of children with rhinitis and cough associated with
milk allergy. Pediatr Allergy Immunol 2007; 18:81–85.
23. Eggesbø M,Botten G,Halvorsen R.,Magnus P. Theprevalenceof allergy to
egg:a population-basedstudy in young children. Allergy2001:56:403–411.
24. Roberts G, Peckittw C, Northstonez K, Strachan D, Lack G, Henderson J,
Golding J and the ALSPAC Study Team. Relationship between aeroaller-
gen and food allergen sensitization in childhood. Clin Exp Allergy 2005;
35:933–940.
25. Ricci G, Patrizi A, Baldi E, Menna G, Tabanelli M, Masi M. Long-term
follow-up of atopic dermatitis: Retrospective analysis of related risk factors
and association with concomitant allergic diseases. J Am Acad Dermatol
2006; 55:765–771.26. Flohr C, Johansson SGO, Wahlgren CF, Williams HC. How “atopic” is
atopic dermatitis? J Allergy Clin Immunol 2004; 114:150–158.
27. Kusel MM, Holt P, de Klerk N, Sly P. Support for 2 variants of eczema. J
Allergy Clin Immunol 2005; 116:1067–1072.
28. Lowe AJ, Hosking CS, Bennett CM, Carling JB, Abramson MJ, Hill DJ,
Dharmage SC. Skin prick test can identify eczematous infants at risk of
asthma and allergic rhinitis. Clin Exp Allergy 2007; 37:1624–1631.
29. van der Hulst AE, Klip H, Brand PL. Risk of developing asthma in young
children with atopic eczema: a systematic review. J Allergy Clin Immunol
2007; 120:565–569.
30. Chehade M, Mayer L. Oral tolerance and its relation to food hypersensitiv-ities.J Allergy Clin Immunol 2005; 115:3–112.
31. Sears MR, Burrows B, Herbison GP, Holdaway MD, Flannery EM. Atopy
in childhood. II. Relationship to airway responsiveness, hay fever, and
asthma.Clin Exp Allergy 1993; 23:949–956.
32. Bergren DR. Tobacco smoke is an adjuvant for maintained airway sensiti-
zation in guinea pigs. J Asthma 2007; 44:723–728.
33. Anthracopoulos M, Mantzouranis E, Paliatsos A, Tzavelas G, Lagona
L, Nicolaidou P, Priftis K. Different effects of sensitization to mites
and pollens on wheezing and spirometry in children: a population-
based cohort study. Ann Allergy Asthma Immunol 2007; 99:122–
129.
34. Høst A. Adverse reactions to foods: epidemiology and risk factors. Pediatr
Allergy Immunol 1995; 6(suppl 8):20–28.
D o w nl
o ad
ed
B y : [
U ni v
e r si
t y
of
T o r o n t o]
A t : 21
:20 1
5
O c t ob
e r 2008
Acta Pædiatrica IS SN 0803–5253
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 44/83
R E G U L A R A R T I C L E
Neonatal characteristics and risk of atopic asthma in schoolchildren:
results from a large prospective birth-cohort studyUrsula Kiechl-Kohlendorfer ([email protected])1, Elisabeth Horak 2, Wilfried Mueller3, Robert Strobl4, Claudia Haberland5,Franz-Martin Fink 6, Michaela Schwaiger6, Karl-Heinz Gutenberger7, Hannes Reich7, Dagmar Meraner2, Stefan Kiechl8
1.Department of Paediatrics, Division of Neonatology, Neuropaediatrics and Metabolic Diseases, Innsbruck Medical University, Innsbruck, Austria2.Department of Paediatrics, Division of Paediatric Cardiology and Pulmonology, Innsbruck Medical University, Innsbruck, Austria3.Department of Paediatrics, Reutte County Hospital, Reutte, Austria4.Department of Paediatrics, Lienz County Hospital, Lienz, Austria5.Department of Paediatrics, Kufstein County Hospital, Kufstein, Austria6.Department of Paediatrics, St. Johann County Hospital, St. Johann, Austria7.Department of Paediatrics, St. Vincent’s Hospital , Zams, Austria
8.Department of Neurology, Innsbruck Medical University, Innsbruck, Austria
Keywords
Atopic asthma, Children, Farming, Hospitalization,Risk factors
Correspondence
Ursula Kiechl-Kohlendorfer, Department of Paediatrics, Division of Neonatology,Neuropaediatrics and Metabolic Diseases,Innsbruck Medical University, Anichstrasse 35,6020 Innsbruck, Austria.
Tel.: +43 512 504 23600 |Fax: +43 512 504 23484 |
Email: [email protected]
Received
18 May 2007; revised 3 July 2007; accepted 6 July2007.
DOI:10.1111/j.1651-2227.2007.00449.x
Abstract
Aim: Asthma is among the most common chronic diseases in childhood and steadily increasing in
prevalence. Identification of risk predictors for a hospitalization for atopic asthma in childhood may
help design prevention programmes and improve our understanding of disease pathobiology.
Methods: An ongoing birth-cohort study prospectively enrolled all liveborn infants in Tyrol. Between
1994 and 1999 baseline data were collected for 33 808 infants. From 2000 to 2005, all children
hospitalized for atopic asthma at an age of 6 years or over (n = 305) were identified in a careful
search of hospital databases. Disease status was ascertained from the typical medical history, athorough examination and proof of atopy.
Results: Male sex (relative risk 2.11, 95% CI 1.65–2.70), urban living environment (vs. rural) (1.93,
1.47–2.54), neonatal admission to hospital (1.70, 1.20–2.40), lack of breastfeeding (1.32,
1.02–1.70), postnatal smoking (1.31, 1.00–1.72) and low birth weight (1.45, 0.94–2.23) all
emerged as adverse risk predictors for hospitalization for atopic asthma whereas a low risk was found
among children living on a farm (0.22, 0.05–0.87) and children with two to three siblings (vs. no or
one sibling) (0.71, 0.51–0.97).
Conclusion: In this study a number of neonatal characteristics and environmental exposures were associated
with hospitalization for atopic asthma in childhood, suggesting that early life is crucial for disease determination
and lending further indirect support to the hygiene hypothesis.
INTRODUCTION
Asthma is among the most common chronic diseases in
childhood and imposes an increasing economic and health
burden (1). Previous studies have suggested that factors
acting early in life and even in the prenatal period may
predispose to or render individuals resistant to childhood
asthma (2,3). In addition, birth characteristics and sociode-
mographic variables like birth weight (4,5), sex (6,7), family
size (8) and parental educational level (9) have been linked
to the risk of asthma manifestation and hospitalization, but
the available results are not entirely consistent In the last
studies may deepen our understanding of disease pathobi-
ology and provide a basis for prevention strategies aimed at
counteracting the steadily increasing burden of asthma (14).
The current study will investigate the association of neonatal
and environmental characteristics with hospitalization for
atopic childhood asthma in a large prospective birth-cohortstudy (n = 33808).
PATIENTS AND METHODS
Study area
The survey area of this study is Tyrol a federal state in the
Kiechl-Kohlendorfer et al. Neonatal characteristics and risk of atopic asthma
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 45/83
and another 2.1% moving within the Tyrol still contributing
to the group of hospitalized asthmatics.
Study population and clinical characteristics
Data on pregnancy, birth and neonatal characteristics wereprospectively collected 4–6 weeks after birth for all infants
liveborn between April 1994 and December 1999 (n =
45 642). The participation rate was high at 74.1% (n =
33808).
The following variables were assessed by means of a stan-
dardized questionnaire completed by the infant’s mother:
(a) infant’s data: age, sex, birth weight and gestation (<37
and ≥ 37 weeks); (b) sociodemographic background: mar-
ital status, educational level and profession of the parents;
(c) pregnancy characteristics: mother’s age at delivery, num-
ber of previous pregnancies and deliveries, maternal smok-
ing habits during pregnancy and after birth of the child;
(d) postnatal factors: neonatal hospitalization, child care
practices such as the usual sleeping position and feeding
practices; (e) place of residence. The questionnaire used in
the current study is part of the public health programme in
Tyrol and was approved by the appropriate committee of
the local Board of Health. Categorization of all variables
was prespecified. Low birth weight was defined by a cut-off of <2500 g. Season of birth was categorized as ‘cold
months’ (November–February) or ‘other months’. Feeding
practice during the first 4–6 weeks of life was coded ‘breast-
fed’ if the main type of milk given was breast milk. Parental
educational level was documented in years of education
(<12 years, 12 years,and>12 years). A farming environment
was assumed if farm work was given as a full-time occu-
pation. Living environment was coded ‘urban’ (large towns
with>
100 000 inhabitants), ‘suburban’ (towns with 10 000–100 000 inhabitants), or ‘rural’ (villages with <100 00 inhab-
itants).
Ascertainment of asthma cases
Data on hospitalization for atopic asthma of children aged
6–10 years (born April 1994 to December 1999) were ex-
tracted from computer databases of all children’s hospitals
in Tyrol. Children aged 5 years and under were not included
in the current investigation, because there are many differ-
ent causes for wheezing in this age range. In a first step,
a search was made for diagnoses 493.0 (ICD-9) and J45.0
(ICD-10) which identified 721 infants. In a second step the
diagnosis was confirmed by two paediatric pulmonologists
(E.H., D.M.) based upon a critical review of the hospital
medical records A diagnosis of asthma required a medi-
The level of specific IgE was measured by radioallergo sor-
bent assay (RAST) and considered elevated if the level was
≥ 3.5 kU per litre for the specific allergen. A total of 367
(50.9%) out of 721 children satisfied these criteria.
Statistical analysis
Data analysis was performed using SPSS 12.0 for Windows.
Associations between categorized baseline variables and
hospitalization for atopic childhood asthma were analyzed
with the 2 test. Relative risks were estimated from logis-
tic regression analysis applying hospitalization for asthma as
the dependent variable. Models were fitted with a forward-
stepwise selection procedure (probability values for entry
and removal of variables, 0.10 and 0.15) and variables se-lected for inclusion from all those listed in Table 1. The
test procedure was based on maximum-likelihood estima-
tors, and the goodness of fit of each model was assessed
by the test of Hosmer and Lemeshow. Differential associa-
tions in subgroups were tested by the inclusion of interaction
terms. All p-values were two-sided.
RESULTS
A critical review of hospital admission records identified 367children with atopic asthma in the survey area, who required
at least one hospitalization in the period between 2000 and
2005. Neonatal questionnaires had been returned in 305 of
the cases.
Descriptive characteristics of the overall birth cohort
(33 808) are shown in Table 1.
Univariate analysis showed hospital admission for asthma
to be significantly associated with male sex, low birth weight,
urban living environment, parental educational level, neona-tal admission to hospital, lack of breastfeeding and smoking.
Detailed analyses indicated that the increase in asthma
risk equally applied to women who smoked during and after
pregnancy (n = 4645, RR [95% CI] 1.36 [1.01–1.84]) and
those who abstained during pregnancy but restarted smok-
ing after delivery (n = 1445, RR [95% CI] 1.57 [0.98–2.52])
whereas the number of women who smoked during preg-
nancy and quitted thereafter was too low to assess reliable
risk estimates (n = 259).
In addition, the risk for hospitalization for atopic asthma
was lower in children who live on a farm and those having 2–
3 siblings, and in children of parents with a low education
level. In detail, relative risks of asthma amounted to 1.00,
0.71 and 1.25 for children with 1, 2–3 and 4 siblings when
compared with those without a sibling Finally there were
Neonatal characteristics and risk of atopic asthma Kiechl-Kohlendorfer et al.
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 46/83
Table 1 Association of neonatal, sociodemographic and environmental charac-
teristics with hospitalization for atopic childhood asthma (n = 38 808)
Hospitalizations Frequency of
Subjects for asthma hos pita lizations
Variable at risk per 1000 PYs∗ p-value‡
Sex
Male 17 219 210 6.3 <0.001
Female 16 589 95 3.0
Season of birth
Cold months 10 126 92 4.8 0.935
Other months 23 682 213 4.6
Birth weight
<2500 g 1666 26 8.4 0.004
≥2500 g 32 142 279 4.5
Mother’s age at delivery
<23 years 3886 33 4.2 0.711
≥23 years 19 922 272 4.7
Number of siblings
≤1 sibling 25 655 248 5.0 0.028
2 or 3 siblings 7171 46 3.3
≥4 siblings 982 11 5.7
Single parenthood
No 30 381 278 4.8 0.455
Yes 3427 27 3.6
Educational level†
<12 years 22 032 163 4.3 0.042
12 years 4012 45 6.8
≥12 years 4441 38 5.1
Farming
No 32 658 303 4.8 0.004
Yes 1150 2 1.0
Living environment
Urban 4186 70 8.1 <0.001
Suburban 4010 24 3.3
Rural 25 612 211 4.2
Neonatal admission to hospital
No 31 269 263 4.3 <0.001
Yes 2539 42 8.7
Initial breast feeding
No 4580 54 6.0 0.033Yes 29 228 251 4.4
Prone sleeping position
No 32 071 289 4.7 0.932
Yes 1737 16 4.1
Postnatal smoking
Table 2 Multivariate association between risk variables and hospitalization for
atopic childhood asthma (n = 33 808)
Variable RR (95% CI) p value
Male sex 2.11 (1.65–2.70) <
0.001Living environment <0.001
Rural 1.00
Suburban 0.70 (0.46–1.07) 0.097
Urban 1.93 (1.47–2.54) <0.001
Neonatal admission to hospi tal 1.70 (1.20–2.40) 0.003
Farming 0.22 (0.05–0.87) 0.032
Lack of breastfeeding 1.32 (1.02–1.70) 0.034
Postnatal smoking 1.31 (1.00–1.72) 0.048
Number of siblings 0.058
≤1 sibling 1.00
2 or 3 siblings 0.71 (0.51–0.97) 0.033
≥4 siblings 1.26 (0.68–2.31) 0.462
Low birth weight 1.45 (0.94–2.23) 0.091
Relative risks (RR) and 95% confidence intervals (95% CI) were es-
timated from logistic regression analysis. The multivariate analysis was
fitted with a forward-stepwise selection procedure allowing for all variablesin Table 1 (probability values for entry and removal of variables, 0.10 and 0.15).
level lost significance after inclusion of smoking and lack of
breastfeeding. The risk profiles identified were virtually iden-
tical in boys and girls and subgroups according to different
level of parental education.
DISCUSSION
This prospective population-based study identified a numberof significant risk predictors for hospitalizations because of
atopic childhood asthma (Tables 1 and 2). Some of the as-
sociations have been described in previous birth cohorts fo-
cusing on asthma manifestation or hospitalization (9,11,15–
18), whereas others like a farming environment have not
yet been addressed in prospective studies. Notably, the bulk
of risk predictors were neonatal characteristics indicating
that early life is crucial for the determination of childhood
asthma. The positive association with postnatal smoking and
urban living environment indicates a detrimental effect of
in- and outdoor air pollution on asthma manifestation and
course, while the strong inverse association with the number
of siblings (surrogate for microbiological burden) and living
on a farm and in a rural environment may be considered
further support for the hygiene hypothesis (8 10)
Kiechl-Kohlendorfer et al. Neonatal characteristics and risk of atopic asthma
d th i k f i f ti i d d i b t ti (19) f ilit t th d l t f i t ill (25) d
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 47/83
reduces the risk of infection-induced airway obstruction (19)
andhas profound effects on the infant’s intestinalmicroflora,
and anti-inflammatory as well as immunomodulatory prop-
erties (20). The higher risk for asthma among low birth
weight infants is in accordance with previous studies (4,21)and probably results from irreversible constraints in airway
development (21) and subsequent small airway size (22).
The majority (80%) of the neonatal hospital admissions in
our cohort were due to respiratory distress, and this was es-
pecially true for preterm infants. Therefore, the increased
risk for severe asthma with subsequent hospitalization in
this group might be explained by chronic lung disease, as
commonly observed in premature neonates, and/or by lung
injury after mechanical ventilation. There is evidence that in-
fants who survive a respiratory distress syndrome and those
with a lower than normal lung function are at greater risk for
developing respiratory disease in later life (23). Vice versa, a
family history of asthma (genetic predisposition) was shown
to increase the risk for severe neonatal respiratory distress
leading to hospital admission (24).
As anticipated, environmental exposure early in life may
interfere in the maturing immune system and play a crucial
role in the development and course of atopy and asthma.
An effective innate immune system (involving its signallingpathways), induction of a profound shift in the TH1/TH2
balance towards TH1 preponderance and more complex
changes in immune responsiveness like the development of
immune tolerance and subtle modifications in TH2 response
are considered protective against atopy and asthma (11–13).
The type and amount of allergens, window of allergen con-
tact and associated circumstances (e.g. level of airway in-
flammation) may determine whether exposure causes sensi-
tization or immune tolerance. Early exposure to infectionstransmitted by contact with older siblings has repeatedly
been related to a decreased risk for atopic disease (8,11). This
theory, first postulated by Strachan in the 1980s, was termed
the hygiene hypothesis (10). Lifestyle changes in western
society, especially the decreasing number of children in a
family, may well have contributed to the increase in allergic
diseases. The current study provides further evidence that a
larger number of siblings may afford protection against the
development of severe asthma leading to hospitalization, but
this finding does not extend to families with four and morechildren. Observation of a protective effect of farming in our
study is in accordance with a few cross-sectional evaluations
(12,13).
Early-life contact with microbial compounds that are
highly prevalent on farms may account for this phenomenon
se facilitates the development of respiratory illness (25) and
promotes sensitization to environmental allergens and may
trigger severe asthma attacks. Analogue effects have to be as-
sumed for indoor exposure to tobacco smoke. Actually, the
results of the current study support the notion that parentalsmoking is an important determinant of atopic asthma in
children (16,26).
Apart from inducing airway inflammation and priming
sensitization (27), smoking impairs lung development and
function in utero and in the postnatal period (28).
Merits and limitations The main merits of our study are
its prospective design and size (n = 33 808). Size clearly
exceeds that of comparable birth-cohort studies focusing
on neonatal characteristics, most of which included a few
thousand infants (9,11,15–18). Furthermore, ascertainment
of asthma did not rely on the parents’ responses to the
questionnaire or the patient’s self-report but on a thorough
in-hospital examination by a specialist in paediatric pul-
monology. In each case the diagnosis was confirmed by
proof of atopy (positive skin-prick test reaction to common
aeroallergens and/or increased specific IgE levels). There
are limitations as well. Firstly, this study was primarily de-
signed to assess risk factors for sudden infant death syn-
drome (SIDS) and apparent life-threatening event (29,30)and later extended to childhood asthma. Accordingly, some
variables meaningful for the current analysis like family his-
tory of asthma and pet contact were not systematically as-
sessed. Secondly, the focus on asthma hospitalization does
ultimately not allow to differentiate whether risk predictors
are related to asthma manifestation or exacerbation. In this
context it should also be mentioned that in Tyrol health
care is free of charge and easily accessible throughout the
survey area and does not differ between sociodemographicgroups, making confounding by selective hospital admission
unlikely. Finally, a minor proportion of families moved out
of the survey area during the study period (1.2% per year).
Because these families do no longer contribute to the pop-
ulation of hospitalized asthmatics relative risks in Table 2
may be modestly underestimated.
In conclusion, our study identifies a broad array of neona-
tal and environmental risk predictors for hospitalization for
atopic asthma in childhood, suggesting that disease man-
ifestation and expression is in part determined early inlife, and provides further indirect support to the hygiene
hypothesis.
ACKNOWLEDGEMENTS
The questionnaire used in the current study is part of
Neonatal characteristics and risk of atopic asthma Kiechl-Kohlendorfer et al.
3 Strachan DP Is allergic disease programmed in early life? Clin for asthma from childhood to young adulthood: a
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 48/83
3. Strachan DP. Is allergic disease programmed in early life? Clin
Exp Allergy 1994; 24: 603–5.4. Svanes C, Omenaas E, Heuch JM, Irgens LM, Gulsvik A. Birth
characteristics and asthma symptoms in young adults: resultsfrom a population-based cohort study in Norway. Eur Respir J
1998; 12: 1366–70.5. Shaheen SO, Sterne JAC, Montgomery SM, Azima H. Birth
weight, body mass index and asthma in young adults. Thorax
1999; 54: 396–402.6. Osman M. Therapeutic implications of sex differences in
asthma and atopy. Arch Dis Child 2003; 88: 587–90.7. Wever-Hess J, Kouwenberg JM, Duiverman EJ, Hermans J,
Wever AMJ. Risk factors for exacerbations and hospitaladmissions in asthma of early childhood. Pediatr Pulmonol
2000; 29: 250–6.8. Strachan DP. Family size, infection and atopy: the first decade
of the “hygiene hypothesis.” Thorax 2000; 55: 2–10.9. Hancox RJ, Milne BJ, Taylor DR, Greene JM, Cowan JO,
Flannery EM, et al. Relationship between socioeconomicstatus and asthma: a longitudinal cohort study. Thorax 2004;59: 376–80.
10. Strachan DP. Hay fever, hygiene and household size. Br Med J
1989; 299: 1259–60.11. Ball TM, Castro-Rodriguez JA, Griffith KA, Holberg CJ,
Martinez FD, Wright AL. Siblings, day-care attendance, andthe risk of asthma and wheezing during childhood. N Engl J
Med 2000; 343: 538–43.12. Braun-Fahrlander C, Riedler J, Herz U, Eder W, Waser M,Grize L, et al. Environmental exposure to endotoxin and itsrelation to asthma in school-age children. N Engl J Med 2002;347: 869–77.
13. Riedler J, Braun-Fahrl ander C, Eder W, Schrever M, Waser M,Maisch S, et al. Exposure to farming in early life anddevelopment of asthma and allergy: a cross-sectional survey.Lancet 2001; 358: 1129–33.
14. Asher MI, Montefort S, Bjorksten B, Lai CK, Strachan DP,Weiland SK, et al. Worldwide time trends in the prevalence of
symptoms of asthma, allergic rhinoconjunctivitis, and eczemain childhood: ISAAC Phases One and Three repeatmulticountry cross-sectional surveys. Lancet 2006; 368:733–44.
15. Rothenbacher D, Weyermann M, Beermann C, Brenner H.Breastfeeding, soluble CD 14 concentration in breast milk andrisk of atopic dermatitis and asthma in early childhood: birthcohort study. Clin Exp Allergy 2005; 35: 1014–21.
16. Kull I, Almqvist C, Lilja G, Pershagen G, Wickman M.Breast-feeding reduces the risk of asthma during the first 4years of life. J Allergy Clin Immunol 2004; 114: 755–60.
17. Rasmussen F, Taylor DR, Flannery EM, Cowan JO, Greene JM, Herbison GP, et al. Risk factors for hospital admission
for asthma from childhood to young adulthood: alogitudinal population study. J Allergy Clin Immunol 2002;110: 220–7.
18. Hagendorens MM, Bridts CH, Lauwers K, van Nuijs S, EboDG, Vellinga A, et al. Perinatal risk factors for sensitization,
atopic dermatitis and wheezing during the first year of life(PIPO study). Clin Exp Allergy 2005; 35: 733–40.
19. Friedman NJ, Zeiger RS. The role of breast-feeding in thedevelopment of allergies and asthma. J Allergy Clin Immunol
2005; 115: 1238–48.20. Hoppu U, Kalliomaki M, Laiho K, Isolauri E. Breast
milk-immunomodulatory signals against allergic diseases. Allergy 2001; 56: 23–6.
21. Barker DJ, Godfrey KM, Fall C, Osmond C, Winter PD,Shaheen SO. Relation of birth weight and childhoodrespiratory infection to adult lung function and death fromchronic obstructive airway disease. Br Med J 1991; 303:671–5.
22. Inselman LS, Mellins RB. Growth and development of thelung. J Pediatr 1981; 98: 1–15.
23. Haland G, Lodrup Carlsen KC, Sandvik L, Devulapalli CS,Munthe-Kaas MC, Morten Pettersen MD, et al. Reduced lungfunction at birth and the risk of asthma at 10 years of age. N
Engl J Med 2006; 355: 1682–9.24. von Mutius E, Nicolai T, Martinez FD. Prematurity as a risk
factor for asthma in preadolescent children. J Pediatr 1993;
123: 223–9.25. D’Amato G, Liccardi G, D’Amato M. Environmental riskfactors (outdoor air pollution and climatic changes) andincreased trend of respiratory allergy. J Investig Allergol Clin
Immunol 2000; 10: 123–8.26. Lau S, Nickel R, Niggemann B, Gruber C, Sommerfeld C, Illi
S, et al. The development of childhood asthma: lessons fromthe German Multicentre Allergy Study (MAS). Paediatr Respir
Rev 2002; 3: 265–72.27. Ronchetti R, Macri F, Ciofetta G, Martinez FD. Increased
serum IgE and increased prevalence of eosinophilia in 9 year
old children of smoking parents. J Allergy Clin Immunol 1990;86: 400–7.
28. Tager IB, Weiss ST, Munoz A, Rosner B, Speizer FE.Longitudinal study of the effects of maternal smoking onpulmonary function in children. N Engl J Med 1983; 309:699–703.
29. Kiechl-Kohlendorfer U, Pupp Peglow U, Kiechl S, OberaignerW, Sperl W. Epidemiology of sudden infant death syndrome(SIDS) in the Tyrol before and after an intervention campaign.Wien Klin Wochenschr 2001; 113: 27–32.
30. Kiechl-Kohlendorfer U, Hof D, Pupp Peglow U, Traweger-
Ravanelli B, Kiechl S. Epidemiology of apparent lifethreatening events. Arch Dis Child 2005; 90: 297–300.
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 49/83
From:
Epidemiology: An Introduction
Kenneth J. Rothman
Oxford University Press
2002
ISBN 0-19-513554-7
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 50/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 51/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 52/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 53/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 54/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 55/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 56/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 57/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 58/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 59/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 60/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 61/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 62/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 63/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 64/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 65/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 66/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 67/83
From:
Epidemiology: An Introduction
Kenneth J. Rothman
Oxford University Press, 2002
ISBN 0-19-513554-7
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 68/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 69/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 70/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 71/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 72/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 73/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 74/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 75/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 76/83
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 77/83
ir =
s;s
v+s
=
5
Z
-O
=c-
F H
6r
:H
t
e E
s
,3tS
E
:
,a
S
3iS
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 78/83
I
.-ia
ra
v
a
-
OJ
-
'-o-qO
lrO
'b
o
.aE
E
-5
+
+
+s
=(
.
g
l
c
-aP
o
f5
0
Iq'dF
\.hEG
€
a
O
E
-cJ
5
c
Se-nC
d
>
e+
EE
:
:a
=
"
;
=E
-
"*e
eT
;
tr6
"E:X
-e
Y
q
€
jz
s
E
g
E
F
y2
55
=JEy
-
4
t eE
F
8
$EEy
.gH
g
-*
Et€E
P
EscbF
:E
,Eo
E=
A
sFE?E;F
eE
€E+
E
Y
Ea
Z
;
;gH
E
-
*
E;z -P*(H
q
oo(g
.=i
I
rb=cZ
(aYF
" -
l g
.
=
' \
c
+<
^
F
(
l;Flp-\EX
-v"99RA{)e.
aq
xv=
T
*a
I
-
JFu
vor
^
ca.9
R
aE6F
4
r
;z\
t=
v-
/<4=
r
v
X\JO4U
q(aqO*6-(U+
+Ft
t=:3YaCafoO
-K=^5aaP9=f=YfSUl{ac
ai
.=z9.oPa
o
.c?hYho
-e*?-Z-
.=\
n
cqz3r
{8q
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 79/83
.ar(---
t
a
lr
'(J
i
-+
q
l-
R
L
l-
-
O
:J
*?t5q.H
+
-2-
E
q
C-dp
dE
c
F
.=
(
q2
#
:
'
-
Y
I
h
s
)
q
=
.Z^
s-
1
-
t
;
-
iJ
ia
Sa
r
.=E
),g
G
't=
F
_
o
a
=
;
:1
q
a
f
-
F
2
u
:-b
E"FE
5
:a
'l aE
E
.E
'A
1
q
"a
II
E
a
tavJ=
l-E
A
'S*
,
l:
)
|
=(
'
.?s
F
,
s
:r
'i;{Ar r
E
:
?
c:
,1-y
++
y:
^
I
F=
,9
F
iz
x1
dt
3z
--
1
v
^1
i6
4
d
.3
i(
r
a
d
a
=
:1a
i
l
J
lr
:
c
tr
u
=
v
t-I+
*e
9.(a 9+Io
1-FqSIOC
C
a\E4\Gr
bE
+r
F
nenX
t*
i
af.v-O+I4"u+I
=EU
z=2Ea9.(o
cV
(
y
EcJE=2A
==E?ECt9
.?
€
E
IE E
f i
c a
*z
c=
2
+
2
jr
:=
o
(
a
f9
-
=5
+
J
6=
€E
;
*
i
{
;3
l
=
"
=E
c
i\
l=S
=+A
*+
H
la
X-
i-c
2=
,
,d
' a
-
t
tu
t
I+9
h
v=
{u
{?
sE
<
=
q
q
s?3
sisea
g
I g
a
"a>
v*
-c
'E
a=I
6
.
1
-
--d(Jo
4
g
C
(r
*>o
d
G
c
C
a3O
ocO'6
:-
t
'YO
q
b3
aF
i
4
d
L7
rc::
F
c.
IcJ
.
.
4E
-L"e
:.
c
=
n
'4rZ
dEs
.=
-
\
-J
+
L
*
cE
1
=
\<7
a
a
=Uo
rVP
.?=
L
.u
=2
-_
t+D
{
Y
-
=
.JZ
c
'
6
,
-
u
\E
d
u
G
=
-c9
iAa
i
c
c
_
L
\
Os
"
.
.P
F
=J
=
.Le
CvZ>
%
;
d
9Y-
-c
d
F-o
-
-L
u
-9
:
uEyZ
n^
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 80/83
eF
: I
{
I iE
s
igg
E
i; l
:
ifg
| |
g
sj
l Ez g
g
gE
O
V
s
rcr
1E=
3
; €E;:n=
iIB
E E
E
F
iF:
q
sr
g:rE
a
E
}Eg
f
i;e
E*?
as
r-gE
Eg
g
+F
Vc
Oot>>a
AzL
O
a@
oi>F=E(
OzL
O
a
tttIal,=a.
a
>
rXa1H\
>oE
^
.
a
F
hi>;Er
4\
z>o
=
:
@
IS
SO
OA
Soo
oo
OUacqtIt
FF
E
lo'aK"<
F
LU
c
Y3
#U
:^
:adF=zA
LU_u
:?
ir
u
=
o@*
*6C
a*>tYrEkn
I
JZ
A
L
U
.
e
v
E+
RYY
V
(
IS ^-s
v
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 81/83
==
E
>cJu
c
c
b
-1o
=,=.=
E
=
:1=n
*-FE
9=
.E
'
I
-
_
=
=
-=
'
=E
.=
.y
itzz7
't+
-E
c=EE=
fr4
i"
E
=
z
3=
4
s&=e
IGo
Q
&
ErI=
E
{_
=
c
;-E
i>a
a
,1
E
dF
-==
io-cS
-c=
E
=
a=F
zE.
==
hE:9
T
a
Y
E
E
=
'r2
-=dE
"=
==
?c-
7
.S-=>
>)
'=
3
eX
q
.
I
o
o
83- 6E
E-o
Hg
5E
-d
S
=
ifE
2z7Y
e
==
' -E
6 -E>
y 54
9
oc
9UHo
oE>-o
g
r?>
f.Je
E
?=
"q:
n
.=
-
o
:
ad
JP
A
-
1
c
h^
"
@ Q
q*
=
=
.=I3
1
Z-Ee
5E3
-d=^=w
?
B
3
4
=
u"=
'=
-=
luE
-EZ
€
o
i=E
=FE
B
.2
;=
s:*.
E
t€F=
e
(o
=
i =9=r
Jb
F-FE
"<
E
F=
-
IcPF?P
^a
aA
.€
63
A
5
Qc
;uX
rn
>1\
:
;-":u1
0O
l
gH
:
X
J
o
c4aF
c
g
A
UE
o
t,n<>Uz>
a
n
O
S--SO
<E+
znn
gs>a.=
=
E
E
;E
;e;gE
;FE
r
qa
e€5t
?
E3E:rE
: -
=*i
P
E
y=e
=E
f=5
:sg
+Z
f f g
iE
F
s€
sgEEtE
E
svEE
+?
=
teIE
E€
rc;E z
E
EE
gEE E
++
d
J
nv
^-COu oc
dt
"odaac
6
lsloo
tH?
lE
sq
lf
E
le
g
I
EE
lr
o
lF#S
IXE
lFI$FB
:B5O
E
[S
F
E
=Q
h
>
E
,
s,"X
q
:*HX
;?
t
EXd
5
g
F
'5
e&
qo
c<
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 82/83
>
>
.9Rqa6u->qEO
JoS9
EEU
C+>a
qx
F
c
c
s'Eq-,Ou9
-F
H
v=O
H
Cv
-F=-AvA*ELA
€
:4
)
R
L
r
=
l
F8
N
Uq
fn
d
9(J
c
a
9-.
.u
qo*U
9U
L
vF
q+e .
aP
IrFF
$
Nra{q(h
I
t?r=
)XI c
2 +
HAYFH
*
J
io\o\ol
H
r5+
V
O
o
X{L(
.
c
roqr
V
€o
hA
rc\+
oIIJV
L^
V-
QvE
va
a
.
c
>'=.=c
-e
--9
tR
F
'
-
t
R
9v
*P
-a
e
-
s=
2
J6
-C
a
a
S
P
=
;
;
'-
34
o
i =eS-
9
€
-Ez:
O
':
=
t=
C
sB9
-
-4=
d
-v
X
E
4
d_h:3
a>5E
<
Aa
n5Y
';1>8
>
-Y+
A
-
e
co
>9=
:Le
vTO
-cctsK
=o
iv
dX9rE
-d
i
o-o
93
EA
tF
i-=at=,f*a
i-
S
sa,O.=,ob
u
.
a<
^
E6
:oo.EadF.9a-a
Ecc
nU;U;>=QUa
2u
l
a4fN
wN
8/13/2019 GREAT Lecture1 Epid Study Design Handout
http://slidepdf.com/reader/full/great-lecture1-epid-study-design-handout 83/83
*a
7a
.=d
&
7
-a=
L
L
==Y
99
p
n
2
9
=
r
c
C
;
a
-
v
-
-
4
&
ta
-
Y
,
-
,e
-c=
a?u
=
e
9
3^
:
^
i=n
i
-=:3
-
-Q
="
i
?
t
Y
9
)
-_-=_
i=
}1
L
tQ
n^7
P
z+
23€6
F
M
=
Y
>
^
2
t=
-
ia<
E
??
;=.E
_
3
t
;Lo
1=
ipE.?c
_YL
r9
=
I2
Y
"A
.
Y
9
.=
iZi
E
2z=Y
=r1
-
;-=7
=
994^
d
c
-
O
Y
C
'.2
L
-
.2F=E
4
,4
=
-CZ
i
'a
a
=fr
a
14=t
'
rt
:=-
'=-
?o
E
=
P
XF
c
'
'
J
-
-
oL-
oQ
n
u
#+
rn
'o
L
.=-
dDF
m
f(
rc
(+
+d
Irn
s
C
E
F
;E
Edd
?
Z=,rE
€
*;F?
E:aE
E Ecr
-
5 ti :2E
(P
o
E
OA
E
= a
1F
tn*
E;F3E
EFP
FYE
-E
=c
-opc
=?
c
p
V
-y
i=
-
:;#aE
E
E
:E
:r6
;
EF
€
E
rr,
H
E
E
+C
E
n
:E
se€
e
:eE;
c E-a
; :
f*E
?
E€:z
t
fE ;g
E+9*tE
9
E=
=E
;dE
E-E
:€Er=t$E3
:=7
i=E
j
€d
;e
s
eE
=E
ed
eE
S
Egg aA
Z
uuE
Ei i
rnA;
L
= a=
x>
U-
u
-=
ugu*
P
*
lLC
gS
g