comparison of caregiver otitis media risk factor knowledge in suburban and urban primary care...

International Journal of Pediatric Otorhinolaryngology (2005) 69, 49—56

www.elsevier.com/locate/ijporl

Comparison of caregiver otitis media risk factorknowledge in suburban and urban primarycare environments$

Joseph E. Kerschnera,b,*, D. Richard Lindstroma,Albert Pomeranzc, Robert Rohloffc

aChildren’s Hospital of Wisconsin, Department of Otolaryngology and Communication Sciences,Medical College of Wisconsin, 9000 W. Wisconsin Avenue, Milwaukee, WI 53226, USAbDivision of Pediatric Otolaryngology, Medical College of Wisconsin, Milwaukee, WI, USAcDepartment of Pediatrics, Medical College of Wisconsin, Milwaukee, WI, USA

Received 28 April 2004; received in revised form 4 August 2004; accepted 6 August 2004

KEYWORDSOtitis media;Risk factors;Caregiver knowledge

Summary

Objective: There are many risk factors for otitis media. Some of these, such aspassive tobacco smoke exposure and childcare arrangements; have the potential to bemodified. The purpose of this study is to assess caregiver knowledge deficits about riskfactors associated with otitis media and their willingness to modify behaviorsassociated with those risks.Research design and methods: This study is a prospective survey study investigatingknowledge deficits of parents or guardians of children ages 6—36months about the riskfactors of otitis media. The patients were consecutively drawn from a suburban and anurban pediatric practice. Any difference in survey results between these two groupswas also assessed. Participants completed a survey of 21 questions with contentincluding demographic and OM risk factor data.Results: A total of 401 caregivers completed surveys, with 213 from an urbanpediatric practice and 188 from a suburban practice. There was a significant differ-ence in the ethnic distributions of the two populations. The suburban population had asignificantly greater family history of ear infections, number of ear infections in thepast 12 months, and number of previous ventilation tubes placed. The urban popula-tion had a significantly greater number of smokers in the household and decreasedknowledge about day care as a risk for OM. The urban population’s question responses

$ Portions of this manuscript were presented at the Society for Ear, Nose and Throat Advances in Children (SENTAC) Annual Meeting,New Orleans, LA, 31 October 2003.

* Corresponding author. Tel.: +1 414 266 6476; fax: +1 414 266 6989.E-mail address: [email protected] (J.E. Kerschner).

0165-5876/$ — see front matter # 2004 Elsevier Ireland Ltd. All rights reserved.doi:10.1016/j.ijporl.2004.08.004

50 J.E. Kerschner et al.

suggested a greater willingness to change day care arrangements to reduce the risk ofotitis media.Conclusions: Both populations demonstrated knowledge deficits regarding risk fac-tors associated with OM and both populations exhibited willingness to modify beha-viors to reduce risk. These findings demonstrate that there are opportunities forimproving education regarding OM risk factors and that this education could poten-tially reduce risk for OM and in turn reduce the incidence of OM in children.# 2004 Elsevier Ireland Ltd. All rights reserved.

1. Introduction

Otitis media (OM) is the most common reason for achild to see a physician for illness in the UnitedStates and has an estimated annual cost of greaterthan US$ 5 billion [1]. Many risk factors (RF) for OMhave been documented. Some are unalterable suchas family history, craniofacial abnormalities andmale gender. Others are potentially modifiableincluding breast-feeding, tobacco exposure, child-care arrangements, and immunizations. While theseRF have been well documented through numerousstudies [2—4], there has been limited study of howwell information regarding RF is communicated tocaregivers and what knowledge of these RF care-givers retain. One strategy for reducing OM inci-dence is to reduce RF associated with the disease[5]. In order to accomplish this, there must first be agood understanding of what themodifiable RF are bycaregivers, and also a willingness to modify behaviorto reduce those RF.

The purpose of this study was to assess caregiverknowledge of OM RF and to assess our hypothesisthat there is a general knowledge deficit regardingRF of OM by caregivers. We also assessed our hypoth-esis that caregivers would also express some will-ingness to change behavior if these changes resultedin less risk for OM. We assessed caregiver knowledgeregarding RF and attitudes regarding behavior mod-ification from two different practice settings todetermine if there was a difference between thesetwo settings. To answer these questions, a ques-tionnaire was administered to caregivers at an urbanand a suburban pediatric practice.

2. Methods

2.1. Study design

This study is a prospective, comparative surveystudy investigating knowledge deficits of parentsor guardians of children ages 6—36 months aboutrisk factors for OM. The survey (Appendix A) wasdesigned with the assistance of the Children’s

Hospital of Wisconsin National Outcomes ResearchCenter and was reviewed and approved by theChildren’s Hospital of Wisconsin Institutional ReviewBoard. The survey items were chosen to assesscaregiver knowledge of known risk factors for OM.Due to space and time limitations, an attempt wasnot made to questions caregivers about every riskfactor previously described in association with otitismedia. The questionnaire assessed caregiver knowl-edge of modifiable and non-modifiable RF for OM.The questionnaire also included items assessing thewillingness of caregivers to modify their behavior inorder to reduce the risk of OM for their child.Further assessment of willingness to modify beha-vior was measured by including questions pertainingto avoidance of surgical intervention through beha-vior modification. In addition, questions concerningantibiotic efficacy and vaccines were included toassess overall knowledge about otitis media.

Results were compared for the urban populationand the suburban population. The urban practiceconsists of general pediatricians with an academicappointment to the Medical College of Wisconsinand is a setting for medical student and pediatricresident training. Patients seen in this setting aregenerally cared for by a team of general pediatri-cians, residents and medical students. The suburbanpractice consists of a four person general pediatricpractice. These pediatricians have a clinicalappointment to the Medical College of Wisconsinand outpatients seen in this setting are generallycared for by the attending physician without resi-dent or medical student involvement.

2.2. Patient recruitment and questionnairedevelopment

The patients were consecutively drawn from a sub-urban and an urban pediatric practice from April2002 to September 2002. Caregivers were asked toparticipate in the survey at the time of registeringfor their child’s appointment. Caregivers (parents orlegal guardians) of children 6—36 months were eli-gible to enroll in the study. The caregivers were notrequired to have a child being seen for a problem

Caregiver otitis media risk factor knowledge 51

related to OM or a history of OM to enroll. Anydifference in survey results between these twogroups was also assessed. Participants completeda survey of 21 questions with content includingdemographic data of the caregiver’s child and OMRF questions. Demographic data corresponding tothe caregivers completing the survey was not col-lected.

2.3. Data analysis

Statistical analysis of survey results includeddescriptive statistics, Levene’s test for equality ofvariances, t-test for equality of means and Mann—Whitney testing. Survey design and statistical ana-lysis was done in consultation with the NationalOutcomes Research Center at Children’s Hospitalof Wisconsin.

Table 2 Risk factors

Urban(%)

Suburban(%)

Siblings with OM 18.5a 47.6a

Parent with OM history 7.2a 37.8a

3. Results

3.1. Demographics

A total of 401 caregivers completed surveys with 213from an urban pediatric practice and 188 from asuburban practice. There was a significant differ-ence in the ethnic distributions of the two patientpopulations (Table 1). The suburban population wasoverwhelmingly Caucasian, whereas the urban prac-tice was primarily African American. The genderdistribution was similar for each location.

3.2. Risk factor exposure

The overall data for RF exposure revealed that28.6% of caregivers noted smokers in the household.Caregivers (38.7%) utilize day care for their chil-dren. Patients (56.1%) indicated that they had beeneducated regarding the RF for OM at their pediatri-

Table 1 Demographics

Urban (%) Suburban (%)

GenderFemale patients 52.7 45.7Male patients 47.3 54.3

Ethnicity/raceAfrican American 79.5a 0.0a

Caucasian 6.9a 95.9a

Hispanic 9.6 2.5Native American 2.7 0.8Asian 1.4 0.8a Indicates statistically significant difference between

populations.

cian’s office. There were no significant differencesbetween the suburban and urban groups regardingday care use and previous discussion of RF at theirpediatrician’s office.

In comparing the two cohorts in regard to the RF,the urban population had a significantly greaternumber of smokers in the household (p < 0.0001).The suburban population had a significantly greaterfamily history of ear infections (p< 0.0001), numberof ear infections in the past 12 months (p = 0.004),and number of previous tympanostomy tubes placed(p = 0.0003) (Table 2).

3.3. Knowledge of risk factors

There was no significant difference between cohortsfor knowledge about tobacco exposure, breast-feeding or family history of OM as OM RF.

However, the urban population responses sug-gested less knowledge about day care as a risk factorfor OM (p< 0.004), but greater knowledge about thepossibility of a vaccine to limit OM risk (p = 0.05).The urban population responses to the questionregarding antibiotic efficacy also suggested lessknowledge regarding antibiotic resistance (p <0.0001) (Table 3).

An overall assessment of caregiver knowledge ofrisk factors suggested that the urban cohort hadsignificantly less knowledge of OM risk factors com-pared with the suburban cohort (p = 0.0003).

The overall knowledge about OM RF were alsoassessed within each cohort separately. Knowledgeabout individual, modifiable risk factors was great-

Patient with history ofventilation tubes

4.7a 15.5a

Siblings with history ofventilation tubes

14.1 10.2

Smokers in household 43.4a 11.8a

Patient in daycare 41.8 35.1Discussion of OM risk factors

with doctor53.4 59.0

OM episodes in past 12 months0 45.6 32.41—2 35.0 40.13—4 15.5 18.15—6 1.5 5.07 or more 2.4 4.4a Indicates statistically significant difference between

populations.


Table 3 Knowledge of risk factors

Survey question Mean scoresa (five-point scale)

Urban Suburban

Contact with cigarette smoke increases OM risk 1.98 1.94Breast-feeding decreases OM risk 2.17 2.08Day care attendance increases OM risk 2.27b 1.91b

Family history of OM increases OM risk 2.43 2.58There is a vaccine which decreases OM risk 2.76b 3.04b

Antibiotics are more effective than ever against OMc 4.00b 3.06b

a Scale from 1 to 5 with 1 indicating strong agreement and 5 indicating strong disagreement.b Indicates statistically significant difference between populations.c Note: Analysis of scale was reversed for this question so that, similar to all other questions, a higher score correlated with a

greater knowledge deficit concerning risk factor association with otitis media.

est concerning the risk of tobacco exposure inthe urban population. Compared to this understand-ing of tobacco as a RF, there was a significantlylower understanding of the risks associated withdaycare attendance (p = 0.001) and that therewas the potential for a vaccine to limit OM (p <0.001). In the suburban population there wasequivalent understanding concerning the risk oftobacco and daycare attendance but a significantlylower understanding among suburban respondentsabout the potential for a vaccine to limit OM(p < 0.001).

3.4. Risk factor modification

The mean for all responses to the scaled questionsregarding behavior modification are listed in Table 4.There was a significant difference between theurban and suburban populations in their willingnessto change daycare arrangements to reduce OM risk orthe need for ventilation tubes (p = 0.012 and 0.024,respectively), with the urban population more will-ing to change. There were trends suggesting that thesuburban population would be more willing to quitsmoking to reduce OM risk or the need for ventilationtubes.

Table 4 Behavior modification

Survey question

Willing to change smoking habits to reduce OM riskWilling to change smoking habits to reduce need for ear tuWilling to change day care arrangements to reduce OM riskWilling to change day care arrangements to reduce need fo

a Scale from 1 to 5 with 1 indicating strong agreement and 5 indb Indicates statistically significant difference between populatio

4. Discussion

Studies examining risk factors associated with OMhave demonstrated characteristics that are amen-able to risk factor modification and those that areintrinsic and cannot be modified [2—4]. UnalterableRF include family history, gender, race and anatomiccharacteristics [2—4]. Potentially modifiable RFinclude tobacco exposure, childcare arrangements,and breast-feeding [2—4]. Knowledge of these RFamong healthcare providers is well accepted, how-ever, little has been done to study how well thisknowledge is passed on to caregivers of young chil-dren. Even less has been written about the will-ingness of caregivers to modify their behavior tolessen their child’s risk of OM or sequelae associatedwith OM.

There is only one previous large survey analysis inthe literature dealing with caregiver knowledge ofOM RF that was identifiable using a Medline search ofthe English literature [6]. This study examinedmaternal knowledge of OM RF in mothers of infants9—15 months of age using a large cross-sectionaldesign. Results indicated that only 25% of caregiversin a population-based sample had been educated asto the RF of OM. Logistic regression revealed that aninfant’s OM history and Caucasian ethnicity were

Mean scoresa (five-point scale)

Urban Suburban

1.62 1.35bes 1.53 1.19b 2.20b 2.62b

r ear tubesb 2.10b 2.42b

icating strong disagreement.ns.


significantly associated with increased caregiverknowledge regarding RF for OM.

In this current survey, caregivers reported morewidespread education about OM RF than the pre-vious report. However, approximately 44% of care-givers reported that they had not been educatedabout OM RF by their primary care physician. Thisrepresents an important segment of the populationwith the potential for knowledge deficits. Theincreased percentage of caregivers whom reportedreceiving education regarding OM RF in this survey,compared with the previous study, might representof number of factors. These include that olderchildren were included in this survey giving moreopportunity for education, or that more emphasishas been placed on OM and OM education in thesurvey sites assessed. This study did not find adifference between the suburban or urban practicesregarding caregiver’s perceptions of OM risk factoreducation being done by the primary care physician.

4.1. Demographics

Gender distribution of patients of caregivers wassimilar for both cohorts in this study. The ethnicbackground in the different cohorts revealed a sig-nificant difference with more African Americanpatients in the urban practice and more Caucasianpatients in the suburban practice. Previous reportshave documented a difference in rates of OM basedon race and socioeconomic situation [3]. This surveysupports previous studies with a reported increasedincidence of OM and previous ear tube placement inour suburban, primarily Caucasian, population.Despite these differences in race and OM incidencethere was not a reported difference in educationregarding OM RF at these two sites.

4.2. Risk factor incidence and knowledge

Exposure to modifiable RF differed between urbanand suburban caregivers. Patients from an urbanpopulation were more likely to be exposed totobacco smoke despite the fact that there was nodifference in the knowledge of tobacco as a riskfactor for OM. Day care usage was no different forthe two cohorts, but knowledge of day care usage asa risk factor for OM was significantly greater in thesuburban cohort. Awareness of antibiotic resistanceas a potential contributing factor in OM was alsogreater in the suburban cohort. The urban popula-tion demonstrated increased knowledge about avaccine to potentially protect against OM.

For each separate, modifiable RF, in both popula-tions there was a greater knowledge about theassociation of tobacco exposure as a RF for OM than

there was about the potential to decrease OM with avaccine. In addition, the urban population had abetter understanding of the association of tobaccoand OM than daycare and OM. These findings wouldsuggest that educational materials and strategiesfocused on these less well understood RF mightprovide greater impact in RF modification.

The differences in understanding regarding theRF associated with OM between the two sites existeddespite caregivers reporting similar levels of educa-tion provided by healthcare professionals regardingRF on their survey questions. There are severalpossible explanations for the different level ofunderstanding regarding RF despite apparent equalattempts at education. These include that neitherparticipating site employs standardized materialsfor education regarding OM RF and that the qualityand scope of education might therefore varybetween patients. The increased incidence of OMin the suburban cohort might also suggest that morefrequent or more in-depth discussions regarding OMmight have taken place with this group of patients.Additional factors that could be important in leadingto these differences are the potential impact ofeducational background, access to healthcare andcontinuity of care provided to patients. These areasof interest deserve further study.

4.3. Behavior modification

Perhaps the most salient finding in this study is thatas a raw score on a five-point scale, the willingnessto change behaviors regarding tobacco exposure andchildcare arrangements was very affirmative in boththe suburban and urban populations (Table 4). In allcases, the willingness to modify behavior was evenmore affirmative if this modification could be linkedto avoidance of surgery with placement of ventila-tion tubes to treat the OM. This indicates a signifi-cant willingness on the part of caregivers to modifybehaviors to lessen the risk of OM for their children.These findings are important in that they suggestthat programs targeted at OM risk factor educationand linked with risk factor modification strategiessuch as smoking cessation programs will find willingparticipants and could have the potential to posi-tively affect risk factor exposure of young childrenand reduce OM incidence. The questions regardingrisk factor modification were only directed towardscaregivers whose children were exposed to the riskin question, smokers and users of daycare, andtherefore this willingness to change behavior existsin the populations at risk. It was beyond the scope ofthis project to examine which type of educationalefforts might have the most impact on changingbehavior, but multiple avenues should be explored


in addition to education by physicians includingmedia campaigns and community efforts.

Caregivers from the urban cohort were signifi-cantly more willing to consider changing childcarearrangements to reduce OM risk for their children.This difference from the suburban cohort existeddespite a greater understanding of the suburbanpopulation regarding the association of daycareand OM. As daycare is often considered the mostimportant risk factor associated with OM these find-ings warrant further investigation.

4.4. Potential study weaknesses

This study also did not attempt to standardizepatient populations in terms of OM incidence,patient age or other medical conditions. As withall survey instruments this study relied on caregiverperceptions rather than more objective measure-ments. This survey also had a focus on past eventsrather than prospective measurements which per-mits recall bias. The survey used for this study didnot undergo validation in other settings, did notassess a broad cross-section of the population andwas not assessed for possible cultural and educa-tional biases. In addition, this survey did notattempt to correlate caregiver knowledge aboutOM RF, or willingness to change behaviors, to theirown child’s particular incidence of OM and thesewould be areas of interest in future investigations.

5. Conclusions

There is a knowledge deficit regarding the RF for OMby caregivers in both urban and suburban pediatricpractices. However, the caregivers of these childrendo exhibit a significant willingness to modify beha-

viors that might reduce the risk of OM to their child.Standardized teaching materials could be madeavailable to assist with education of caregiversand could be a first step in behavior modificationin an attempt to reduce the incidence of OM. Thissurvey study certainly has some limitations, how-ever, the data obtained from this investigationdemonstrates a number of important points andsuggests some clear avenues of future investigationin an area that has been scientifically neglected.Further study is needed to assess if education alonecan be effective in risk factor modification or if moreintensive programs are needed to assist with riskreduction. Further study is also needed to assess ifdifferent strategies are required to erase knowledgedeficits in different types of practice locations.Some of this work has already been initiated. Giventhe high prevalence of OM and its potential formorbidity, the significant health care expendituresinvolved in treating this disease and themore recentdeclining efficacy of antimicrobial therapy due toantibiotic resistance, increased study of risk reduc-tion with implementation of strategies to reducethis risk should be pursued.

Acknowledgements

Special thanks to all of the physicians, residents,medical students and staff at the Downtown HealthClinic, Milwaukee, WI, who assisted in collectingdata for this project and to Dr. Rohloff’s colleagues(Drs. Gutzeit, Jeruc, Richter and Young) as well astheir staff at Southwest Pediatrics, New Berlin, WI.Also, special thanks to Stephanie Frisbee and EvelynKuhn from the National Outcomes Research Centerat Children’s Hospital of Wisconsin for their assis-tance with survey design and data analysis.


Appendix A. Ear infection survey


References

[1] J. Bondy, S. Berman, J. Glazner, D. Lezotte, Direct expendi-tures related to otitis media diagnoses: extrapolations from apediatric medicaid cohort, Pediatrics 105 (2000) e72.

[2] P. Auinger, B.P. Lanphear, H.J. Kalkwarf, M.E. Mansour, Trendsin otitis media among children in the United States, Pediatrics112 (2003) 514—520.

[3] K.A. Daly, M.L. Casselbrant, H.J. Hoffman, L.B. Ingvarsson,K.J. Kvaerner, M. Tos, P.B. Van Cauwenberge, Recent

advances in otitis media. 2. Epidemiology, natural historyand risk factors, Ann. Otol. Rhinol. Laryngol. 188 (Suppl.)(2002) 19—25.

[4] K.A. Daly, Epidemiology of otitiis media, Otolaryngol. Clin.North Am. 24 (1991) 775—785.

[5] J.V. West, Acute upper airway infections, Br. Med. Bull. 61(2002) 215—230.

[6] K.A. Daly, R.E. Selvius, B. Lindgren, Knowledge and Attitudesabout otitis media: implications for prevention, Pediatrics100 (1997) 931—936.

comparison of caregiver otitis media risk factor knowledge in suburban and urban primary care...

Documents