To the Editor:
Accurate measurement of health behavior is critical to reliably estimating population prevalence, trends, and intervention effects. Although some objective measures exist, self-report of health behavior via questionnaires or interviews is often the only feasible method of collecting health behavior data in large population studies. Self-reported assessments of alcohol consumption and tobacco use typically underestimate actual use,1 whereas self-reported vegetable and fruit intake2 and physical activity3 typically are overestimates. A variety of survey design and administration strategies have been recommended to reduce such misreporting but few have been formally studied. We conducted a pilot randomized controlled trial to assess how providing introductory information and instruction to participants might affect their reporting of health behaviors.
The sampling frame was an existing research cohort. To be eligible, participants had to be at least 18 years old; speak English; and reside in New South Wales, Australia. Participants completed a computer-assisted telephone interview (CATI). Following a brief greeting, participants were randomized to receive a brief introductory statement and instruction (information-statement group) or to not receive this introductory information (no-information-statement group), using a random number function embedded in the CATI software. Participants, but not interviewers, were blind to allocation.
Interviewers telephoned potential participants, saying they were from the local health service and inviting them to participate in a 15-minute telephone survey. No other information was provided to participants randomized to the no-information-statement group. Participants allocated to the information- statement group were then read the following: “The call today will ask questions about a range of health behaviors. There are no right or wrong answers. Accuracy of the information is important to us as the information will be used to help health services. Some of the questions you may find difficult to answer, but please take your time and answer to the best of your knowledge.”
All the participants then completed a telephone survey including eight questions from the Active Australia survey concerning participation in physical activity in the previous week, 10 items of the Alcohol Use Identification and Disorders Test, two questions assessing usual daily vegetable and fruit consumption, and a single item about current smoking status; current smokers were also asked how many cigarettes they smoked per day. These behavioral assessments are commonly used in epidemiologic research in Australia.4,5 Of the 597 adults assessed for eligibility, 482 completed the telephone survey. The demographic characteristics of participants were similar between groups. Reported health behaviors by group are presented in the Table. The findings suggest that providing introductory information and instructions makes respondents more likely to disclose some undesirable health behaviors.
It is unclear why the intervention influenced some behaviors and not others. Previous studies have found that more comprehensive, multi-item assessments of health behaviors (such as the physical activity and alcohol use questionnaires we used) typically provide more valid measures of health behavior.6,7 In addition, reporting of stable health behaviors (such as tobacco or alcohol use) is thought to be more accurate than reporting of less stable behaviors (such as the quantity of food consumed or amount of exercise per occasion).8 When a health behavior is unstable and assessed with a single item (such as number of cigarettes or servings of fruit and vegetables) an information statement like the one we used may improve recall. A larger study is required to verify the effects reported here and to test hypotheses concerning the mechanisms of misreporting.
We acknowledge the contribution of telephone interviewing staff of Hunter New England Population Health for their review and piloting of the telephone survey script.
Erica L. James
Jeryl L. Francis
School of Medicine and Public Health
University of Newcastle
1. Stockwell T, Donath S, Cooper-Stanbury M, Chikritzhs T, Catalano P, Mateo C. Under-reporting of alcohol consumption in household surveys: a comparison of quantity-frequency, graduated-frequency and recent recall. Addiction. 2004;99:1024–1033
2. Amanatidis S, Mackerras D, Simpson JM. Comparison of two frequency questionnaires for quantifying fruit and vegetable intake. Public Health Nutr. 2001;4:233–239
3. Rzewnicki R, Vanden Auweele Y, De Bourdeaudhuij I. Addressing overreporting on the International Physical Activity Questionnaire (IPAQ) telephone survey with a population sample. Public Health Nutr. 2003;6:299–305
4. AIHW. The Active Australia Survey: a Guide and Manual for Implementation, Analysis and Reporting. Cat. no. CVD 22.. 2003 Canberra, Australia AIHW
5. Fleming J. The epidemiology of alcohol use in Australian women: findings from a national survey of women’s drinking. Addiction. 1996;91:1325–1334
6. Thompson FE, Subar AF, Smith AF, et al. Fruit and vegetable assessment: performance of 2 new short instruments and a food frequency questionnaire. J Am Diet Assoc. 2002;102:1764–1772
7. Dollinger SJ, Malmquist D. Reliability and validity of single-item self-reports: with special relevance to college students’ alcohol use, religiosity, study, and social life. J Gen Psychol. 2009;136:231–241
8. Bachman JG, O’Malley PM. When four months equal a year: Inconsistencies in student reports of drug use. Public Opin Quart. 1981;45:536–48