Original research articleAudio computer-assisted self-interviewing in reproductive health research: reliability assessment among women in Harare, Zimbabwe☆,☆☆
Introduction
The challenge of accurately measuring reproductive health behaviors has received substantial attention in the context of HIV prevention research [1], [2]. With no gold standards, reliability studies have aimed to improve the accuracy of self-reports through the modification of interview mode (e.g., self-administered vs. interviewer-administered instruments) and to strengthen assessments of self-reported behavioral risks, such as condom-use practices, with biological markers of sexual activity [e.g., Y chromosome, prostate-specific antigen (PSA)] [3], [4], [5]. Relatively few studies, however, have addressed the reliability of self-reports on the use of hormonal contraceptive methods. Accurate data on contraceptive use are needed both to calculate the effectiveness rates of methods and to minimize misclassification in observational epidemiologic studies, which, particularly for low-magnitude associations, can introduce biases that render interpretation of such associations tenuous [6].
Most methodological studies of contraceptive method use have investigated the extent of recall bias in self-reported use by comparing reports (both in self-administered and interviewer-administered paper questionnaires) to clinic records [7], [8], [9], [10]. An evaluation of reported oral contraceptive use in the Nurses' Health Study against pharmacist records found that reports of pharmacology and dose were concordant for 75% of sampled respondents [11]. Another study on compliance to oral contraceptive method, which compared the diary data of daily pill use to data from electronic dispensers, demonstrated that diary data were susceptible to social desirability and recall biases (depending on when respondents actually completed their diaries). Reports of missed pills were concordant for only 45% of participants, with women reporting fewer pills missed in diaries than in electronic data [12].
Research on measurements of the use of barrier methods for HIV prevention has shown that interview methods that increase both the privacy and the anonymity of the interview context can increase reports of high-risk behaviors, particularly those that may be embarrassing, stigmatizing or illegal [1], [13]. Audio computer-assisted self-interviewing (ACASI), which achieves the privacy of a self-administered questionnaire [14] but allows for complex and consistent questionnaire administration, has become more widely used because it appears to yield more accurate self-reports for the most sensitive behaviors [15], [16], [17], [18], [19], [20], [21], [22], [23].
Nearly all studies examining ACASI, though, focused primarily on HIV prevention behaviors by using a cross-sectional design and were conducted in the United States or Great Britain. With the exception of one recent study among Kenyan adolescents [24], which found inconsistent effects of ACASI on reports of HIV risk behaviors, this study is among the first investigations of ACASI to be carried out in sub-Saharan Africa and the first to examine hormonal contraceptive use.
In the context of a prospective cohort study of hormonal contraceptive use and HIV acquisition (HC–HIV Study) [25], we conducted a prospective, randomized, cross-over study among women in Zimbabwe to examine misclassification of exposure, hormonal contraceptive use and related reproductive health risks. We compared self-reports obtained through ACASI and face-to-face (FTF) questionnaires and examined the concordance of self-reports and clinical data on contraceptive method disbursement.
Section snippets
Study population
This study was ancillary to the multisite HC–HIV Study conducted among a cohort of HIV-negative women in Zimbabwe, Uganda and Thailand. Because previous research conducted in Zimbabwe had indicated the acceptability of ACASI methods [26], this substudy was added to the Zimbabwe site only. The ACASI Study was conducted in two periurban communities outside Harare: Epworth and Chitungwiza. Eligible women were 18–35 years old, sexually active, HIV-negative and agreed to be tested for HIV every 3
Study participant characteristics
The HC–HIV Study Zimbabwe cohort enrolled 1092 women from the two clinical sites where we conducted the ACASI Study. The ACASI Study included 655 women, or 59.9% of the total seen in these clinics. ACASI Study participants appear to be representative of the overall HC–HIV cohort; none of the background characteristics or sexual history factors examined differed between ACASI Study participants and the remaining cohort members. Follow-up rates were 90.2% for Visit 2, 89.3% for Visit 3 and 82.5%
Discussion
Assuming that higher reports of sensitive behaviors are more accurate, our findings generally suggest that ACASI can minimize social desirability bias. Reports of hormonal contraceptive use — the main exposure examined by the parent study — appeared to be influenced by interview mode, with lower reports of use in ACASI interviews than in FTF interviews. That lower reported use of hormonal contraceptives was found in ACASI for the pill and Depo-Provera cohorts, but not for the nonhormonal cohort
Acknowledgments
We would like to acknowledge the study interviewers and data clerks for their tremendous contributions to recruitment and data collection, and Mavis Rimai for superb data entry and management assistance. Dr. Minnis also would like to thank Drs. Arthur Reingold and Lorraine Midanik, professors at UC Berkeley, for reviewing an earlier version of this manuscript (as part of her doctoral dissertation) and for guiding its analysis and interpretation. Thanks are extended to Markus Steiner for his
References (36)
- et al.
Use of prostate-specific antigen (PSA) to measure semen exposure resulting from male condom failures: implications for contraceptive efficacy and the prevention of sexually transmitted disease
Contraception
(2003) - et al.
Reproducibility of oral contraceptive histories and validity of hormone composition reported in a cohort of US women
Contraception
(1997) - et al.
Self-reported honesty among middle and high school students responding to a sexual behavior questionnaire
J Adolesc Health
(1998) Interview mode and measurement of sexual behaviors: methodological issues
J Sex Res
(1999)- et al.
Methodological problems in AIDS behavioral research: influences on measurement error and participation bias in studies of sexual behavior
Psychol Bull
(1990) - et al.
Using Y chromosome PCR as a biomarker for sexual intercourse: studies of condom efficacy
- et al.
Efficacy of the female condom as a barrier to semen during intercourse
Am J Epidemiol
(2003) Bias in the evaluation of low-magnitude associations: an empirical perspective
Am J Epidemiol
(2000)- et al.
Agreement rates between oral contraceptive users and prescribers in relation to drug use histories
Am J Epidemiol
(1978) - et al.
Agreement between women's histories of oral contraceptive use and physician records
Int J Epidemiol
(1983)
Comparison of recalled and validated oral contraceptive histories
Am J Epidemiol
Validity of contraceptive histories in a rural community in Kenya
Int J Epidemiol
Measuring compliance among oral contraceptive users
Fam Plann Perspect
Mode effects in surveys of gay men: a within-individual comparison of responses by mail and by telephone
J Sex Res
Adolescent sexual behavior, drug use, and violence: increased reporting with computer survey technology
Science
Audio-computer interviewing to measure risk behavior for HIV among injecting drug users: a quasi-randomised trial
Lancet
Under-reporting of alcohol consumption in sample surveys: the effect of computer interviewing in fieldwork
Br J Addict
Audio-CASI vs. interview method of administration of an HIV/STD risk of exposure screening instrument for teenagers
Int J STD AIDS
Cited by (0)
- ☆
This project has been funded with federal funds from the National Institute of Child Health and Human Development, National Institutes of Health, Department of Health and Human Services through a contract with Family Health International (contract no. N01-HD-0-3310).
- ☆☆
The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services or of Family Health International, nor does mention of trade names, commercial products, or organizations imply endorsement by the US Government.