Discrepant abortion reporting by interview methodology among men from the United States National Survey of Family Growth (2015-2017)

Abstract

Objective: To examine discrepancies in men's abortion reporting when queried via face-to-face interview versus audio computer-assisted self-interviewing (ACASI) in the National Survey of Family Growth (NSFG). Study Design: The NSFG collects nationally representative data on family life, sexual behavior, and reproductive health in the United States. The questionnaire is administered to participants via face-to-face interview (FTF), with selected items also asked of the same participant via ACASI for direct comparison. As the 2015-2017 NSFG queried individuals’ abortion history via both methods, we examined discrepant reporting among respondents. We additionally explored sociodemographic and reproductive characteristics associated with discrepant abortion reporting in a multivariable logistic regression model. Results: Of 4540 male respondents ages 15-49, 45.3% reported a pregnancy. Via FTF, 12.3% reported an abortion, compared to 19.9% via ACASI (p < 0.01). With respect to discrepancies in the number of reported abortions, 8.5% of respondents reported more abortions via ACASI versus FTF. Multivariable logistic regression modeling noted independently greater odds of abortion reporting in ACASI among non-Hispanic Black men (aOR 2.31, 95% CI 1.19-4.45), men living below the Federal Poverty Level (less than 100% FPL: aOR 3.65, 95% CI 1.93-6.89; 100-400% FPL: aOR 2.04, 95% CI 1.20-3.45), and those desiring more children in the future (aOR 1.91, 95% CI 1.20-3.04). Conclusion: Men were more likely to disclose their abortions in ACASI compared to FTF interview. Disproportionate, discrepant abortion reporting among low-income, minority men who report desiring more children in the future warrants further research. Implications: Surveys utilizing ACASI as an adjunct to FTF interviews may more accurately capture men's abortion experience.