Assessing nonresponse bias at follow-up in a large prospective cohort of relatively young and mobile military service members

Abstract

BackgroundNonresponse bias in a longitudinal study could affect the magnitude and direction of measures of association. We identified sociodemographic, behavioral, military, and health-related predictors of response to the first follow-up questionnaire in a large military cohort and assessed the extent to which nonresponse biased measures of association.MethodsData are from the baseline and first follow-up survey of the Millennium Cohort Study. Seventy-six thousand, seven hundred and seventy-five eligible individuals completed the baseline survey and were presumed alive at the time of follow-up; of these, 54,960 (71.6%) completed the first follow-up survey. Logistic regression models were used to calculate inverse probability weights using propensity scores.ResultsCharacteristics associated with a greater probability of response included female gender, older age, higher education level, officer rank, active-duty status, and a self-reported history of military exposures. Ever smokers, those with a history of chronic alcohol consumption or a major depressive disorder, and those separated from the military at follow-up had a lower probability of response. Nonresponse to the follow-up questionnaire did not result in appreciable bias; bias was greatest in subgroups with small numbers.ConclusionsThese findings suggest that prospective analyses from this cohort are not substantially biased by non-response at the first follow-up assessment.

Relationships between dental appearance, self-esteem, socio-economic status, and oral health-related quality of life in UK schoolchildren: A 3-year cohort study

“…Although we did achieve the estimated sample size of 198 in the follow-up examination, because CHQ-CF87 was not included in the baseline data collection for two schools early in the study, the response rate for the longitudinal element was 58% (n = 173/296) for those with baseline self-esteem data, but below half of the original baseline sample (46%; n = 173/374). There is the risk that those participants who provided data for the follow-up examination were different to those who did not provide data, thereby introducing the possibility of non-response bias (42). We found it increasingly difficult to obtain the full cooperation of schools as the children got older, due to the pressures of formal examinations and assessments.…”

Design, methods and demographics from phase I of Alberta's Tomorrow Project cohort: a prospective cohort profile

“…These characteristics are typical of those commonly reported for participants thought to be more engaged in research. 45 Further work to elucidate patterns of response is required, taking into account health outcomes that may reduce the likelihood of questionnaire completion. Loss to follow-up is a major concern of longitudinal cohorts, and thus using retention strategies continues to be a priority for Alberta's Tomorrow Project.…”

Sesgo de respuesta en una encuesta postal poblacional sobre calidad de vida relacionada con la salud (SF-12v2)

“…Además, una alta tasa de respuesta no siempre garantiza la representatividad de la muestra y, por consiguiente, la validez de los resultados por lo que sería aconsejable analizar siempre las caracterís-ticas de los que no responden 8,24 independientemente de la tasa de respuesta. No obstante, la frecuente falta de información disponible sobre los que no participan impide llevar a cabo este análisis 25,26 .…”