Conferences

Results of mental health studies could potentially be biased by non-participation of respondents. Non-contact, refusal, ineligibility or death are main causes of non participation and especially in longitudinal (mental) health studies biased results can occur because non-random attrition is to be expected at each wave. In order to understand this bias it is important to determine the differences between those respondents who will continue participating in the study and those who do not. Also, such samples may have reasons for refusal which are specific to groups with mental health problems, such as lack of general motivation and concentration problems.

In studying non-response, refusal is of particular interest because it is amenable to change. The more we know about those who have been contacted, but refuse to take part, the more we are able to tailor our approach and hopefully lower our refusal rates. Especially in longitudinal studies much more is known about those who do not participate in second and subsequent waves and therefore longitudinal studies are prone to study characteristics of respondents and determinants of those who stop participating in the study.

The aim of this paper was to examine the relationship of refusals in the second and third wave, to the demographic, physical, psychological and social state of participants at subsequent waves in The Netherland Study of Depression (NESDA). A longitudinal mental health survey among 2981 persons at baseline (2006). Reasons for refusal are discussed as well as possible solutions to reduce these refusal rates. Also consequences for longitudinal data-analysis and imputation methods are introduced.