The requirement for prior consent to participate on survey response rates: a population-based survey in Grampian

Abstract

Background: A survey was carried out in the Grampian region of Scotland with a random sample of 10,000 adults registered with a General Practitioner in Grampian. The study complied with new legislation requiring a two-stage approach to identify and recruit participants, and examined the implications of this for response rates, non-response bias and speed of response.

Who are we missing? Area deprivation and survey participation

“…This opens the possibility of examining response rates (strictly 'proportions') according to postcode-derived ecological characteristics such as area deprivation. Several large population-based epidemiological surveys in the UK have used this strategy in recent years, looking at health and non-health outcomes in a range of study designs (1,(6)(7)(8)(9)(10). All eleven examples found strong evidence of lower response rates in more deprived areas, while to our knowledge no studies have found the reverse.…”

“…Most studies simply report non-participation, and no study goes beyond a binary categorisation of non-participation (usually non-contact vs. non-cooperation, although one study compares consenting to receive a questionnaire with subsequently completing that questionnaire (10)). Yet without knowing why people are not taking part, it is very hard to know best how to increase response rates; the best strategy for dealing with non-response because of out-of-date postal addresses will not be the same as the best strategy for dealing with non-response because potential participants do not feel motivated to take part.…”

“…Finally, while these examples include studies with a focus on both adults (1,8,10) and children (6,7,9,11), the surveys of children sought only to interview the children's parents, and not the children themselves. To our knowledge, no investigation has investigated child participation, or surveys in which some informants act as 'gatekeepers' determining whether other informants may be approached.…”

“…Several use deprivation measures based on relatively few area-level characteristics collected up to a decade previously (1,(8)(9)(10)(11). Others simply dichotomise areas into 'advantaged' and 'disadvantaged' wards (6,7), or look for heterogeneity within the category of 'deprived' areas (10); only one study provides sufficient raw data to examine whether area deprivation has a dose response effect on response rates across the whole spectrum of deprivation (9). As such, it is unclear whether it is only in the most deprived areas that non-response is lower.…”

Factors influencing the response to postal questionnaire surveys about respiratory symptoms

“…Issues regarding consent to participate in NHS research were explored in a Scottish study, 23 revealing serious implications for this kind of research if prior agreement to be approached is required. Agreement to be approached for research (keeping the option to decline consent) as a part of the 'contract' for receiving/providing NHS healthcare would allow the wider re-use of information and would substantially improve the accuracy of epidemiological research.…”