Year: 2013
Type: article
Study design: Observational Study
Abstract: Background Reminders are routinely applied in surveys to increase response rates and reduce the possibility of bias, but it is uncertain whether these efforts actually affect response patterns. This study examines the effect of multiple reminders on response rates and derived estimates of prevalence and association in a national health survey. Methods Data derive from the Danish National Health Survey 2010 where 298,550 individuals (16 years or older) were invited to participate in a cross-sectional survey. Minimum two reminders were used and 177,639 (59.5%) individuals completed the questionnaire. Response rates, prevalence estimates and exposure-outcome relations were compared between four groups: those responding after first mailing (n = 109,479), second mailing (n = 40,575), third mailing (n = 27,336), respectively, and non-respondents (n = 120,911). A total of 249 (0.1%) respondents were excluded due to missing response date. Results The cumulative response rates after the first mailing, second mailing and third mailing were 36.7%, 50.3% and 59.5%, respectively. The inclusion of second and third mailing respondents didn’t change the overall characteristics of respondents compared with the target population; e.g. males (45.2% of first mailing, 45.8% of second mailing) were still underrepresented (49.2% of sampling frame), as were individuals with only basic schooling (26.5%, 28.2% and 31.5%, respectively). Only small changes in prevalence estimates were seen when including second and third mailing respondents compared with only first mailing respondents; e.g. daily smoking prevalence changed from 18.4% (95% CI: 18.1-18.6) to 19.9% (19.7-20.0). The observed exposure-outcome relationships were nearly identical; e.g. first mailing respondents that smoked daily had 2.25 (95% CI: 2.15-2.36) times the odds of having a high alcohol intake than first mailing respondents that didn’t smoke daily. The corresponding odds ratio was 2.28 (95% CI: 2.20-2.36) among all respondents. Conclusions Multiple reminders do increase response rates, but it is unlikely that they provide an effective way to reach the ‘hard core’ non-respondents. Overall, reminders have some effect on response rates and only minor impact on estimates of prevalence and association.
Authors: AI Christensen, O Ekholm, C Glümer, K Juel
Language: English
Subfield: Sociology and Political Science
Field: Social Sciences
Domain: Social Sciences
Open Access status: bronze
OpenAlex ID: W2586949033