Original articles
Effect of Mailed Reminders on the Response Rate in Surveys Among Patients in General Practice

https://doi.org/10.1016/S0895-4356(99)00024-4Get rights and content

Abstract

Randomized trials were performed in Denmark and The Netherlands to determine the effect of mailed reminders on the response rate in surveys among patients in general practice. In both countries, general practitioners handed out questionnaires to 200 adult patients who came to visit them. An intervention group of 100 patients received reminders at 3 weeks after the visit, whereas a control group of the remaining 100 patients did not receive reminders. The response rate was significantly higher in the intervention groups than in the control group in The Netherlands (86% versus 55%, respectively) but not in Denmark (87% versus 81%, respectively). Mailed reminders can improve the response rate in surveys related to a general practice, but they are not effective in all situations.

Introduction

Response rates in surveys among patients vary considerably. A systematic review of mailed surveys among patients reported a standard deviation of 21% and a mean of 60% [1]. Nonresponse reduces the statistical power of a study, and it may introduce selection bias. For instance, nonresponders were more often seriously ill 2, 3, 4, were less satisfied with care provided [4], and were less frequent users of health care [5] compared with responders.

Several factors may influence the response rate of a survey: the motivation of the clinician to recruit patients [3], the attractiveness of the questionnaire 6, 7, the method of administering the questionnaire to patients 8, 9, the use of monetary incentives 10, 11, and even the type of stamps on the return envelopes [12]. Effective methods to improve the response rate in surveys among patients are needed. Research on survey methodology, mainly performed outside health care settings, points out a range of options, such as preliminary notification, repeated contacts, and personalization 6, 13, 14, 15, 16, 17. It is unclear to what extent these methods are also effective in health care.

Reminding patients to fill in a questionnaire, which was distributed earlier, could be an effective method [17]. However, the effectiveness of the use of reminders in surveys among patients in general practice has hardly been studied and may depend on the local situation or national culture. A notable exception was a study that compared telephone and recorded delivery reminders in a randomized trial [18]. The response rate among patients with asthma or diabetes mellitus increased from 61% to 71% if telephone or mailed reminders were used, with mailed reminders providing the best results. We formulated the following research question: What is the effect of mailed reminders in surveys among patients in general practice on the response rate?

Section snippets

Methods

This study is part of Europep, an international study on patients’ evaluations of general practice care. Randomized trials were performed in Denmark and The Netherlands to determine the effect of mailed reminders on the response rate. Four general practitioners in Denmark handed out 50 questionnaires, and five general practitioners in The Netherlands handed out 40 questionnaires to adult patients who consecutively visited them (200 patients per country). The patients were asked to fill in the

Results

Table 1 shows the results of the randomized trials. The response rate in The Netherlands was 86% in the patients who received reminders and 55% in patients without reminders (P < 0.001). The response rate in Denmark was 87% in the reminded patients and 81% in the patients without reminders (P = 0.25).

Discussion

Achieving a high response rate in surveys among patients is always a challenge for clinical researchers. This study showed that reminding patients in surveys related to general practice improved the response rate considerably in The Netherlands. It is reasonable to assume that some people may have lost the questionnaire or simply have forgotten to fill it in. The reminders can help to overcome these problems. It is also possible that the announcement of the general practitioners to the patient

Acknowledgements

This study was conducted with a European Union (Biomed) grant.

References (18)

  • D.A Asch et al.

    Response rates to mail surveys published in medical journals

    J Clin Epidemiol

    (1997)
  • J.F Etter et al.

    Analysis of non-response bias in a mailed health survey

    J Clin Epidemiol

    (1997)
  • M Schweitzer et al.

    Timing payments to subjects of mail surveysCost-effectiveness and bias

    J Clin Epidemiol

    (1995)
  • M Ehnfors et al.

    Patient satisfaction surveys subsequent to hospital careProblems of sampling, non-response and other losses

    Qual Assur Health Care

    (1993)
  • V Peto et al.

    Factors affecting general practitioners’ recruitment of patients into a prospective study

    Fam Pract

    (1993)
  • H.R Rubin

    Can patients evaluate the quality of hospital care?

    Med Care Rev

    (1990)
  • Ventura M Crosby Fm et al.

    Examination of a survey methodologyDillman’s Total Design Method

    Nurs Res

    (1989)
  • J.F Etter et al.

    Unexpected effects of a prior feedback letter and a professional layout on the response rate to a mail survey in Geneva

    J Epidemiol Community Health

    (1998)
  • C Bamford et al.

    Development of patient satisfaction questionnairesI. Methodological issues

    Qual Health Care

    (1992)
There are more references available in the full text version of this article.

Cited by (26)

  • Most response-inducing strategies do not increase participation in observational studies: a systematic review and meta-analysis

    2018, Journal of Clinical Epidemiology
    Citation Excerpt :

    Two reports were excluded from the meta-analyses due to insufficient information provided in the paper and no response from the authors. Therefore, 159 reports were available for analysis [22–180], in which 159 RCTs (with 358,528 potential participants) and nine observational studies (with 9,088 potential participants) were reported. The characteristics of the 168 eligible studies are shown in Appendix C on the journal's web site at www.elsevier.com.

  • Initial nonresponders had an increased response rate after repeated questionnaire mailings

    2005, Journal of Clinical Epidemiology
    Citation Excerpt :

    Response rates of about 70% have been achieved in surveys in other populations of older adults [3], so the response rate in this study was relatively low. It may be possible that a ceiling effect caused the limited effect in our study, although response rates of over 80% are achievable in surveys among patients in general practice [4]. Perhaps the effect of more intensive follow-up procedures is different, if the general practitioners hand out the first questionnaire themselves.

View all citing articles on Scopus
View full text