ARCHIVED - AN INTRODUCTORY LETTER IN ADVANCE OF A TELEPHONE SURVEY MAY INCREASE RESPONSE RATE
Telephone interviewing is an economical and effective way to conduct epidemiologic studies in the population. However, the validity of the results is often threatened by non-response bias stemming from refusal to participate. Introductory letters sent in advance of telephone survey contact have been suggested as a method of increasing response rates(1); however, few studies documenting this effect have been published.
In June 2002, we began a retrospective, cross-sectional telephone survey entitled "Magnitude of Enteric Illness in the Province of British Columbia: a Population Telephone Survey". The objective of the survey was to estimate the magnitude and describe the distribution of self-reported gastrointestinal illness in British Columbia. An introductory letter was used in an attempt to increase response rate.
The study area consisted of three regions, chosen to represent the province of British Columbia, which corresponded to one urban, one mixed, and one rural public health authority area. The sampling frame consisted of a randomized list of residential telephone numbers obtained from a commercial database (SelectPhoneTM, InfoUSA, Inc.). Once telephone contact had been made, one individual from each household - the one out of all household members whose birthday fell next - was randomly selected to participate in the survey.
All households in the sampling frame for which there were mailing addresses were sent an introductory letter approximately 1 to 2 weeks before the first telephone call attempt. The letter briefly explained the purpose of the study as well as the reasons why participation and random selection of one respondent from all household members were important.
Results from the 12-month survey are presented here. Of those reached to participate in the survey, 24% (1951/8090) received an introductory letter. The overall response rate was 57% (4,611/8,090). Table 1 shows response rates and the association between survey participation and receipt of the introductory letter.
The response rate among those who received an introductory letter was approximately 1.58 times greater than the response rate among those who did not. The association between receipt of an introductory letter and survey participation did not vary across the three regions of the study area, indicating its potential effectiveness in both rural and urban settings. The cost of including the introductory letter in the study was approximately 7% of the total cost of data collection.
The findings presented here are similar to those published in two Australian studies, one that used an introductory letter before recruitment of control subjects in a case-control study(2) and the other in advance of telephone interviewing(3). Both studies reported that introductory letters increased participation rates among those receiving them. In recent years, there appears to be a decreasing trend in response rates for epidemiologic research surveys conducted by telephone owing to a growing aversion to telemarketing and increased use of answering machines(4). The identification of effective techniques for increasing response rates in telephone surveys has therefore become more critical.
Table 1. Association between receipt of an introductory letter and survey response, by region, from the survey "Magnitude of Enteric Illness in the Province of British Columbia: a Population Telephone Survey", 1 June, 2002, to 31 May, 2003 (n = 8090)
(n = 2659)
(n = 2747)
(n = 2684)
(n = 8090)
|Overall response rate||58%||56%||57%||57%|
|Response rate among those who received letter||83%||76%||78%||79%|
|Response rate among those who did not receive letter||50%||50%||50%||50%|
|Crude relative risk (95% confidence interval)||1.66
|Mantel-Haenszel relative risk (95% confidence interval)||1.58
The main limitation of the results presented here is that this study was not designed to determine the impact of introductory letters on response rate, and thus the letters were not randomized to the sample population. Therefore, it is not possible to be certain that the better response in those who received letters was not due to other factors.
The use of introductory letters in advance of telephone surveys may provide an economical and effective way to increase survey response in future studies. This phenomenon deserves further research in a more controlled study to better determine whether introductory letters indeed cause increased response rate.
Kelsey JL, Whittemore AS, Evans AS et al. Methods in observational epidemiology, 2nd ed. New York: Oxford University Press, 1996.
Robertson B, Sinclair M, Forbes A et al. The effect of an introductory letter on participation rates using telephone recruitment (letter). Aust N Z J Public Health 2000;24(5):552.
Smith W, Chey T, Jalaludin B et al. Increasing response rates in telephone surveys: a randomized trial. J Public Health Med 1995; 17(1):33-8.
Hartge P. Raising response rates: getting to Yes. Epidemiology 1999;10(2):105-7.
Source: SE Majowicz, MSc, VL Edge, MSc, J Flint, MPH, P Sockett, PhD, K Doré, MHSc, Foodborne, Waterborne, and Zoonotic Infections Division, Public Health Agency of Canada, Health Canada, Guelph and Ottawa, Ontario; L McDougall, MSc, British Colum-bia Centre for Disease Control, Vancouver, British Columbia, and Field Epidemiology Training Program, Health Canada; V Remple, RN, MSN, M Fyfe, MD, British Columbia Centre for Disease Control; S Henson, PhD, Department of Agricultural Economics, University of Guelph; K Gaebel, MSc, Centre for the Evaluation of Medicines, St. Joseph's Hospital, Hamilton, Ontario.
Report a problem or mistake on this page
- Date modified: