Factors influencing the response to postal questionnaire surveys about respiratory symptoms.

Background: Response rates to postal questionnaires have been falling in recent years.

Aim: To examine factors affecting the response to five postal respiratory questionnaire surveys.

Design: Cross sectional study.

Setting: General practice.

Method: Five surveys were conducted in all adults registered with two UK general practices using an ECRHQ-based questionnaire, with two reminders at 4-week intervals.

Results: Response rates declined over time (1993 - 71.2%; 1995 - 70.5%; 1999 - 65.5%; 2001 - 65.3%; 2004 - 46.9%). Age and gender of non-responders were available for 2001 and 2004: responders were older (mean 48.8 years vs 37.6, p<0.001; 50.5 vs 38.8, p<0.001) and more likely to be female (54.9% vs 44.9%, p<0.001; 55.3% vs 48.5%, p<0.001). The response rate was increased by 18% (2004) and 23% (2001) by the use of two reminders. Early responders were older and more likely to be females, but were less likely to smoke than late responders after reminders. There was no important association between respiratory symptoms and associated feature prevalence and stage of response.

Conclusion: Declining response rates may represent reduced motivation and reluctance to share personal information. Qualitative exploration of late/non-response could help reduce bias when planning and analysing such surveys. The use of two reminders is an important factor in improving response.