BACKGROUND:Surveys of the population are commonly used to obtain information on health status. Increasingly, researchers are linking self-reported health status information to primary care consultation data. However, it is not known how participating in a health-related survey affects consultation behaviour. The objective of this study was to assess whether completion of a health-related questionnaire changes primary care consultation behaviour. METHODS:Participants were 3402 adults aged 50 and over from the general population in North Staffordshire, UK, who completed a health-related postal survey received in April 2003. The survey was predominantly about occurrence and severity of knee pain in the last year. Primary care attendance for the three months following response was compared to three control periods: i) the three months prior to the survey, ii) the same time period in the previous year and iii) the same time period in the following year. Comparisons were made on consultations for any problem, consultations for musculoskeletal disorders and consultations for knee problems. RESULTS:The percentage of subjects consulting for any condition was marginally higher for the three months directly after receipt of the questionnaire but the difference was only statistically significant in comparison to the three months before the survey (64% v. 62%, p = 0.05). There was little difference in consultation prevalence for musculoskeletal problems immediately after the survey compared to the three control periods. There was an increase of 37% in knee disorder consultations for the three months after the survey compared to the three months directly before the survey (p = 0.02). However, consultation prevalence for knee problems was identical for the three months after the survey to the same time periods in the years prior to and following the survey (both p = 0.94). CONCLUSION:The results from this study suggests that questionnaires related to physical health do not affect the standard consulting behaviour of patients, even for the symptom under investigation. This should reassure researchers who wish to link self-reported health status and medical care utilisation and clinicians whose patients are involved in such research.