Healthy Volunteer Bias in Recruiting Periodontitis Patients by Mail

BACKGROUND: Participants in clinical studies often have greater health awareness and compliance to treatment recommendations compared to non-participants. This healthy volunteer bias is more common when recruiting individuals from the general population. OBJECTIVE: To assess whether a healthy volunteer bias is present in the recruitment of patients for a study on periodontal treatment outcomes. METHODS: For each individual that was invited to participate, it was determined how long he/she had been eligible for dental services, and the rate of dental treatment utilization (number of procedures/year) was calculated. Dental treatment utilization rates were subgrouped using the CDT-4 treatment procedure classification codes, and differences in rates between participants and non-participants estimated using Poisson regression models. RESULTS: Between December 2002 and July 2003, we sent out 1737 invitation letters; 679 individuals have elected to participate so far. Individuals electing to participate had a 10% higher rate of dental treatment utilization rate (Relative Rate: 1.10, 95% confidence interval: 1.08-1.11; p < 0.0001). Participating individuals had a 12% higher usage rate of diagnostic services (p < 0.0001), a 13% higher usage of preventive services (p < 0.0001) , a 12% higher usage of restorative procedures (p < 0.0001), a 29% higher usage rate of endodontic procedures (p < 0.0001), and a 65% higher usage rate of orthodontic procedure codes (p < 0.0011). Participating individuals had a 9% lower rate of oral surgery procedures (p-value < 0.04). This difference in oral surgery procedure utilization was almost entirely due to a 13% lower rate of extraction procedures of erupted teeth (p-value < 0.0049). CONCLUSION: Individuals with periodontitis who agree to participate in a longitudinal study on periodontal treatment effectiveness are different from non-participants. Analysis of dental care utilization suggests a healthy volunteer bias – more prevention and fewer extractions of teeth. NIDCR R-01 DE13912