2006 HSR&D National Meeting Abstract
3019 — Patient Characteristics Associated with Participation in DNA Banking: National Veterans ALS Registry
DiMartino LD (VA Center for Health Services Research in Primary Care)
Allen KD (VA Center for Health Services Research in Primary Care)
Kasarskis EJ (Lexington VA Medical Center )
Coffman CJ (VA Center for Health Services Research in Primary Care)
Lindquist JH (VA Center for Health Services Research in Primary Care)
Oddone EZ (VA Center for Health Services Research in Primary Care)
A primary goal of the National Registry of Veterans with Amyotrophic Lateral Sclerosis (ALS) is to build an associated DNA Bank, so that stored samples from Registry participants can be used in future studies. Little is known about patient characteristics that may influence participation in DNA Banks. This study examines demographic, clinical, and military characteristics associated with participation in the Registry DNA Bank.
Registry participants were initially contacted about the DNA Bank via telephone and then mailed a consent form. Participants (N=959) were 98% male, 90% white, and the mean age was 64 years. 60% of participants were current VA health system users, and the branches of service were: Army (46%), Navy (23%), Air Force (23%), and Marines (8%). Participants were categorized as consented, active refusals, or passive refusals (did not return a consent form after >3 months and multiple reminders). Chi-square and t-tests examined bivariate associations of gender, race, military branch, VA health system use, age, time since symptom onset, and years of military service with consent or refusal status. We also examined associations using a multivariable logistic regression model.
A total of 13.5% of Registry participants refused DNA Banking (8.5% active, 5% passive). The proportion of refusals varied significantly according to: VA health system use (users=18%, non-users=7%, p<0.01), race (non-white=28%, white=12%, p<0.01), and military branch (Army=16%, Navy=12%, Air Force=14%, Marines=4%, p<0.05). In the multivariable logistic regression model, only VA health system use was significantly associated with refusal, where current users had 2.37 times the odds of refusing as compared to non-users (95% CI: 1.43–3.96).
Race, branch of service, and VA health system use were associated with higher rates of refusal to participate in the Registry DNA Bank. However, in the adjusted analysis, VA health system use was the only variable significantly associated with refusal. Reasons for this association are unclear.
DNA Banks can be a valuable adjunct to both clinical trials and health services research. In this study, refusal rates differed according to some patient characteristics. This may influence generalizability of DNA Banks, and further efforts are needed to understand and intervene to reduce these differences.