Does Personal Genome Testing drive service utilization in an adult preventive medicine clinic? N. Hoang1, R. Hayeems1, 4, J. Davies2, L. Velsher2, J. Aw2, S. Pu1, S. Wodak1, S. Chenier3, J. Stavropoulos1, R. Babul-Hirji1, R. Weksberg1, C. Shuman1 1) Clinical and Metabolic Genetics, Hospital for Sick Children, Toronto, Ontario, Canada; 2) Genetics, Medcan, Toronto, Ontario, Canada; 3) Département de pédiatrie, Centre Hospitalier Universitaire de Sherbrooke, Sherbrooke, Quebec, Canada; 4) Institute of Health Policy Management & Evaluation, University of Toronto, Toronto, Ontario, Canada.

   Background: Genome-wide single nucleotide polymorphism arrays that assess individual genetic risk for common polygenic diseases can guide the use of preventive health care services, but outcome data are limited. We aimed to understand how personal genome testing (PGT) risk results relate to subsequent health service utilization in a preventive medicine clinic. Method: We conducted a retrospective medical record review for two groups of clients at Medcan, an executive health clinic in Toronto. Ascertainment of cases over a 1 year period included those who pursued PGT at their comprehensive health assessment (CHA) (denoted CHA1); controls underwent CHA but with no PGT. We measured condition specific services used post CHA1 up to CHA3 (typically a 2 year period). PGT risk estimates for nine conditions were examined including: abdominal aneurysm, atrial fibrillation, celiac disease, colon cancer, type 2 diabetes, glaucoma, heart attack, melanoma, and prostate cancer. Client data were collected on: age, sex, family history, health risk status at CHA1 and CHA2, environmental modifiers (exercise level, alcohol intake, and smoking) and Medcan membership. Of 448 PGT ordered, 388 (86.6%) met inclusion criteria; a random sample of 388 matched Medcan clients were selected as controls. Analysis for all 9 conditions will be presented but the data here pertain to risk for heart attack. Results: Binomial log link regression analyses of relative risk (RR) indicate that being identified as at risk for heart attack on CHA1 and/or CHA2 is strongly associated with increased heart related services used compared to not being at risk [RR=4.44, 95% CI (3.57, 5.53)]. Similarly, age over 40 is associated with increased heart related services used compared to age under 40 [RRage41-50=2.03, 95% CI (1.33, 3.10); RRage50+=2.85, 95% CI (1.92, 4.23)]. Pursuing PGT itself was associated with increased services used [RR=1.48, 95% CI (1.20, 1.82)]. With respect to risk conferred by PGT, having an average risk for heart attack was associated with increased services used compared to the control group [RR=1.53, 95% CI (1.24,1.89)], but there was no difference between the increased risk group and the control group [RR=1.16, 95% CI (0.76, 1.79)]. Conclusions: For heart attack, the strongest drivers of related services used were CHA health risk status and age. It would appear that individuals who pursue PGT are likely to use more health related services regardless of the actual PGT test result.

You may contact the first author (during and after the meeting) at