Performance of the disease risk score in a cohort study with policy-induced selection bias

Mina Tadrous; Muhammad M Mamdani; David N Juurlink; Murray D Krahn; Linda E Lévesque; Suzanne M Cadarette

doi:10.2217/cer.15.40

Performance of the disease risk score in a cohort study with policy-induced selection bias

J Comp Eff Res. 2015 Nov;4(6):607-14. doi: 10.2217/cer.15.40. Epub 2015 Nov 3.

Authors

Mina Tadrous^{1

2

3}, Muhammad M Mamdani^{1

2

3

4}, David N Juurlink^{2

5}, Murray D Krahn^{1

2

4

6}, Linda E Lévesque^{2

7}, Suzanne M Cadarette^{1

2}

Affiliations

¹ Leslie Dan Faculty of Pharmacy, University of Toronto, 144 College Street, Toronto, ON, M5S 3M2, Canada.
² Institute for Clinical Evaluative Sciences, 2075 Bayview Avenue G1 06, Toronto, ON, M4N 3M5, Canada.
³ Applied Health Research Centre, Li Ka Shing Knowledge Institute, St Michael's Hospital, 30 Bond St, Toronto, ON, M5B 1W8, Canada.
⁴ Institute of Health Policy, Management & Evaluation, University of Toronto, 27 King's College Cir, Toronto, ON, M5S, Canada.
⁵ Sunnybrook Research Institute, 2075 Bayview Avenue, Toronto, ON, M4N 3M5T, Canada.
⁶ Toronto Health Economics & Technology Assessment (THETA) Collaborative, Toronto, ON, M5S 3M2, Canada.
⁷ Department of Public Health Sciences, Queen's University, 99 University Ave., Kingston, ON K7L 3N6, Canada.

PMID: 26529307
DOI: 10.2217/cer.15.40

Abstract

Aim: To examine the performance of the disease risk score (DRS) in a cohort study with evidence of policy-induced selection bias.

Methods: We examined two cohorts of new users of bisphosphonates. Estimates for 1-year hip fracture rates between agents using DRS, exposure propensity scores and traditional multivariable analysis were compared.

Results: The results for the cohort with no evidence of policy-induced selection bias showed little variation across analyses (-4.1-2.0%). Analysis of the cohort with evidence of policy-induced selection bias showed greater variation (-13.5-8.1%), with the greatest difference seen with DRS analyses.

Conclusion: Our findings suggest that caution may be warranted when using DRS methods in cohort studies with policy-induced selection bias, further research is needed.

Keywords: cohort studies; confounding factors (epidemiology); data interpretation; multivariate analysis; osteoporosis/epidemiology.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Bone Density Conservation Agents
Cohort Studies
Diphosphonates
Female
Hip Fractures
Humans
Insurance, Health, Reimbursement
Models, Statistical
Ontario
Organizational Policy*
Osteoporosis / drug therapy
Policy*
Risk Adjustment*
Selection Bias*

Substances

Bone Density Conservation Agents
Diphosphonates

Abstract

Publication types

MeSH terms

Substances

Grants and funding