Doctor of Philosophy
Epidemiology and Biostatistics
Garg, Amit X
Background and objectives: In Ontario, primary care reform was initiated in the early 2000s with an aim to improve the quality of primary care. Hence, the provincial government restructured family physicians’ remuneration package. Prior to the reform, most physicians received majority of their income through fee-for-service (FFS). In Ontario, Family Health Group (FHG) and Family Health Organization (FHO) are dominant post-reform primary care models that remunerate family physicians through blended FFS and blended capitation, respectively. In three studies, we compared physicians in FHGs and FHOs in terms of their care provision for persons with diabetes mellitus (1st study), congestive heart failure (CHF) (2nd study) and chronic kidney disease (CKD) (3rd study).
Methods: All data were obtained from the ICES (formerly known as the Institute for Clinical Evaluative Sciences). For the first and second studies, we employed propensity score-based weights and fixed effects regressions on a balanced panel of physicians spanning 10 years; all analyses were conducted at the physician level. In these two studies, the comparison was between physicians in FHG who never switched to FHO or other models (i.e., non-switchers); switchers were physicians who switched from FHG to FHO. For the third study, we performed two cross-sectional analyses at the physician level; lack of data availability for patients with CKD over time deterred us from conducting longitudinal analyses as in the first two studies.
Results: We found that switching from FHG to FHO was associated with an improvement in some aspects of diabetes care. We found that CHF care—in terms of physicians’ follow-up of patients who are discharged—was not different between switchers and non-switchers. We found that some aspects of CKD care were better with physicians in FHG relative to their counterparts in FHO.
Conclusions: Compared to blended FFS, blended capitation payment is associated with a small but statistically significant improvement in some aspects of diabetes care. Our findings suggest that follow-up care for patients with CHF is similar in Ontario’s blended FFS and blended capitation models. Though we found that blended FFS is associated with greater adherence to some CKD process measures, future studies could employ longitudinal regressions to account for more confounding.
Summary for Lay Audience
In the early 2000s, the Canadian province of Ontario embarked on reforming its system of primary care; a major aim of the reform was to improve the quality of care that family physicians provide to their patients. To achieve this aim, the government partially relied on changing family physicians’ remuneration and introducing pay-for-performance (P4P) incentives. Prior to the reform, most of Ontario’s family physicians were paid through fee-for-service (FFS), a mode of remuneration where the unit of payment is the service. The post-reform primary care models are partly characterized by blended payment, which refers to a remuneration system based on multiple sources of income—of which one source predominates. For instance, in blended FFS, physicians are mainly remunerated through FFS and have secondary sources of income through bonuses, P4P incentives, and premiums. Family Health Group (FHG) and Family Health Organization (FHO) are two popular post-reform primary care models that pay physicians through blended FFS and blended capitation, respectively. In capitation, the unit of payment is a person. Over time, many physicians switched from FHG to FHO (i.e., switchers); however, some remained in FHG (i.e., non-switchers). The evidence on physician’s performance in FHGs and FHOs is limited; thus, we examined the impact of these two payment models on physicians’ provision of primary care services to Ontarians with diabetes mellitus, congestive heart failure (CHF), and chronic kidney disease (CKD). We also investigated patients’ health outcomes. We found that physicians’ switch from FHG to FHO was associated with moderately better care for some diabetes-related services; there was no significant difference between switchers and non-switchers in terms of the CHF care measures we examined. We also found that adherence to some CKD-related care processes was higher for physicians in FHGs relative to their counterparts in FHOs. Collectively, our studies provide some evidence to support that family physicians’ level of care can be associated with how they are paid.
Bamimore, Mary Aderayo, "Investigating quality of care for diabetes mellitus, congestive heart failure and chronic kidney disease in Ontario’s Family Health Group and Family Health Organization models" (2020). Electronic Thesis and Dissertation Repository. 7407.