Enhanced fee-for-service model and physician productivity: Evidence from Family Health Groups in Ontario

https://doi.org/10.1016/j.jhealeco.2010.10.005Get rights and content

Abstract

We study an enhanced fee-for-service model for primary care physicians in the Family Health Groups (FHG) in Ontario, Canada. In contrast to the traditional fee-for-service (FFS) model, the FHG model includes targeted fee increases, extended hours, performance-based initiatives, and patient enrolment. Using a long panel of claims data, we find that the FHG model significantly increases physician productivity relative to the FFS model, as measured by the number of services, patient visits, and distinct patients seen. We also find that the FHG physicians have lower referral rates and treat slightly more complex patients than the comparable FFS physicians. These results suggest that the FHG model offers a promising alternative to the FFS model for increasing physician productivity.

Research highlights

▶ We study Family Health Groups, an enhanced FFS model for primary care physicians in Ontario. ▶ Our sample includes a 17-year panel of claims data before and after FHG model was introduced in 2003. ▶ We use matching and difference-in-differences with fixed effects and linear trends. ▶ We find that FHG doctors provide more services and visits and see more patients than FFS doctors. ▶ FHG doctors also have fewer referrals and treat more complex patients that FFS doctors.

Introduction

Understanding how primary care physicians respond to payment incentives has been an important policy question for decades. The early literature has focused in large part on how three main methods of payment – salary, fee-for-service, and capitation – influence physician behaviour1. These traditional methods of payment have been recently reformed in many countries to include incentives for desired healthcare outcomes, such as reaching preventive care targets, improving chronic disease management, and attaching patients with no family doctor2. Despite the emerging empirical literature, however, it is still largely unknown how physicians respond to payment incentives in these new payment models3.

In this paper, we provide new empirical evidence on this question by studying a primary care model known as the Family Health Group (FHG) that was introduced in Ontario, Canada in 2003. The FHG model is an enhanced fee-for-service (FFS) model that includes payment incentives for improving patient access and quality of care, such as premiums for extended hours, bonuses for chronic disease management, and incentives for patient enrolment. We study the impact of joining the FHG relative to the traditional FFS model on three measures of physician productivity: the number of clinical services, visits, and distinct patients seen. Our analysis is based on claims data for almost all primary care physicians in Ontario for eleven years before and five years after the FHG model was introduced.

Our study contributes to the emerging literature on how physicians respond to new payment incentives in several ways. First, we study a primary care model that is based on the targeted healthcare outcomes that are at the front and centre of recent primary care reforms in many countries, such as chronic disease management, enhanced access, and comprehensive care. In addition, we develop a stylized economic model of physician behaviour in the FHG model. This model is useful as a framework for understanding how physicians respond to the FHG incentives and as a guide for our empirical analysis. The model can also serve as a starting point to study incentive structures in other jurisdictions. Lastly, we use an empirical methodology that can be fruitfully exploited to evaluate how physicians respond to payment incentives when only observational data is available. Specifically, we use the propensity score matching to select control groups of FFS physicians and we use the difference-in-difference model with fixed physician effects and linear trends to evaluate the FHG impact. We also explore multiple ‘experiments’ and dynamics of the FHG impact to further validate the interpretation of changes in physician behaviour.

We find that joining the FHG model has a meaningful impact on physician productivity relative to the traditional FFS model, as measured by the number of services, visits, and patients. The estimated productivity gain is about six to ten percent, equivalent to about two to three additional weeks of work per year. Furthermore, the impact occurs within the first year of joining the FHG model and persists over time. The impact is also stable across physician groups defined by age, sex, and location of practice. We also find that FHG physicians have significantly lower referral rates to specialists and treat slightly more complex patients than the comparable FFS physicians. These results suggest that the payment incentives in the FHG model significantly improve physician productivity relative to the traditional FFS model.

The rest of the paper is organized as follows. The next section provides a brief introduction to the primary care models in Ontario, including a detailed comparison between the traditional FFS model and the FHG model. Section 3 presents a theoretical analysis of the decision of FFS physicians to join the FHG model and the impact of this decision on their practice profile. Section 4 describes our data and empirical strategy. Section 5 discusses the results and Section 6 concludes.

Section snippets

Institutional background

Primary care physicians in Ontario participate in a wide spectrum of patient enrolment models (PEM). These models were introduced in a recent primary care reform that aimed to provide alternatives to the traditional FFS model. The reform dramatically changed how primary care is provided in Ontario. Between 2002 and 2008, the percent of primary care physicians participating in the PEMs increased from less than 5 percent (400 physicians) to over 70 percent (8000 physicians)4

Economic model of Family Health Groups

To analyze how differences between the FFS and FHG models affect physician behaviour, we utilize a stylized labour supply model that distinguishes between regular hours of work and after hours. We then use this model to analyze the decision of FFS physicians to join the FHG model and the impact of this decision on their practice profile.

Data

The data comes from the Ontario Health Insurance Plan (OHIP) fee-for-service claims for the fiscal years 1992–2008. This period includes eleven years before and five years after the FHG model was introduced in 2003.

The OHIP data has several advantages for our analysis. It includes virtually all family physicians in Ontario who are potentially affected by the introduction of the FHG model13

Initial estimates

Our initial estimates of the FHG impact are presented in Table 4. These estimates are based on the sample that excludes years with annual billings below C$30,000, a common income threshold used to identify physicians with the minimum attachment to the labour force. For comparison, we present results from the OLS model, the fixed effects model, and the correlated random trend model. These models progressively add more physician-specific effects: the OLS model includes two observed fixed effects:

Conclusion

In this paper, we compare productivity of primary care physicians in a new payment model introduced in Ontario known as the Family Health Group to the traditional fee-for-service model. Our results indicate that the FHG physicians provide more services and visits and treat more patients than the comparable FFS physicians.

We believe that these results are important for at least two reasons. First, the results show that how physicians are paid can affect their total productivity. This insight is

Acknowledgements

We thank the editor, two anonymous referees, and seminar participants at York University in Toronto, Canada, McMaster University in Hamilton, Canada, and the 2010 Annual Meeting of the Canadian Health Economics Study Group in Montreal, Canada for useful comments. As usual, all errors are ours.

References (19)

  • R.A. Devlin et al.

    Do physician remuneration schemes matter? The case of Canadian family physicians

    Journal of Health Economics

    (2008)
  • T. McGuire et al.

    Physician response to fee changes with multiple payers

    Journal of Health Economics

    (1991)
  • S. Campbell et al.

    Quality of primary care in England with the introduction of pay for performance

    New England Journal of Medicine

    (2007)
  • Crossley, T.F., Hurley, J., Jeon, S., 2006. Physician labour supply in Canada: a cohort analysis. McMaster University....
  • R.H. Dehejia et al.

    Propensity-score matching methods for nonexperimental causal studies

    The Review of Economics and Statistics

    (2002)
  • R.H. Glazier et al.

    Capitation and enhanced fee-for-service models for primary care reform: a population-based evaluation

    Canadian Medical Association Journal

    (2009)
  • J. Li et al.

    Physician response to pay-for-performance—evidence from a natural experiment

  • Leuven, E., Sianesi, B., 2003. PSMATCH2: Stata module to perform full Mahalanobis and propensity score matching, common...
  • T. McGuire
    (2000)
There are more references available in the full text version of this article.

Cited by (44)

  • Team-based primary care practice and physician's services: Evidence from Family Health Teams in Ontario, Canada

    2020, Social Science and Medicine
    Citation Excerpt :

    Among those who switched to FHT, 157 physicians reverted back to a non-FHT model, and were excluded from our analysis (inclusion of these observations in our analysis did not change the results). We further restricted our sample to those physicians with enrolled patients in each year and working in a group of three or more physicians with an annual gross income above $30,000 – a threshold used to identify physicians with a minimal attachment to the labour force (Kantarevic et al., 2011). Our final sample contains 5200 physicians, 2655 FHTs and 2545 non-FHTs, over a period of 10 years (Appendix Table A1.3 presents the sample selection flowchart).

  • The impact of team-based primary care on health care services utilization and costs: Quebec's family medicine groups

    2017, Journal of Health Economics
    Citation Excerpt :

    Principal-agent models provide clear predictions on the influence of payment schemes on physician behavior (McGuire 2011): fee-for-service incentivizes higher volume of services, capitation may lead to stinting and over-referral to specialists and hospitals, and pay-for-performance may result in increased use of targeted services. Numerous empirical studies have confirmed these predictions and have shown that physicians respond to payment incentives, including in the primary care context (Gaynor and Pauly 1990; Kantarevic et al., 2011). However, since both the mechanism and levels of payment are the same for physicians inside and outside FMGs (fee-for-service payment in accordance with the provincial fee schedule), we do not expect this channel to be relevant in affecting changes in utilization or costs under Quebec’s primary care reform.

View all citing articles on Scopus
View full text