The essence of primary care reform is the creation of a formal relationship between a physician provider and a patient. The idea is that a patient will sign on with a particular provider who will then take medical responsibility for accessibility, quality of care and continuity of care. However, various aspects of physician remuneration, specifically negation, are perceived as a disincentive for physicians. Negation refers to the deduction of a specified amount from the contracted physician if a patient sees another family physician or general practitioner not associated with their practice. How often and to what extent patients see different family physicians or general practitioners (FPs/GPs) becomes an important question in the debate. Scientists at ICES are examining several aspects of family medicine and general practice, including the use of physician claims from the Ontario Health Insurance Plan (OHIP), to learn how often patients see different primary care physicians. In one study, a random sample of 538 primary care physicians, having more than 500 OHIP claims in fiscal 1997/98, was examined. Only office-based visits were included in the study. This group of physicians saw 738,910 patients over 6,493,345 visits. In one year, about 31.0% of patients saw one FP/GP, 29.8% saw two FPs/GPs, 18.3% saw three, 10.0% saw four and 10.9% saw five or more FPs/GPs. When specialist consultation visits were included, the proportion of patients seeing only one physician dropped to 17.7% while the proportion of patients visiting more than five physicians a year grew to 26.0%.

Using OHIP consultation data, patients were "assigned" to various FP/GP practices using different assignment rules. For the "75% Rule," a patient was assigned to a primary care physician if at least 75% of his or her visits were to this primary care physician. The "Majority Rule" assigned a patient to a primary care physician if at least 50% of visits were to this primary care physician. The "Plurality Rule" assigned the patient to the primary care physician who accounted for the highest proportion of total visits made by the patient. In Table 1, the Majority Rule and Plurality Rule assigned the most patients to the randomly selected FP/GP physician (53.3% and 56.7% respectively), with the plurality rule also assigning more patients to other FPs/GPs (29.2%). The 75% Rule had the most "unassignable patients" (45.5%). The Plurality Rule had the highest proportion of random practice volume devoted to the assigned patients (82.1%). When the sample was restricted to patients having more than three visits (N=602,469), fewer patients were assigned to the random FP/GP for all assignment rules. However, the proportion of the random FP/GP practice volume devoted to assigned patients remained over 75% for both the majority and plurality rules.

These assignment rules are based on outpatient visit data and do not represent a formal agreement between a physician and a rostered patient. Visits made as part of a group practice versus those made in walk-in clinics are also difficult to differentiate with administrative claims data. With this in mind, the study found assigned patients saw fewer numbers of physicians than unassigned patients. For example, only 3.4% of the patients assigned using the Majority Rule saw more than five physicians in the year. Further work continues at ICES utilizing the National Population Health Survey to examine the characteristics of patients seeing multiple FPs/GPs. In addition, models examining the economic consequences of negation are also being conducted.

In summary, a significant proportion of the population (just over 10%) saw more than five GPs/FPs in one year. However, if a patient identifies with one physician provider, then they are less likely to see multiple physicians. In our study, which was based on a fee-for-service system, a patient identified with a physician provider by having the majority of their visits with that provider. Therefore, a formal process of patient rostering may reduce the number of different FPs/GPs a patient sees.

Table 1: Comparison of Patient Assignment Rules to Primary Care Physicians
Patient share assignment rule Number of patients assigned to a FPs/GPs Number of patients assigned to other FP/GP Number of unassignable patients % of random PCP's practice volume devoted to assigned patients
75% Rule 334,804 (45.3%) 67,987 (9.2) 336,119 (45.5%) 66.8%
Majority Rule 393,630 (53.3%) 50,742 (20.4%) 194,538 (26.3%) 77.9%
Plurality Rule 419,008 (56.7%) 215,856 (29.2%) 104,046 (14.1%) 82.1%

About the Author

Liisa Jaakkimainen is a family physician practising in Toronto. She is also a funded researcher in the Department of Family and Community Medicine at the University of Toronto and a scientist at ICES. Her research interests include primary care health services research, continuity of care and primary care chronic disease management.