“I tried to find another doctor, but as soon as they heard I was on ODSP (Ontario Disability Support Program), they didn’t want to see me.”
I saw this man regularly early in my career, when I was working at a primary care clinic serving people who were unhoused and he was living in the shelter system. When he was housed far from the city’s core and the clinic, we both hoped he would be able to transition his primary care closer to home, but here he was again. His frustration was evident at what was so blatantly a denial of care based on his poverty and disability.
Yet, this kind of inequity is baked into our current primary care system in Ontario. I think about his words often, and was reminded of them recently when the Ministry of Health and Long Term Care in Ontario ratified a new Physician Services Agreement with the Ontario Medical Association. The agreement includes a plan to introduce a complexity modifier to the capitation payment system for primary care physicians and may help address some of the ongoing structural inequities in access to primary care. This change has the potential to make a big difference, but the devil is in the details.
This change has the potential to make a big difference, but the devil is in the details.
In more than a decade of serving marginalized communities, I have seen that the system often is poorly set up to meet the needs of those at the margins, whether they be Black, Indigenous or racialized; queer, trans and non-binary; people who are unhoused; people who use drugs; people with severe mental illness; sex workers; or migrants with a range of immigration situations. We know that societal structures of systemic racism, sexism, classism, homophobia, transphobia, etc., create health inequities in both outcomes and access. Those who are marginalized and minoritized are the most likely to be unattached to primary care, yet often need it most, and therefore are more likely to present with complications or preventable conditions in our emergency departments. Many marginalized individuals have experienced trauma and have good reason not to trust people in positions of authority, including and sometimes especially physicians.
There are many factors involved in building anti-racist, anti-oppressive, culturally safe primary care systems that serve marginalized communities well, but certainly one of the biggest is adequate time in both length and frequency of visits. Time allows us to build trust and relationships; to sort through unaddressed physical and mental health issues; to help navigate and access income and other social supports; and to help people engage in preventative care that may be long overdue. This form of primary care can mean that a woman with sexual trauma feels safe getting a pap smear after 10 years with her provider, preventing cervical cancer; or a young man with few social supports getting diagnosed and treated for schizophrenia, improving his function and quality of life. We need a system that incentivizes serving those who most need it and doing it well to mitigate ongoing inequities, but existing incentive structures often do the opposite.
Before the early 2000s, Ontario’s primary care physicians were primarily paid through a fee-for-service model. Unfortunately, this model poses challenges for those with complex needs due to incentives to see patients for short visits, addressing one issue at a time and prioritizing procedural care over other types. For many patients, this model poses barriers to adequately expressing themselves, building trust and discussing multiple issues to meet their needs.
Community Health Centres (CHCs) and certain clinics funded through an Alternate Payment Plan (APP) have payment structures geared toward serving marginalized communities, with physicians compensated on a salaried or hourly basis. These models also allow for innovative services such as funding for co-located harm reduction, health promotion and case management services, or outreach models where primary care co-locates within existing social service agencies. These factors, along with the payment structure, create the conditions for low-barrier access and proper care for marginalized groups with complex needs, but we need the mainstream health-care system to be able to do the same.
Many primary care providers in this payment model are facing structural barriers, and marginalized communities are paying the price.
Since the 2000s, Ontario has made a large shift away from fee-for-service as the primary mode of payment for primary care physicians toward patient enrollment models such as Family Health Organizations (FHOs) and Family Health Groups (FHGs). In FHOs, physicians are paid primarily through capitation, receiving an annual payment for each individual rostered to their care, based on age and sex. This means a physician is paid the same annually to care for a 45-year-old recently immigrated Bangladeshi single mother living in social housing on ODSP with poorly controlled diabetes and chronic pain as a 45-year-old Canadian-born white woman working as a lawyer with no dependents or chronic medical issues. One of these individuals will require monthly long appointments upfront to address her multiple needs using an interpreter and the other comes in every three years for her pap smears. Rather than incentivizing rostering the individual who needs care more urgently due to the complexity of her medical and social needs from an equity lens, capitation incentivizes rostering the individual with fewer needs, thus freeing up time to roster more patients and thereby be paid more overall. Working now primarily in a capitation-based model, I am very aware that intentionally choosing to roster people with more complex needs will lead to lower compensation overall. Many primary care providers in such a payment model are facing the same structural barriers, and marginalized communities are paying the price.
The new Physician Services Agreement plans to bring in a complexity modifier to capitation payment based on CIHI’s Population Grouping Methodology, currently being used by health-systems planners and policymakers to understand health conditions and costs associated. The methodology uses hospital and physician billing data to assign a complexity category to individuals based on diagnostic codes during various health-care encounters. This categorization would then be added as a factor beyond age and sex to determine the annual payment provided to a primary care physician based on anticipated care needs. Based on its implementation, this can be a significant improvement over the current method, but there are some important considerations worth noting.
Firstly, this methodology utilizes data from health-care usage but may not capture well the needs of individuals who may avoid seeking care. Sometimes, those with the most complex needs are the hardest to engage with, potentially due to negative treatment, mistrust or mental illness. Someone with undiagnosed serious medical conditions who avoids seeking care may be misclassified by this methodology as having low needs. Ideally, we need a system that captures complexity despite low usage and incentivizes continuing to attempt to connect with these individuals, which often can take additional effort on the part of physicians and team members.
Secondly, while billing data will capture the number of visits that individuals may engage in to determine complexity, it will not capture the time taken per visit or number of issues presented, which can vary widely for a visit coded similarly. Notably, there is also data to suggest female primary care physicians spend more time with their patients, contributing to a pay inequity which would not be taken into account.
As well, a complexity modifier must account for social factors. In the example given in CIHI’s materials, a 57-year-old male with hypertension, diabetes and depression is presented, suggesting that CIHI’s methodology can estimate the health-care usage of someone with this profile. However, individuals with the exact age and sex demographics, as well as medical conditions, can have significantly different complexity of needs and poorer outcomes if they are experiencing homelessness, are new immigrants, are unemployed without access to drug coverage, or are Black, Indigenous or racialized. Additional factors such as postal code data is important as it can provide a crude proxy for income levels, but many other social factors are not captured consistently in health care.
Some medical diagnostic codes can also provide clues to social complexity. The weighting of different disease entities will matter significantly in how complexity is calculated, and medical conditions that have a stronger association with social complexity, such as mental illness and addictions, should be prioritized in terms of payment to incentivize serving those who are most marginalized from an equity lens. While there are, in fact, some OHIP diagnostic codes pertaining to social issues (e.g., economic and/or educational problems, unemployment, legal problems), these are likely not well known nor widely used.
Billing codes are also geared toward procedures and often do not capture interventions commonly provided for marginalized individuals, such as completion of certain forms, writing letters of advocacy or collaborative care planning with social service agencies. Often, much of this administrative work falls outside the actual clinical visit.
Incentivizing physicians for serving those who most need care can help mitigate historic inequities and address ongoing health gaps
Finally, regardless of how the complexity modifier is rolled out – even if it most perfectly captures the usage and complexity of the individuals served – structural barriers beyond payment systems as well as systems of oppression and discrimination will still impact many communities requiring attention from our health-care institutions. Alongside the work being done to appropriately incentivize serving marginalized individuals, we need to continue to shift the culture and tangible supports around addressing social issues in primary care.
Adding a complexity modifier to capitation-based primary care models in Ontario can be an important step toward equity with some of the above considerations in mind. The stakes are high and appropriately incentivizing and rewarding physicians for serving those who most need care can help mitigate historic inequities and address ongoing health gaps facing marginalized communities in Ontario today. I can only hope that such a change would mean that care for people who most need it will be one step easier in the future.
The author would like to acknowledge Drs. Katie Dorman, Rick Glazier, Samantha Green, Tara Kiran and Eileen Nicolle, all of whom helped inform her conceptualization of this article.