Editor’s Spotlight/Take 5: Are Income-based Differences in TKA Use and Outcomes Reduced in a Single-payer System? A Large-database Comparison of the United States and Canada

“Medicare for all.” It’s a tempting thought, one that’s been discussed in CORR [2], and one that has moved beyond bumper stickers. United States Senator Bernie Sanders recently introduced S. 4204, “A bill to establish a Medicare-for-all national health insurance program,” with 14 cosponsors. If nothing else, the proposed legislation would mark the end of the United States’ position as the only OECD country without some form of universal health coverage. Why not move forward, so that every individual can be as healthy as possible?Bella Mehta MBBS, MDAs it turns out, simply providing insurance isn’t enough. In a fascinating study in this month’s Clinical Orthopaedics and Related Research® [13], lead author Bella Mehta MBBS, MD and her colleagues compared TKA usage by patients across different income brackets in Ontario, Canada (where patients have universal coverage) to those who had the procedure performed in the United States, specifically Pennsylvania. As the authors point out, the relative difference in TKA use between the highest-income and lowest-income quintile was larger in Ontario. Stated another way, the disparity in usage of this elective procedure between patients of high and low income levels was greater in Canada than it was in the United States, contrary to my expectation. In its recent study, “Key Health Inequalities in Canada: A National Portrait,” the Public Health Agency of Canada (Agence de la Santé Publique du Canada, if you’re a francophone) found that Canadians with less than a high school education live 11 fewer healthy years than do university graduates, that Canadians in the lowest income group live 11 fewer healthy years than those in the highest income groups, and that Canadians living in areas of both high social and high material deprivation have a life expectancy 6 years shorter than those living in areas with low social and material deprivation [10]. Lest Americans feel any schadenfreude from this, the situation in the United States is no better, and in some ways worse [17, 18]. These findings convey some important messages, perhaps none more than this: Healthcare disparities related to patients’ resource limitations or levels of educational attainment are not unique to the United States or Canada, and they’re not limited to elective orthopaedic care. In short, whether we’re looking at TKAs, as Mehta et al. have done here, or various other measures of health, the story is the same—socioeconomic status is a major influence on health. And this is true in countries with every form of universal health insurance. Access to healthcare is not sufficient. A recent study from the United Kingdom looked at 4 million individuals in the Clinical Practice Research Datalink and found that socioeconomically deprived individuals were substantially more likely to develop heart failure than were affluent individuals, did so earlier in life than those from the most affluent group, and had more comorbidities, despite their younger age [3]. Furthermore, the disparity got worse from 2002 to 2014. Similarly, Whitehall II, a longitudinal study of British civil servants, found that lower socioeconomic status increased the risk of multimorbidity, frailty, and disability [6]. And disparities occur in countries with relatively homogeneous populations. Surveys of health in Germany since 2000 show persistent and increasing differences, based on socioeconomic status, in the prevalence of chronic illness among men and women [11]. Even in Japan, researchers find differences in health and life expectancy by prefecture [14]. In both countries, the differences are worsening over time. Although having insurance is clearly an advantage in terms of avoiding economic hardship and finding care when needed, it’s not enough. The “Oregon Experiment,” in which Oregon expanded Medicaid based on lottery drawings from a waiting list, generated no substantial improvements in measured physical health outcomes [1]. Subsequent similar studies found that access and utilization are better with insurance, but there was little to show in terms of measurable improvement in health [12, 16]. In short, as our Spotlight article demonstrates, addressing disparities in healthcare may require substantially more than the “simple” answer of “Medicare for all.” Join me now as I talk with Dr. Mehta, the lead author of “Are Income-based Differences in TKA Use and Outcomes Reduced in a Single-payer System? A Large-database Comparison of the United States and Canada” in the Take 5 interview that follows. Take 5 Interview with Bella Mehta MBBS, MD, lead author of “Are Income-based Differences in TKA Use and Outcomes Reduced in a Single-payer System? A Large-database Comparison of the United States and Canada” Paul A. Manner MD:There’s an assumption that universal health insurance helps reduce or eliminate inequality in healthcare. In your study, that doesn’t seem to be the case. Assuming your study is not some kind of outlier, how do your findings reflect a larger picture of health inequality in Canada? Bella Mehta MBBS, MD: Our findings indicate that Canada’s universal health insurance system did not eliminate income-based disparities for TKA. This was contrary to our hypothesis when we started the study. There is an assumption that Canada’s universal health insurance will create more equitable access across income strata, but the number of empirical studies is really quite limited. In a 2021 study, Dr. Peter Cram and his team found larger income-based differences in cancer surgery rates in the US (and Australia) than in Canada [15], but we are unaware of prior studies that have looked at joint replacement. Studies have also looked at US-Canada disparities in screening, care, and survival in cancer outcomes, demonstrating smaller wealth-based disparities in Canada [7-9]. In considering our results, it is important to recognize that mortality is different from utilization and joint replacement is different from cancer surgery. In considering our results, it is also important to recognize that our study focused on one Canadian province (Ontario), one state (Pennsylvania), and one procedure (TKA). Further studies are needed to determine whether the results are consistent across other procedures such as THA and in different geographic regions. Dr. Manner:Many countries have universal health insurance but use a combination of public and private systems. How does Canada compare to these countries in terms of equality of healthcare? Dr. Mehta: Canada follows a so-called “national health insurance” model, also sometimes referred to as “socialized medicine.” This differs from the Bismark or Beveridge models used in other countries [5]. With the Beveridge model used in United Kingdom, healthcare is provided and financed by the government through tax payments, just like the police force or the public library. Most hospitals and clinics are government owned, and the ones that are private still bill the government. But in the Bismark model, which is used in Germany, where healthcare is based on an insurance system, the insurers are called “sickness funds” and are usually financed jointly by employers and employees through payroll deductions. In Canada, most medical care is free at the point of care. Canada relies on providers who are technically in private practice, but payment comes from a government-run insurance program funded through general tax revenue. Thus, it tends to be cheaper and much simpler administratively than the American-style for-profit insurance. National Health Insurance plans also control costs by limiting the medical services they will pay for, or by making patients wait to be treated. The United States, however, has both public and private insurance plans. The US insurance system differs substantially from Canada and commonly has some degree of cost-sharing. In a previous study comparing these healthcare systems [4], Canada was found to have higher total joint replacement utilization, lower readmission rates, and higher mortality than the United States. However, we are not aware of studies specifically looking into the transnational differences regarding income strata in joint replacement. I believe that the impact of social determinants and healthcare system factors on TKA use and outcomes should be considered by surgeons when planning surgery. Dr. Manner:In the United States, the state of Oregon ran what was essentially a lottery for Medicaid enrollment, which became known as the “Oregon Experiment.” Participants did not achieve better health in terms of defined measures than those who did not enroll. How does your study help us learn what determinants we are missing, since insurance status doesn’t adequately explain differences in outcomes? Dr. Mehta: The Oregon studies (along with the RAND health insurance experiment) are some of the few experimental studies which are testing “what is the impact of having health insurance?” While these were landmark studies, it is extremely difficult to conduct large, randomized trials of health insurance for a combination of ethical, economic, and pragmatic reasons. It is unlikely that studies such as Oregon or RAND will be replicated again in the near future, leaving us with limited opportunities to truly test different types of insurance coverage at a population level [5]. International comparisons offer another avenue for such research since it is really hard to conduct randomized trials of health insurance. I would interpret our results pretty modestly here. Overall, I believe that improving access for patients from low socioeconomic strata would help. Dr. Manner:One consistent finding in your study is lower rates of surgery in Canada versus the United States across all quintiles. Why are we seeing this? How might differences in practice between countries or availability of resources contribute to this? Dr. Mehta: Previous studies [8, 9] have demonstrated problems with access to care in single-payer health systems like Canada, likely because of factors like limited hospital capacity and limited numbers of orthopaedic surgeons resulting in long wait times, especially for elective nonemergent procedures like TKA. Although TKA is an effective procedure, it is costly, and single-payer systems typically restrict access. It is also unknown whether residents of low- and high-income neighborhoods have different needs for and severity of osteoarthritis, and whether higher healthcare usage in the United States may actually reflect an overuse of those services. While our study focused on the differences between the two countries, several cultural-, individual-, and surgeon-level factors also influence the decision for surgery that a large-scale study like ours cannot address. Dr. Manner:The two areas you studied are very different in terms of population density; Pennsylvania has about 300 people per square mile, Ontario about 30. (Admittedly, the northern portion of Ontario is almost uninhabited). Are the lower rates of surgery in part a function of proximity to surgeons? Is there a similar problem in Pennsylvania? Dr. Mehta: Yes. Ontario had lower overall rates of TKA, but we observed no reduction in income-based disparities in either utilization or outcomes. Let’s try and unpack this a bit. The lower rates of TKA in Ontario are a bit surprising, since a prior study by Cram et al. [4], which compared Ontario with New York, actually showed higher utilization in Ontario. We suspect that what we are seeing is that Pennsylvania has higher utilization of TKA than Ontario, and Ontario has higher utilization than New York. The lesson here is that your comparators really matter. Regardless, it is really important to understand payment policy for TKA in both the United States and Canada. In the United States, even with the evolution toward “value-based purchasing,” hospitals and physicians have incentive to perform TKA. Indeed, even with prior authorization and other efforts to restrain TKA utilization, the reality is that in the United States, it is relatively easy to perform a TKA from an insurance standpoint. In Canada, things are really different. Canadian healthcare is organized and financed at the level of the province; this is important since one province (Ontario, for example) could decide to budget more generously for musculoskeletal services while another (say, Quebec) could devote more resources to mental health. Access differs widely by province. Generally, all provinces have fixed annual budgets, and a substantial part of provincial healthcare planning in Canada is devoted to balancing (and restraining) the use of costly discretionary procedures. Nobody in Canada argues that TKA is a great procedure, but TKA is expensive and provincial governments need to make hard decisions. It is also important to understand how orthopaedic surgeons and hospitals are paid in Canada. Most Canadian physicians—exclusive of primary care—are reimbursed on a fee-for-service basis just like in the United States. Thus, the more care a doctor provides, the more money that doctor gets paid. Orthopaedic surgeons get paid for each TKA they perform. However, Canadian hospitals are globally budgeted, which is much different from US hospitals. A typical Canadian hospital gets allocated a fixed sum of money from the provincial government at the start of each fiscal year; the precise amount each hospital gets is largely based on historical precedent (how much the hospital received the prior year). Hospitals typically do not get additional money for the additional care that they provide. What does this mean? If a hospital is budgeted to provide 500 TKAs, but instead decides to perform 700, the hospital cannot expect incremental money for the excess 200 TKAs performed. As a result, hospitals in Canada have a strong incentive to limit volumes for discretionary services like TKA. Therefore, we would guess that the lower rates of TKA in Ontario reflect differences in funding and policy related to TKA between the two countries.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.