An American Physician in London

Depending on what side of the Atlantic they are on, Americans and Brits may be familiar with one of two versions of suspicion of the other side's approach to healthcare: while Americans fear "socialized medicine", Brits tend to fear "privatisation of our NHS". The reality is that we have much to learn from each other as both countries seek to find a better way to pay for healthcare that is more focused on rewarding good outcomes rather than just productivity or staying within a budget. Ultimately, both countries are working to figure out how to provide high quality care efficiently, and to enable their residents to experience better health.

I moved to the UK three years ago with my family with the belief that I could bring my experience as an American physician and health systems consultant to the UK, and that my experience in the UK would help me think differently upon my return. Now back in the US, I have been reflecting on a patient I cared for early in my career as a physician in training at the Massachusetts General Hospital, in Boston, USA as a way of thinking through what the differences in our two healthcare systems would mean for a real individual with complex needs.

I admitted Mr. G, an immigrant from Guatemala with diabetes and a drinking problem, many times to the hospital. He was one of the first patients I treated for diabetic ketoacidosis, where sugar levels skyrocket and a deficiency of insulin leads to a dangerous acidification of the blood, dehydration, and other life-threatening complications. Once he was in the hospital, our team worked smoothly to start intravenous fluids, administer insulin, and closely monitor his laboratory results to help him recover during the dangerous first few hours of treatment. We also managed his alcohol withdrawal, another condition that can lead to hallucinations and seizures, and can be life-threatening. 

I was convinced we were saving his life and was thrilled when his mental status improved and his glucose was in better control. When I went to see him early on the third day in the hospital, happy to hear from his nurses that he was conscious and conversing, I cheerily started to ask questions, one Latin American to another, about why he had come to the US in the first place. He didn’t mirror my cheer. He was withdrawn and depressed. He didn’t have family in the area. He didn’t want to talk much about his home country or his past. He drank to forget. He sometimes could stay with people he knew, but sometimes didn’t, and he lost his medications. By the time we were discharging him from the hospital, I had worked with our team to give him information on shelters and innovative organizations like Healthcare for the Homeless in Boston. He took them and left. 

That was the first time I saw him. By the fourth time I admitted him, we treated him with as much care as the first time, but I was much less convinced we were “saving his life”. His nurses had seen him many more times between my rotations, and we knew that he had been to other hospitals in the area. We were merely treating the manifestation of his disease, without addressing any of the causes of the exacerbations such as homelessness, depression, alcohol use disorder, and likely post-traumatic stress. When he was acutely ill, the substantial costs of his care were borne by the hospital and taxpayers, but once he recovered, he returned to a life with very limited support that ensured his return to the hospital. 

His story, and the stories of many like him, drove me to explore opportunities to improve healthcare delivery systems. I spent three years as a McKinsey consultant based in Boston, and in 2016 I transferred to the London office in an effort to learn from the UK’s National Health Service, a taxpayer-funded public system of healthcare delivery that is free at point of care to all UK citizens and residents. On its surface, the UK has some enviable numbers to recommend it. Despite spending less than half, per capita, on healthcare than the US (US$3,958 in 2016 in the UK compared to US$9,870 in the US, equal to 9.8% of GDP in the UK vs. 17.1% in the US)[1], many outcomes are better: for example, “mortality amenable to healthcare”- a measure of avoidable deaths- is 85 per 100,000 population in the UK vs. 112/100,000 in the US, and the ability to get same day appointments while sick is 57% in the UK vs 51% in the US[2]. 

Interestingly, despite starting from different legacies, the English NHS and the US have been coming closer together in how they pay for healthcare. From a tradition of budget-based block grants, in 2003 NHS England responded to productivity problems by starting to move toward a system closer to fee-for-service in the acute care sector (an initiative called “Payment by Results”).[3] Meanwhile, the US struggled with the opposite problem- too much productivity and runaway healthcare costs. Building off of private sector innovation, the 2009 US Affordable Care Act promoted innovation that moves away from fee-for-service, enabling a variety of value-based contracts and entities like Accountable Care Organizations (ACOs). More recently, NHS England has been pushing toward Integrated Care Systems at a regional level, inspired in no small part by ACOs in the US.[4]

Reflecting on Mr. G, I find myself thinking about how his care could have been different in the UK. The unit prices for care provider time, drugs, and equipment would be lower, certainly. But there is no simple answer as to how he would have experienced care differently. Along the three aspects of care discussed below, there are some clear advantages of the UK system, but also significant opportunities to better capture the potential of those advantages. 

Accountability for health outcomes

Mr. G lived in Massachusetts, the state with the lowest percentage of uninsured in the country (at 2.5%, as compared to an 8.8% national average)[6], but he did not have insurance- likely because of his immigration status and because he was unable to follow through with paperwork for registration or application for refugee status. Each hospital bore the costs of his care during his many admissions and passed it on in the form of higher prices for paying patients, but there was not a clear single organization with the mandate or incentives to think about his care comprehensively.

In the UK, he would have had explicit access to care regardless of his immigration status[7], with care paid for by the local CCG in a publicly-paid and publicly-delivered system. However, in most systems that have not matured into Integrated Care Systems, each of the organizations would have had a mandate only over a small part of his care. While the CCG would have paid for his acute care, it would not pay for, or have data on, his life outside of the hospital. 

GP’s, who play a central primary care gatekeeper role for the NHS, could be in a position to take responsibility for his care. But GPs are paid an average of ￡151 / capita/yr[8] in the UK to provide all services, with a small bonus from Quality Outcomes Framework tied to pay-for-performance on largely process measures (e.g., measuring whether blood pressure was recorded in patients over 40)[9], instead of outcomes measures like patient experience, hospital use, mortality, and costs. The per-capita payment is less than the average national tariff for a single visit to Accident and Emergency.[10] There is little room to innovate beyond required services, and there are limited financial incentives to invest in interventions that will keep the GP’s patient panel out of hospital. New funding for multi-disciplinary teams is part of the long-term plan,[11] but it is not tied to performance on outcomes. GP’s are also structurally limited in their ability to take risks and innovate: GPs in the UK are private enterprises with unlimited liability for partners.[12] Individual GPs would not be able to make investments with payoffs conditional on system-wide performance without better protection from personal financial risk.

The push toward Integrated Care Systems, with shared financial incentives, shared data, and explicit collaboration mechanisms, will go a long way toward helping individuals like Mr. G better manage his health outside the hospital. The NHS Long Term Plan from 2018 promises to allow for system-wide budgets and facilitate longer-term investments.[11] A key caveat here: in the US it is common to worry about the need to have a financial buffer to take on actuarial risk- that is, the normal variation of health expenditures year-to-year based on chance. By giving acute trusts and small CCGs a budget that assumes a given volume of activity, NHS England may transfer significant actuarial risk to those institutions with limited financial buffers, leading to periodic financial shortfalls. Risk-bearing entities like CCGs and acute trusts can be bailed out by national funds- but consistent bailouts could lead to premature declaration of failure of the entire venture.

Social determinants of health

In the above discussion of accountability there is a significant clinical focus. But perhaps the most important drivers of Mr. G’s recurrent decompensations are social: homelessness, alcohol use, depression, post-traumatic stress, and a lack of social support. Services may vary more within the US and the UK than between them, and I cannot do justice to the many local and private efforts to address social determinants of health in both countries. 

In Mr. G’s case in Boston, he was unable to access Medicaid in the time I knew him due to access barriers - so he would have depended on largely charitable local initiatives. Had he managed to access Medicaid in the state of Massachusetts in its current incarnation, he could have benefited from the State’s push to care for all beneficiaries in partnership with an ACO[13]. That ACO, by taking on accountability for health outcomes and managing healthcare costs, would be able to justify making investments to address those underlying social factors that drove his recurrent hospital use. Both Partners Healthcare, the parent organization of MGH, and Boston Medical Center, routinely screen for social determinants of health and are exploring ways to address them, from improving housing and decreasing food insecurity, to affording medications.[14,15]

In the UK, much of social care is driven by local government, and would vary as much as it would in the US. At this point, with the political turmoil generated by Brexit, it is uncertain what direction the national government will take, at least in England. Prime Minister Boris Johnson’s confidently-presented plans for social care, for example, have been delayed in the turmoil of Brexit.[16]

My hope in this dimension is that in the future years Mr. G will receive better attention to these issues in either country than he would receive today.

Data-driven care

Had Mr. G been “attributed” to a payer or provider in the US with the ability and incentives to use his data to understand his very high risk of returning to the hospital, as well as drivers of that risk, he could have received services targeted to his needs. Many advanced payers, providers, and other service providers in the US have been using dynamic segmentation and targeting models to help prioritize both who receives services, and what services they could benefit from to most efficiently and effectively modify their risk.[17]

In the UK, despite having a right to services meeting his needs regardless of his ability to pay,[18] those scarce services are often not optimally targeted to Mr. G’s needs. The NHS has enviable data available to identify his high overall risk and his specific modifiable risks—data that includes nationwide encounter-level data on all hospital and specialty visits available for over two decades, national databases feeding planning and attribution for primary care, and electronic health records across almost all primary care—but that data is currently underused. Each attempt to use data has an extensive Information Governance review for that particular use, and any sharing between institutions (even within government institutions) can take months and significant red tape. 

New attempts to create system-wide data warehouses interacting with health information exchanges are a step in the right direction, but clearer regulatory coverage to support appropriate data sharing would help. The level of sophistication of analytics-based insights also has room to improve. Many regions use static “risk-stratification” models based on years-old data- and there are many opportunities to learn from international experience to leapfrog the costly process of full in-house development. 

The NHS is viewed by many with patriotic fervor for good reason- it delivers healthcare for all at very reasonable cost and without the pain of deductibles, out-of-network fees, copays, and complex and varied health insurance benefits. In the US now, I will bring with me a broader set of examples of approaches to healthcare, which will help me think more broadly about innovation here. In the setting of apparently increasing political traction for proposals like “Medicare-for-all”, the example of the NHS will be relevant to understand what it would take. I also know that the NHS in England can be even better—and that similarly learning from the rest of the world will help it get there. 

References:

1.          Current health expenditure per capita (current US$) | Data. Available at: https://data.worldbank.org/indicator/SH.XPD.CHEX.PC.CD. (Accessed: 13th June 2019)

2.          International Profiles of Health Care Systems | Commonwealth Fund. Available at: https://www.commonwealthfund.org/publications/fund-reports/2017/may/international-profiles-health-care-systems. (Accessed: 13th June 2019)

3.          Payment by Results in the NHS: a simple guide. GOV.UK Available at: https://www.gov.uk/government/publications/simple-guide-to-payment-by-results. (Accessed: 13th June 2019)

4.          NHS England ? Integrated care systems. Available at: https://www.england.nhs.uk/integratedcare/integrated-care-systems/. (Accessed: 14th June 2019)

5.          Leadership in today’s NHS. The King’s Fund (2018). Available at: https://www.kingsfund.org.uk/publications/leadership-todays-nhs. (Accessed: 13th June 2019)

6.          Bureau, U. C. Health Insurance Coverage in the United States: 2017. Available at: https://www.census.gov/library/publications/2018/demo/p60-264.html. (Accessed: 20th June 2019)

7.          NHS entitlements: migrant health guide. GOV.UK Available at: https://www.gov.uk/guidance/nhs-entitlements-migrant-health-guide. (Accessed: 20th June 2019)

8.          NHS Payments to General Practice, England, 2016/17 - NHS Digital. Available at: https://digital.nhs.uk/data-and-information/publications/statistical/nhs-payments-to-general-practice/nhs-payments-to-general-practice-england-2016-17. (Accessed: 6th June 2019)

9.          Quality Outcomes Framework (QOF) - NHS Digital. Available at: https://digital.nhs.uk/data-and-information/data-tools-and-services/data-services/general-practice-data-hub/quality-outcomes-framework-qof#more-information. (Accessed: 6th June 2019)

10.        National tariff payment system | NHS Improvement. Available at: https://improvement.nhs.uk/resources/national-tariff/. (Accessed: 6th June 2019)

11.        Plan, N. L. T. Online version of the NHS Long Term Plan. NHS Long Term Plan Available at: https://www.longtermplan.nhs.uk/online-version/. (Accessed: 6th June 2019)

12.        Iacobucci, G. New business models would boost GP partnerships, says review. BMJ 364, l222 (2019).

13.        MassHealth Partners with 18 Health Care Organizations to Improve Health Care Outcomes for Members. Mass.gov Available at: https://www.mass.gov/news/masshealth-partners-with-18-health-care-organizations-to-improve-health-care-outcomes-for. (Accessed: 22nd September 2019)

14.        Partners Population Health Management. Partners Population Health Available at: https://populationhealth.partners.org/care-redesign-programs/beneficiary-care/medicaid/. (Accessed: 22nd September 2019)

15.        Upstream Healthcare Delivery. Boston Medical Center (2018). Available at: https://www.bmc.org/mission-in-action/upstream-healthcare-delivery. (Accessed: 22nd September 2019)

16.        Boris Johnson’s ‘ready to implement’ social care plan now delayed. Available at: https://inews.co.uk/news/politics/boris-johnson-social-care-overhaul-495176. (Accessed: 22nd September 2019)

17.        HealthITAnalytics. 10 High-Value Use Cases for Predictive Analytics in Healthcare. HealthITAnalytics (2018). Available at: https://healthitanalytics.com/news/10-high-value-use-cases-for-predictive-analytics-in-healthcare. (Accessed: 22nd September 2019)

18.        Principles and values that guide the NHS. nhs.uk (2018). Available at: https://www.nhs.uk/using-the-nhs/about-the-nhs/principles-and-values/. (Accessed: 6th June 2019)

19.        A Brief Look at Commercial Health Insurance Market Share in Select New York Metro Areas. Mark Farrah Associates (2018). Available at: https://www.markfarrah.com/mfa-briefs/a-brief-look-at-commercial-health-insurance-market-share-in-select-new-york-metro-areas/. (Accessed: 20th July 2019)