tech prescribed

This past Spring, I had the privilege of taking a Kellogg MBA course with Dr. James Weinstein, former CEO of the innovative Dartmouth-Hitchcock health system. Throughout this course, professor Weinstein guided us through often contentious discussions focused on improving healthcare delivery in the US.

During our second meeting, the discussion turned to Social Determinants of Health. In a class with students from all parts of the political spectrum, the debate spanned a canyon of arguments on how we might address these issues. While I was familiar with the buzzword, I quickly realized my knowledge on the subject was shallow, inspiring the research leading to this post. Here I will attempt to provide an overview of social determinants, the impact they have on health outcomes, how we have attempted to address them in both the public and private spheres, and the role technology can play going forward. For me, it all started with this chart:

Bradley EH, Elkins BR, Herrin J, et al, Health and social services expenditures: associations with health outcomes; BMJ Qual Saf 2011;20:826-831

As attention to our healthcare system has grown, we have all heard the complaints about medical costs reaching an astronomical, unsustainable 18% of GDP. Other countries have figured out how to keep costs low while delivering better outcomes, why can’t we? There are inevitably a large number of differences between the US and our international counterparts (universal coverage, free medical education, price differences) that we should absolutely explore, but this chart is an indication of one area where the US has become a significant outlier. If we assume that our end goal is to improve the health outcomes of our population (as measured by all out mortality), then social programs may have a much more significant impact in the long run than health services.

“As a country, we have made the decision, implicitly or explicitly, to value the best possible medical care for those that can afford it over baseline social care for those who cannot.“

Courtesy KFF

In this research done by the Kaiser Family Foundation, “Health Care” has only a 10% impact on the risk of premature death while “Social and Environmental Factors” have a 20% impact (double?!?). Yet, in the US, our spending on Health Services is 1.2x our spending on Social Services (conceding there may be some categorization difference between the KFF research and the original chart above). There is also a clear difference between the populations that benefit from healthcare spending (wealthy, fully-insured) vs. social spending (low income, underinsured). As a country, we have made the decision, implicitly or explicitly, to value the best possible medical care for those that can afford it over baseline social care for those who cannot. Avoiding the politics of whether this is an ethical approach, there are clear benefits for all populations to shifting this proportion of spending. But first - let’s define the range of “Social Determinants” that could be addressed by dedicated programs.

Overview of SDoH Categories

While different sources may have slight variations, I found the definitions provided by KFF to be the most comprehensive representation of the different social influences on health status.

Courtesy KFF

Ultimately, these unmet needs have adverse impacts on our health system. Here are a couple key examples:

People who live in higher crime areas have higher BMI and higher levels of obesity due to lack of physical activity
Children exposed to violent crime are at risk for mental and behavioral health issues
WHO contributed 11% of US mortality to environmental conditions in 2012
In 2016, 31.6% of low income households were food insecure
21.3M households were cost burdened in 2014, but only 26% of eligible households received federal assistance
And much more...

It doesn’t sound like the US is quite the utopia we all hoped. While each of these categories impacts groups differently, the outcomes driven by unmet social needs are clear: 2x rates of depression, 60% higher prevalence of diabetes, 50% higher cholesterol, 2x rate of ER visits, and 2x rate of clinic no shows. These are some of the most significant drivers of cost in the system as a whole. (Data courtesy: HealthyPeople2020, KFF, Deloitte)

How much does it cost us? Hard to tell. While multiple studies have tried to quantify the amount of money spent as a result of a lack of social needs funding, many have struggled to come up with a finite number given the complexity of attribution. However, it is safe to say that this number is LARGE - potentially as much as $1.7T according to Healthcare IT News. When considering whether or not to invest in programs to address SDoH, we will need to conduct specific research to outline costs addressed by each program. But there is no doubt the opportunity is there. For example, in a recent partnership between USF and Wellcare, the organization was able to achieve a 10% ($2400 PMPY) reduction in costs from routing Patients to social programs (Healthcare Finance News). Many other examples will be cited later in this post.

Trends & Challenges to Addressing SDoH

The US healthcare system has been pursuing new models of care for decades, focused on achieving the triple aim: improving health, improving quality, and lowering costs. While the transformation has occurred slowly, we are reaching an inflection point (maybe?) where pure fee for service payment is in a death spiral. In a Change Healthcare 2018 report, only 37.2% of 2017 reimbursement was provided through pure FFS arrangements. Instead, a variety of value based models have taken hold, designed to align provider incentives and drive the costs of care downward. Additionally, these same Providers are also facing increased penalties from missing key quality measures like hospital readmissions.

Core to success in value based programs is the ability to risk stratify a population and identify the individuals that are the highest utilizers of unnecessary or improper care. Individuals that suffer from lack of social support often fall into this category, and hospitals are starting to catch on. Having two or more value based programs in place at a hospital is a strong indicator that a SDoH strategy will be implemented. (Deloitte)

Given all we know about the impact of social programs, why is it taking so long for programs that address them to be implemented? In short - many are trying, but it is really, really hard. 9 in 10 hospitals actively screen for social needs, but only 62% of them do so in any consistent manner. (KFF) A SDoH strategy requires Providers to take on new activities for which they are not accustomed, namely partnering with community programs and referring Patients outside the health system. Fragmented insurance programs and funding for public initiatives makes it extremely difficult for Providers to understand what is actually being paid for. And, on top of all these challenges, the Trump administration is continuing to make it more difficult for individuals who need social support to get it (e.g., work requirements).

Courtesy Deloitte

Additionally, there remain significant technical challenges to running a SDoH strategy in the provider ecosystem. Gaps or inconsistencies in data make it difficult to aggregate and use reliably. Less than ⅓ of hospitals actually integrate any SDoH data into their EMRs (Deloitte). Finally, individuals suffering from these issues tend to be some of the lowest utilizers of technology (52% of people with 2+ chronic diseases even go online!), requiring the need for significant operational programs. Startups across HCIT are beginning to attack these issues (see below), but their progress has not yet accelerated.

Public Sector Initiatives

With the goal of improving overall population health, federal and state governments have launched a series of programs to incentivize and fund the launch of SDoH initiatives in both the public and private sectors.

In the federal sphere, the push has been led by the Center for Medicaid and Medicare Innovation (CMMI) to drive the development of new delivery models. In 2014, the organization launched the “State Innovation Models” initiative with nearly $950M in funding being awarded to states that develop revolutionary payment models. Importantly, many of these models have an emphasis on Population Health (required for Round 2 of funding) that is further driving the need for a SDoH strategy. CMMI also established the “Accountable Health Communities” program to more deliberately assess the role social determinants play in healthcare utilization and costs. Even CDC has gotten in on the action, with initiatives focused on everything from curbing youth violence to reducing lead content in houses.

With this guidance from the federal government, several states have developed innovative approaches that incorporate SDoH as core tenants to meeting their goals. For example, Oregon Coordinated Care Organizations (CCO) and Colorado Regional Collaborative Organizations (RCCO) are participating the State Innovation Models initiative. They receive capitated payments and utilize community and social programs to stifle their rising costs. Additionally, some states have taken it a step further by linking some Delivery System Reform Incentive Payments (DSRIP) payments directly to steps Providers take to address community needs. (KFF)

Despite movement in the right direction, these models remain in “pilot” stage with fragmented funding sources. Thus, it is also important that we explore how the private sector is approaching these issues as well.

Private Sector Initiatives

I am a big believer in the power of the free market to help solve some of our most pressing challenges - even those challenges focused on social causes. There are absolutely limitations to that statement, but I find that most causes have a potential business model that can facilitate rapid innovation in pursuit of profits while also doing some real good. SDoH is an example of this opportunity.

As mentioned previously, the shift to value based models of payment has put greater pressure on Providers to find new ways to contain costs while improving quality of care for their Patients. Particularly within institutions that serve significant disadvantaged populations, Providers are learning that one of the greatest ways to achieve their cost containment objectives is by partnering with community organizations to address SDoH.

Here are a few notable examples:

Geisinger uses "Community Health Assistants" (CHA) who are non-licensed professionals designed to assess Patients needs and direct them to the right programs
Boston Medical Center recently physicians worked with their law teams to write protection letters keeping utility companies from shutting off necessary services for at risk Patients
Kaiser Permanente recently launched a $200M plan to address homelessness in its service area
Montefiore Health System invested through the Bronx Health and Housing Consortium to provide housing for its homeless Patients, yielding a 300% ROI
UnitedHealthcare has invested millions in grants for community organizations to address a variety of SDoH

And this is just a short list of the programs that currently exist! Private businesses investing in SDoH will undoubtedly lead to safer, healthier communities at a much more rapid rate than the programs currently run by government entities. As the industry continues to shift its method of payment, we can expect this trend to accelerate - benefitting the most disadvantaged across our country.

The Role of Technology

Okay - now to the part where I feel most comfortable - determining how we can leverage technology to help address this pressing issue. First, let me begin by saying that technology by itself has rarely been successful in healthcare. It is a tech-enabled services industry by default, and addressing SDoH is no different. Technology coupled with the right people and the right process can yield massive results.

There are three steps to addressing SDoH in delivery systems, and full stack technology solutions can help to support each step:

Assess: Determine individuals who may be a good fit for inclusion in a social program. This is typically done at the point of care through questionnaires or by algorithmically reviewing Patient data aggregated across sources
Refer: Identify the proper programs in the Patients’ geographic area for which he or she is qualified and make the referral
Close the Loop: Follow up with the social program to ensure the Patient has enrolled and participated, integrating data wherever possible

Given this process, what functionality can technology vendors provide that helps to manage it efficiently? I believe this post from Healthcare Informatics gives the greatest insight into key product capabilities:

Resource Directories: Up-to-date resource directories of local social service Providers and their eligibility requirements. These are going to be highly localized and maintaining them requires significant people-effort on the part of the vendor
Screening Tools: Tools that Providers can employ at the point of care to determine if a Patient is a good fit for a social program
Closed Loop Referral Platforms: Easy to use platforms that can be deployed across social programs to handle coordination of care. Anything that prevents a phone call or fax from being the only mechanism of data transmission is a step up
Reporting and Analytics: It is unlikely that anyone gets it right on the first try, so fine tuning the programs will absolutely be key to future success. Also, it’s always good when you can prove to the hospital administrator that your program actually has a real ROI.

Courtesy Patchwise Labs

Technology vendors in the space seem to be taking many different approaches to solution development. Companies like Healthify and NowPow have developed full stack solutions that touch on all of the above pieces. While both of these organizations are just getting started, they each have real clients that have implemented their solutions. A year ago, for example, Healthify had over 25 enterprise clients, spanning payers, Providers, and government entities(Modern Healthcare). As an alternate approach, a company that has been gaining traction recently is Circulation - an organization that aggregates ride sharing services for at-risk Patients and helps them to get to their appointments on time. This is a meaningful problem for 3.6M people who face a medical transportation barrier each year, and example of how tech can be used to address a single SDoH with real impact.

With the above said, the company that I believe is taking the most meaningful step towards addressing these issues is CityBlock. I will let them explain more clearly what they do:

“We created Cityblock with the simple idea that cities should be healthy places to live -- for everyone. Three trends motivated us: First, underserved populations living in our biggest cities have continued to have disproportionately poor health outcomes, with interventions coming much later and at a significantly higher cost than for other populations. Second, nearly all innovation efforts have been focused on people with means, despite urgent need in underserved lower-income communities. And finally, the business of healthcare has become increasingly transactional, leaving little room for meaningful relationships between Patients and clinicians, driving increased dissatisfaction among both. We set out to challenge this status quo.”

Healthcare is local, and CityBlock has built homegrown technology (called Commons) coupled with a relationship-based approach to address medical and social issues in the neighborhood. The organization combines preventative care with local access to resources to address health disparity in low socio-economic urban communities. Cityblock realized that technology alone was not enough to tackle this problem, and put its money where its mouth is by taking on operational risk simultaneously. CEO Iyah Romm says it perfectly in MobiHealth News:

“Despite the boom in venture-backed and technology-based innovations, people on Medicaid and other low-income urban populations have been largely left behind. By combining a data-driven approach to both operations and clinical service delivery, Cityblock has the opportunity to address those needs and unlock value in entirely new ways. We believe that by investing in tools and care approaches that meet people where they are, we can achieve this goal, while also building a business that drives down the spiraling cost of healthcare”

At this point, we are at the forefront of the explosion in technology designed to address SDoH. The aforementioned Healthcare-Informatics article states that the market for this tech is expected to increase 3-fold over the next 3 years. However, challenges to scaled adoption remain present:

Reimbursement: It remains unclear what Payers will reimburse for Providers that have one foot in the Fee For Service and one foot in the Value Based Payment canoe
Workflows: Many of these tools require integration into the Provider workflow to maximize use. Providers have a lot on their plate, and something that is viewed as “one more tool” will not work
Integration: The best platforms will efficiently integrate back into Provider EHRs - something that is not always very easy to do without significant people time

These are the same challenges that impact every piece of technology implemented in healthcare; and, I believe that these new products will merit the time investment needed to overcome them.

In Conclusion...

Going into writing this post, I didn’t really know what to expect. I feel like most healthcare news these days is about a government agency trying to tear down programs and leave socio-disadvantage populations out in the rain. However, I actually feel uplifted by the amount of investment we see in this space. Government entities, private companies, and technologists are coming together to address an issue that negatively impacts an extremely disadvantaged portion of our society. I am excited to see how these programs progress over the coming years to truly improve population health.

Additional Sources Utilized But Not Quoted Directly (Links Embedded)