".. what’s needed for a sustainable culture is an agreement by its people as to what is truly meaningful in life, what gives life purpose." Amen (Amin) to that! It's up to -- it's the responsibility of -- each party to strengthen a win-win culture not only for business goals but also for society as whole, because we won't be able to address effectively SDOH without societal change and building resources to lift everyone up. The US still spends significantly less than other developed countries on social support, preferring instead to spend more dollars on costlier interventions. Providers should not shoulder all the responsibility and be left without societal resources for their patients.

I always have some trepidation discussing these deep-rooted cultural issues that are based on socio-political-religious beliefs, values and philosophies, and that require fundamental changes in cultural determinants of how human nature is expressed. Some might view these issues as out of scope from a biomedical perspective since they fall into the psychological domain of thoughts, perceptions, attitudes and emotions that influence behaviors. Nevertheless, these non-medical factors have a neurological base in brain and nervous system functions, and even more importantly, they directly affect people’s health and well-being.

So, what role should the healthcare profession take in addressing the need for significant cultural change to address SDH problems? What specifically can we do help realize substantive societal/cultural change?

When assessing the impact of SDH, It's important we recognize it is socio-economic factors, and not culture, that exert more influence on the conditions in which people live. Both the CDC and Healthy People 2020, as well as many other groups clearly express that the conditions where people live, play, learn, work, and worship determine how healthy they may be including health risks and outcomes. Personal resources and supports available in our homes, neighborhoods, and communities account for significant differences as does quality and level of schooling, safety of our workplaces and neighborhoods, the cleanliness of water and air, availability of healthy affordable food, and the presence of positive social interactions and relationships.
I respectfully disagree with the use of 'cultural change' in the context of SDH bur rather want health care professionals to focus on the social and economic conditions of our communities that disadvantage many people and deny them healthy lives.

We can't lead from behind. I believe it's our calling to lead changes that benefit our patients, and not just through medical interventions. As physicians, we can impact SDOH on at the population level by lobbying, volunteering, running for office, etc.

One may say that we have a culture that is impacting our socioeconomic priorities. We should take a leading role in reshaping that environment, otherwise we'll never get off the hamster wheel.

I'll be very interested in hearing others' responses. We are looking at the patient experience of care coordination and in spreading the SDH assessment regionally. I am also working with human factors researchers to understand how nurses in the role of care coordination find and use information to develop the comprehensive shared care plan.

For organizations that deem SDH critical to their care delivery, such as safety-net provider organizations, there are preliminary efforts underway as noted in the following articles.

How 6 Organizations Developed Tools and Processes for Social Determinants of Health Screening in Primary Care An Overview (sirenetwork.ucsf.edu/tools-res...)
The authors describe “the processes recently used by 6 organizations to develop SDH screening tools for ambulatory care and the barriers they faced during those efforts.”

Developing Electronic Health Record (EHR) Strategies Related to Health Center Patients’ Social Determinants of Health (sirenetwork.ucsf.edu/tools-res...)
In regards to Community Health Clinics (CHCs’), the authors noted that, “EHRs present an opportunity to standardize the collection, presentation, and integration of SDH data in CHCs’ clinical records.^45 Toward that end, a national coalition of CHC-serving organizations created the “Protocol for Responding to and Assessing Patient Assets, Risks, and Experiences” (PRAPARE), which included a preliminary SDH data collection tool informed by the IOM’s phase 1 report.^45 PRAPARE includes most of the IOM recommended domains and a few additional questions specific to CHC populations.”

Gold R, Cottrell E, Bunce A, et al. Developing electronic health record (EHR) strategies related to health center patients' social determinants of health. J Am Board Fam Med. 2017;30(4):428-447. doi: 10.3122/jabfm.2017.04.170046.

Additional resources: sirenetwork.ucsf.edu/staff/rac...

There important research to do in this space--both re: the impacts of integrated care, and ways to implement effective programs. Our goal at SIREN is to synthesize and disseminate existing work and to catalyze new research in this rapidly evolving field. Our monthly SIREN newsletter regularly provides a summary of new relevant publications. You can read recent newsletters and sign up to receive them via email here: sirenetwork.ucsf.edu/newsroom/....

I would advocate for the importance of having the patient and caregivers involved in creating the care plan and especially in terms of their goals for care and what are acceptable interventions.

Any ‘shared care plan’ must be patient-centered, which can be ensured by involving the patient, family, and other caregivers. Currently, we see both individual patient-driven and team- driven care plans—but the combination will be most effective.

According to this article (catalyst.nejm.org/making-the-c...), the authors defined the vision for a Comprehensive Shared Care Plan (CSCP) as, “a vehicle for secure, virtual exchange of information among clinicians who all care for a single patient, regardless of the clinicians’ (and the patient’s) locations.”

The article went on to state a description of a CSCP, as described by the National Partnership for Women and Families, “is not setting-specific; it uses information technology to enable the clinical team to collaborate seamlessly as they help address the full spectrum of the patient’s needs across all care settings and over time. Although barriers to widespread adoption still exist, we believe that robust collaboration between the health care and technology communities can help to advance this promising approach.” A number of groups have developed templates or suggested versions of shared care plans (IHI, PCPI, AAP, insurance companies, etc.), but what is fundamental is addressing the coordination of care and most plans vest the leader role in a RN or social worker.

PART II:

In another article (healthdataanswers.net/comprehe...), Dr. Freeman states, “To move forward, we must encourage all disciplines (e.g., nursing, physical therapy, registered dieticians, etc.) to focus less on siloed, discipline-specific care and more on the communication and teamwork that optimizes patient outcomes. The patient’s care plan should reflect that interdisciplinary and longitudinal approach. Nurses can leverage their experience to promote the concept of comprehensive, shared care planning which follows the patient across the boundaries of care from their home to medical facilities and back again.”

Thank you for sharing Dr. Freeman's work. I had the pleasure of hearing her speak last year at the Nursing Knowledge, Big Data Science conference at the University of Minnesota. Her vision for interoperability to improve patient care is inspiring. I think that the other important aspect of developing a quality CSCP will include the meaningful inclusion of the 'patient's voice' into the EHR somehow to promote goal-concordant care.

While I think a majority of clinicians seek to have shared decision-making capability with the patients and families, there are a few impediments that restrict this:
1) The FFS (Fee-for-service) model doesn't reimburse these conversations; its easier to refer to a specialist than to have a true informed consent conversation
2) Medical literacy is a big issue, and most providers don't have real required time to sit down with their patients to truly explain medical conditions and treatment plans, again especially in the FFS model
3) Providers are not always easily accessible for questions after the visit from patients and their families - not a lot of providers or groups have other means for communication after a visit is completed for follow up questions

In a capitated model, like CareMore's, there is no worry about how much time it takes for a provider to help a patient. For example, one of my doctors recently saw a patient in the ER who had just been diagnosed with recurrent lymphoma. The physician spent three hours in the emergency room with the patient, his wife, and his sons on the phone discussing the patient's medical condition and treatment options. The patient and family ultimately decided to go home with hospice.

What I want to stress here, was that time and freedom of fear of reimbursement helped allow this to happen. In our current financial model in healthcare, it would have been much easier to do a basic H+P and admit the patient to the hospital and not spend the time it takes to have true informed consent.

Great example

Because individuals had multiple chronic conditions it is difficult to tease out the impact of a specific disease. In the Medicaid population, behavioral health conditions were more common co-morbidities than you might find in Medicare or commercial populations. We were trying to move toward consideration of complexity of disease rather than a specific disease process.

Re: which ones to choose, go where you get the most bang for your buck is, both cost and revenue. Several determinants are the same across mandatory and voluntary programs, e.g. ACO, MU, DSRIP, etc. Collaboration with CBOs and other organizations can help address patient needs. Risk adjustment models continue to be fine tuned, but need the data collection.

These are interesting questions -- but one thing we haven't considered in this age of growing health information exchange is when we are collecting sensitive information, what are we doing to protect the individual's right to privacy. I believe that providers have a duty to limit the collection and sharing of SDH data to when there is a likely benefit to the patient from sharing the information. That consideration underlies our decision to complete the social and behavioral determinants of health assessment on a very limited population -- those person's with pre-existing chronic conditions during care transitions back into the community. Because of the high prevalence of co-morbidity, we didn't limit to a single chronic condition (such as heart failure). With 25% of the adult Medicaid population having 1 or more chronic condition, that's a large target population so that small differences in hospitalization and emergency department utilization can have a relatively large impact in cost savings without burdening the primary care practice too much. For the practice, the economic benefit equation was in the large number of additional office visits after the proactive outreach. This was sustainable and led to regional spread of the health information exchange.

A second question looked at what social issues were so intractable that they would require shifts in policy -- opioid addiction comes to mind immediately. This problem is one where primary care needs to collaborate closely with behavioral health providers -- thus the role of care coordination is especially important. There is a lot of experimentation of how to best integrate the behavioral and primary care world. In our facility we would negotiate for the best care plan and site for coordination on a case by case basis.

Excellent points and practical suggestions!

It would be wise to focus on SDH data that has the real potential to lower costs and increase revenue, while improving clinical outcomes, patient engagement and well-being. As such, I agree that it makes sense to focus on collection of SDH data that satisfies multiple mandatory (and voluntary) programs, as well as those pertinent to risk-adjustment models.

Protecting patient privacy is a must and limiting requirements as to which patients requires collection and sharing of extensive SDH data is a practical strategy that can be weighed against the benefits of improved outcomes, patient engagement, and practices economic gains.

And opioid addiction is a good example of how the effect of intractable social determinants on human behavior could be addressed with coordination of care between PCPs and behavioral health specialists.

Pushing the bar a bit: What about SDH issues that are influenced by deep-rooted socio-economic and cultural values/norms/mores such as poor housing, unsafe streets, unemployment, poverty, lack of affordable healthy foods, illiteracy, unsafe neighborhoods, environmental pollution, climate change dangers, social isolation, and on and on? Is it even reasonable to assume that clinicians can do more within their practices—beyond collection of SDH data, making referrals, sharing care plans and coordinating care—to bring about positive change in these areas for their patients? If not, then (a) an efficient SDH assessment and referral process is adequate and (b) the degree to which clinical outcomes are shown to be adversely affected by such intractable SDH factors (something where research is still needed) should not negatively affect care team performance metrics.

I wonder if intractable social conditions are like chronic illness where we do not expect cure, but help people to manage better given the limitations.

Love the analogy!

Re: discussion re: performance metrics...I think this article from Ashish Jha might be thought-provoking: blogs.sph.harvard.edu/ashish-j...

Me too, and it raises the issue of barriers from the perspective of the healthcare provider. It is one thing to identify problems but without solutions it can be discouraging and time consuming to address SDH.

Stephen, I appreciate your response and agree with you.

I think that the medical community is expected to address the deficient policies in our country, such as housing, food, safety, education, racial discrimination, etc, that lead to a majority of chronic conditions that our population faces.

I also think that it is a stretch to expect individual providers to do more beyond what you listed above (data collection, coordinating care, referrals to social groups).

Instead, I would suggest that it is a delivery system/payor's responsibility to help address them, whether directly as part of their benefits packages to their patients, or indirectly in advocating for policy changes in our country as a whole.

Piyush, I agree that expecting a delivery system/payor to be responsible to take constructive steps to help advocate for positive policy changes in our nation is reasonable. But how can they be held accountable if they fail to do so? Seems to me that those who pay bills may have business reasons for shirking such responsibility (which has associated costs) and passing blame for poor outcomes on the treating clinicians/providers.

Risk adjustment (RA) for SDH--a complex issue as reflected by Laura's last post--may mitigate clinician blame if valid RA algorithms are established and implemented effectively. However, that side-steps the challenge of having sustainable models and processes that directly address the underlying causes of adverse SDH.

I contend that this daunting issue boils down to the polar aspects of human nature. In case people are interested in examining this issue, I’ll start a new thread with a list of key aspects of human nature that I’ve compiled over the years.

Piyush, I agree with you regarding the importance of payer-provider collaboration. Both collect data that can enhance risk adjustment models. The latter have been improving for many years, and are already on sale by vendors in that space. Payer-provider collaboratives outside of IDNs have yielded positive results for many years. Apart from the data, payers have deeper experience in analytics and in case management and community-based programs. Collaboration with them is not only advantageous but also necessary.

Re: "those who pay the bills," for the majority of patients under different insurance types, the money comes from employers, unions, state or federal governments, etc. Payer intermediaries have been expected, esp. in the past 15 years, to have better outcomes in quality and member satisfaction or lose the membership from those who pay the bills. Such requirements have already reduced both membership size and the amount of quality bonus for under-performance plans. This environment has been generating new opportunities for collaborative care and research between providers and payers.

With that in mind, I'd rather talk about how to influence societal changes and proposed solutions that can reduce, say, homelessness, a strong social determinant of health. I'd rather talk about concurring aspects of human nature, and how to use compassion and ingenuity to delivery better outcomes for our patients.

Along the lines of Sharon's comment - what are the ethical considerations inherent in sharing a patient sensitive information? For example, is it ethical/appropriate to share within the circle of community/care providers the fact that a patient has been a victim of domestic violence (and has had a restraining order against a family member in place at certain points), and may be at risk of having access to care blocked in the future by the perpetrator? Is patient consent required to share that information?

Sharing sensitive information with CBOs within a safe corridor amidst the landmines of federal and state laws plus BAAs is important for successful interventions in the community beyond the provider's reach. In fact, the availability of such data from the providers is an enticement for CBOs because the referrals (which always entailed such data) support their mission and success. A win-win approach for all partners.

While we as clinicians may feel that sharing sensitive data is in the patient's best interest, the patient may not agree with that at all - especially in higher-risk populations.
There is significant data out there on medical mistrust and how it is highest among minority and disenfranchised populations. Although providers, payors, and ACO's, etc. see the benefits in sharing this information (and think we are able to put our biases aside and treat sensitive data objectively), patients really need control of their data if we are going to adequately address the issue of medical mistrust. There are standards to support granular privacy protections (i.e. DS4P, which really hasn't gained widespread acceptance yet) and functionality to support consent repositories accessible through the patient portal. I believe it is essential to build in these types of granular privacy protections as our interoperability capabilities grow and we continue to share data that will likely be of great benefit to many patients - but is a true barrier to care for others.

Think this Canadian article might be of interest to the recent thread: ncbi.nlm.nih.gov/pubmed/211507...

Laura and Hannah, good points to emphasize about security and privacy in sharing PHI esp. as more data have become digital, and as interoperability and HIE expand. Federal and many state laws, and HIE protocols, as complex as they are, allow for sharing minimum necessary information for the purpose of care management. And patient attitudes have changed and keep changing as they experience convenience outside the healthcare system and comfort with new technology.

Interesting study. Seems to me such patient resistance reflects (at least) two core fears in our cultures: 1. Fear that one's personal information may be used against a person by companies or authorities and 2. Fear of being judged by society as being worthless, inferior, undeserving because of one's lack of wealth, low status. The first is caused by lack of trust. The second, which I call this “People Appraisal,” is caused by the belief that a person's "inner-essence" (inherent self-worth, essential value as a person) is measurable by one's SES (and other SDH factors).

There can be valid reasons for the former, e.g., harm from the sale and use of personal information.

The latter, however, is without merit because there is no rational, objective way to measure one’s global value/worth AS A PERSON (a measure of what a actually person IS in totality). All we can measure objectively is what a person DOES (verbal and behavioral expressions) and HAS (materialistically, economically, physiologically, socially). What a person does and has is does NOT reflect what one IS. A person is one thing and one thing only: A fallible human being with the continuum of qualities in human nature.

Unfortunately, many/most cultures breed the belief that it is reasonable to judge a person’s ENTIRETY based on certain expressed thoughts, behaviors, knowledge, appearance, social groups, power, etc. As such, someone who feels inferior because of subjective society judgement (e.g., base on SES, race, nationality, educational/skill level, etc.) is likely to feel shame, humiliation, distrust, angst. They would naturally fear exposing themselves as being inferior if they were to provide personal information.

The solution: 1. Build trust that one’s identity would be protected and personal information used only to provide better healthcare (research, collaboration, decision support, etc.) and 2. Reject all people appraisals as foolish, destructive, and inherently faulty.

I'd like to propose a multicenter patient survey re: HIE sharing with quarterly reporting to stay abreast of evolving trends esp. as government and community organizations move to digital solutions and new payment models expand. The survey would examine also physician attitudes re: PHI sharing and privacy.

The first link below is a recent literature review of patient attitudes re: HIE. One factor worth noting is higher patient acceptance below the age of 40-45 years. The info reviewed in the article was published in 2015 or earlier, i.e. the actual work was done even before that. A lot has changed since that time and the literature can barely keep up with that pace.

The second link is an article addressing the level of privacy patients assign to their medical information including HIV. I believe the first article informs reading the second one.

The third article notes the value of patient-physician/provider relationship in patient views of sharing PHI.

ncbi.nlm.nih.gov/pmc/articles/...

tandfonline.com/doi/abs/10.108...

ncbi.nlm.nih.gov/pmc/articles/...

In my experience, healthcare is very local. Every city/region have their own strengths and weaknesses. Ensuring that an appropriate community asset mapping is completed before undertaking transformational efforts is essential.

However, there is some value to having a standardized approach for part of the equation, then allowing enough flexibility to adjust/adapt/iterate on the process.

This has been a helpful reference for me:
atsdr.cdc.gov/communityengagem...

I would agree that healthcare is local, and maybe even at a practice level. The process that we used was alerting the nurse of discharges that might be affected by social determinants, and using the SDH assessment as a way to identify the subset who needed immediate attention. For those cases, she went to the interprofessional team with the asset map to develop a response that was specific the individual's needs. The team learned with each case and eventually had a number of approaches that worked for them to employ for similar cases.

We measured outcomes using claims data, and that approach is more likely to be scalable than using multiple EHR systems to evaluate outcomes. Also using the existing health information exchange made it easy to spread the intervention throughout the eight-county region.

Creating a culture of health through population health management also requires an understanding and appreciation of the impact of SDHs. The final rule from HHS for the 2015 Edition Health Information Technology (Health IT) Certification Criteria, 2015 Edition Base Electronic Health Record (EHR) Definition, and ONC Health IT Certification Program Modifications (gpo.gov/fdsys/pkg/FR-2015-10-1...), calls for the ability to “address health disparities by providing certification to: standards for more granular capture of race and ethnicity; the collection of sexual orientation, gender identity, social, psychological, and behavioral data; for the exchange of sensitive health information (Data Segmentation for Privacy); and for the accessibility of health IT” (page 62603). The final rule calls for the ability to “enable a user to record, change, and access the following patient social, psychological, and behavioral data: Financial resource strain, Education, Stress, Depression, physical activity, Social connection and isolation, Exposure to violence (intimate partner violence)” on a standardized basis (page 62748-62749).

Nurses recognize, “Despite rapid technological advancements in the U.S. health system, care remains fragmented, uncoordinated, costly, and yields poor population health outcomes. Clinical care currently accounts for only 20 percent of health outcomes. To improve care and population health outcomes, healthcare providers must address other factors, including social determinants of health. To do this, we must first focus on addressing the variability in capturing and documenting social determinants in order to use this health data to benefit patients”( American Academy of Nursing on Policy, Putting “health” in the electronic health record: A call for collective action, 2015 nursingoutlook.org/article/S00...).

Great question. Important scalability work remains to be done. I will add to the great responses above that there is quite a movement towards using clinical decision support tools (in some cases to improve medical treatment decisions and more frequently, to facilitate community resource connections) in a way that may ultimately help scale activities. This includes folks like Healthify, UniteUs, NOWPOW, etc--and now Epic, Cerner, and other EHR vendors, who are creating similar platforms or partnering with these organizations. We do need some evidence examining whether these programs connecting patients with community resources make an impact on health. Other barriers to scalable solutions include payment/reimbursement for these activities. But there are some exciting opportunities out that could lead to payment reform that could help--these include some programs under state waivers that grant more flexibility in payment and other demonstrations like CMMI Accountable Health Communities, which if effective could lead to more permanent funding changes.

Discussion of scalability with the finance team is critical. As noted above by others, effective tools and funding are important. The latter is a challenge because many pilots and studies were funded by grants, then funding dries up for continuation let alone scaling up, and both endeavors often require doing more with less. New risk-adjusted payment models offer extra revenue but it is delayed by many months to two years. So to operationally scale up, one needs to collaborate close with Finance.

This brings up another challenge: evidence. Some studies do not show or address savings and ROI. Studies may not have a comparison group or be randomized to avoid baseline bias, and may be limited pre/post comparisons. And negative results are typically not publicized. The finance team will be looking for evidence, and will take a fresh look at the cost of scaling up from, say, a CHF pilot to diabetes, from alerts re: the most vulnerable patients to the broader population, from one facility or practice to many with varied "local" cultures and agendas. Finance will need to consider scalability in the context of shifting FFS and VBC revenue streams.

Time is also a challenge because the payer stakeholders including government already concluded that SDOH are important to reducing cost and/or efficient population management, and want attention to them now. So while it's still important to continue studying and sharing new ways of using SDOH to optimize our interventions and risk models, evidence from new studies trails the shift in payment models and the evolution of technology. Keeping an eye on the latter two during pilots and studies helps in pivoting as one scales up.

Dr. Akim, I agree with your response. The financial incentives in the FFS model make it difficult to scale even basic, common-sense solutions in addressing SDH. Without a wrap around, fully integrated model that is both clinically and financially aligned, scalability will continue to be a challenge.

This article (link below) about generalizability of study results is an excellent dive into factors that influence study outcomes and the ability to scale up new solutions. In this case the digital solution is telehealth - no spoilers from me about the result of implementing the latter. (P.S.: one can't upload PDF attachment)
ncbi.nlm.nih.gov/pmc/articles/...

I don't know exactly what paper you are referencing but I can talk about some of our experiences with behavioral health issues, particularly those with severe mental illness (SMI).

We run an integrated behavioral health and primary care practice, primarily working with Medicaid patients. SMI patients can be very demanding, time consuming, and intimidating. Appropriate training in trauma informed care, understanding how to de-escalate patients in crisis, having protocols for crisis stabilization, and having coordinated care with between NGO's, outpatient, and inpatient behavioral health institutions are critical to help mitigate these challenges.

Our case management team, equipped with the appropriate training, manages our patients with much more confidence and quality. It continues to be a learning curve for us, but everyday we are identifying more effective ways in building authentic healing relationships with our patients.

I believe Joe is referring to the paper shared by the authors on this group discussion (tagged here). Other papers here and elsewhere note challenges like reluctance, resistance, etc. So two things to talk more about:
1-This particular issue is about managing change particularly when it is rapid. So everyone, what do you do to manage that change?

2-Farhad, please share with us the changes that made engaging resource-intensive patients more effective.

Great insight. Yes this project required a rapid change in the role of the RN care coordinator from managing patients with diabetes to meet clinical guidelines, to proactively dealing with a very vulnerable group of patients at a critical point in their recovery. As we implemented the new process, the research team met with the practice on weekly basis to solve issues with the new clinical and informational workflow. As we move forward, we are looking at developing training for the expanded role in of the nurse in primary care to use data effectively, identify the target population, gain skill in care coordination in a community setting, and participate in an interprofessional team. The goal is to incorporate this in baccalaureate nursing education as the healthcare system evolves.

This article, Developing Electronic Health Record (EHR) Strategies Related to Health Center Patients’ Social Determinants of Health (available at: ochin.org/media/ASSESSDO_proce...), identifies that collection of social determinant health data in EHRs has the potential to lead to improved outcomes. Of note, Community Health Centers (CHCs) may provide patients the connection to the clinicians and appropriate resources in order to increase access to essential health care services. The American Nurses Association’s, Principles for Health System Transformation (available at: nursingworld.org/Principles-He...), calls for, “optimizing primary, community-based and preventive services while supporting the cost-effective use of innovative, technology-driven, acute, hospital-based services.”

Thank you for those resources, Dr. Cipriano. We hope to continue to work towards the goal of not only integrating SDH assessment into clinical workflows to improve transitions of care and health outcomes, but also towards finding ways to integrate SDH information into comprehensive shared care plans that are shared between settings so that the interdisciplinary team is simultaneously involved in patient care planning. We recently completed a review of the literature that was published in Nursing Outlook (open access) titled: Integrating social context into comprehensive shared care plans: A scoping review (Here is the link: doi.org/10.1016/j.outlook.2017...). However, we found that work in this area is only emerging at this time, so there remains much progress to be made. In addition, we feel that the patient's voice needs to come through in the EHR and care plan in a way that builds an authentic collaboration between the provider, patient, and family to promote concordance and improve adherence to a plan of care that is aligned with individual values, goals, and beliefs.

Participants in this discussion also might be interested in the publicly-available resources on the SIRENetwork.ucsf.edu website. The SIREN Evidence Library and Reports/Issue Briefs sections of that site include research studies on both the effectiveness of integrated (social + medical) care delivery and the implementation science that can help clinics improve integration. Our literature review of the effectiveness research done to date also can be found at ncbi.nlm.nih.gov/pubmed/286887.... Though the review was published recently, there are more recent articles in the Evidence Library that would be worth checking out.

Thank you for sharing your work, Dr. Gottlieb. This looks like it will be a very useful review.

I have created a map of the hemoglobin A1c distributions by ZIP Code in the communities served by my hospital. We have a meeting in the near future which includes representatives from the hospital, the Physician Hospital Organization, community health center and local academics where we will try to figure out how we use this information to improve the care of diabetics in the affected ZIP Codes. The data can be looked at using 5 digit or 9 digit ZIP Codes. When looked at using the 9 digit ZIP Code (mail carrier level) the number of data points in any 9 digit ZIP Code is relatively few so that data is statistically problematic.

My original plan was empower/implore the communities leaders to address the issue, using the healthcare professionals as a consultants ie, move the DM care outside the medical establishment and help the community leaders figure out how to solve the problem.

If anybody has done anything like this before, or can point me in the direction where I can learn from others,
it would be greatly appreciated.

You might want to check out the innovative work that has been done at Cincinnati Children's using geospatial data to guide decision-making around social interventions. The leaders for that work include Andrew Beck and Rob Kahn. They have a number of papers that may be relevant to your project. I've flagged a few of them below.

ncbi.nlm.nih.gov/pubmed/272458...
ncbi.nlm.nih.gov/pubmed/285839...
ncbi.nlm.nih.gov/pubmed/274023...

Thank you

Dr. Gottlieb's research on SDH within an EHR provides important insights related to data collection, sharing and reporting and ways to deal with them. It also discusses the practical use of SDH data to identify risk and use of lists to identify appropriate referral resources by proximity to a patient’s address and by patient preferences. Furthermore, it offers a useful examination of five steps involved in with using SDH data domains and maps assessment items that compare ASSESS, PREPARE and IOM questions and indicates potential actions based on the answers. It also describes a way to identify referral options through the creation and management of lists that considered distance, patient needs and preferences, and it describes issues concerning referral outcomes tracking. Bottom line: This research study, imo, offers a comprehensive and well-designed model to guide future SDH research.

I'm just learning about Dr. Gottlieb's work with SIRE network on SDH. The idea of using geomapping to identify potential resources is very exciting.

One of IHMI's benefits is providing a forum whereby clinician-researchers can come together to share their studies, ideas and lessons learned. Drs. Hewner and Gottlieb are a case in point wrt SDH. What we can all learn from this kind of interchange is a great example of a continuous learning health system--supported by a loosely-coupled social network of diverse participants--that builds evidence-based knowledge to inform clinical decisions, to identify technological and workflow challenges and possible solutions, and to spark innovation.

And I hope that it allows us to get a clearer understanding of the different perspectives and strengths that each discipline bring to the challenging problem of social determinants of health.

Your city or county department of health would have geo data to complement yours. It will also be a partner in intervention directly and indirectly by connecting you with community resources and people interested in supporting the effort.

A few years we competed in the DeBeaumont Foundation's to close the divide between primary care and public health. We used American Community Survey (ACS) - 2010-2014 five-year estimates at the ZIP Code Tabulation Area (ZCTA) level to make geographic SDH groups based on tertiles of poverty and of <High School education. The exposure is high risk (vs. low) and the outcome was poor A1c control. We are now re-doing the analysis (with more years of A1c data) at the census tract level to make groups comparing the highest risk quartile to the lowest risk.

Vincent - Would you explain how the geographic SDH groups' data are being (will be) used to bridge the divide between primary and public health. Thanks.

Yes, the local partners are using to target healthy eating/active living programming into the geographic areas with high poor control and high geographic SDH risk.

Vincent, I would also be interested in hearing about how you plan to bridge the divide between primary and public health as it may help to inform our work in developing an interoperable comprehensive shared care plan with the goal of linking healthcare settings to promote goal-concordant care.

Now we are also adding first-year medical students doing home visits (this was not part of the original plan) assessing SDH and connecting families to intersectoral partners around food insecurity, transportation, access to health care, etc. We also are working with the local HIE for the sharing of the SDH interoperable care plan. We will add electronic referrals for intersectoral partners by the end of 2018 in the HIE suite of applications.

The comprehensive shared care plan (CSCP) used the Health Level 7 Consolidated Clinical Document Architecture Release 2 Care Plan as the standard. During the implementation we made the transition from including our goal for the patient to trying to determine the patient's goal for care. This sounds small, but it was a big change that was facilitated by the discussion between the nurse and the recently discharged patient about social factors and their transition from hospital to home. When we think about a care plan that follows the patient over time, the social determinants become much more important than in a short-term single facility care plan. I think that primary care is the ideal place for both collecting and incorporating social concerns into the care plan because of the long-term relationship with the patient and their family.
Our regional health information organization (HEALTHeLINK) is working with stakeholders to implement a care plan that is shared across sectors (including behavioral health). The planning committee is ironing out the details of how and when care plans are shared with the long-term goal (2019) of sharing discrete data using the HL7 C-CDA r2 care plan. It's a journey.

How do you structure this shared care plan and does any group "own" it? In my experience, nurses and case managers have a more formal method of structuring care plan documentation; this may differ from that of behavioral health providers and even more so from MD's/DO's/AP's who have less formal methods of care plan/goal documentation. Additionally, in developing the shared care plan, have any privacy issues come up regarding surfacing behavioral health information to the entire care team?

There are four parts to the care plan: identification of health concerns, the patient's goals for care, setting interventions to reach those goals, and evaluating the outcomes. The PCAM helps the nurse identify health concerns related to SDH, but social work could also participate here. Physical health concerns include the medical diagnoses, but also nursing health concerns (for example risk for falls or limited health literacy). Nurses are skilled in the development and curating of care plans, but the comprehensive shared care plan really transcends any one discipline or setting. As these care plans become embedded in the health information exchange, they need to communicate the longer-term plan, rather than site-specific care plan, and there needs to be careful consideration of what information does any care provider need to know.
Our RHIO is meeting quarterly with community providers in both behavioral and primary care settings to try to resolve these questions as we move forward. For behavioral health we are trying to determine what should be included, however there seem to be some health policies that are encouraging the inclusion of some type of behavioral health participation in the shared care plan.

Our thinking on this evolved over time. Originally the nurse entered the values for each question into a lab test we called the PCAM. However, this created a barrier to acting on the identified social problem. We then developed a web-based interface for the nurse to enter the score for each question. After completion, the app provided the text of each response stated as a health concern (problem). This list was copied and pasted into the transition of care note . This facilitated discussion among the interprofessional team, and social problems could be manually entered into the care plan.
Our next step was to refine the app to be a digital analog of the paper PCAM, which we called the ePCAM. The ePCAM allowed the nurse to score each question anywhere along the continuum from no problem (strength) to big problem (health concern requiring immediate attention). We then worked with the internation LOINC committee to develop LOINC codes for the questions, responses, scoring, and text fields that form the basis of the care plan. I believe that you can access a test version of the ePCAM at: nursent5.nurse.buffalo.edu:133...

We are working on making the connection from the ePCAM to the comprehensive shared care plan automatic. Our goal is to have this clinical decision support tool bring social determinants of health into the team discussion of complex chronically ill persons and to share the results of that discussion across settings.

Designed to assess risk, PCAM measures SDH factors only in terms of quantity and/or frequency metrics. Since it does not ask any WHY questions, it doesn't provide insights into WHAT can be done to deal with those factors nor HOW to do it. That would require an in-depth assessment of people's thoughts, affect, behaviors, as well as details about the environmental, psychosocial, physiological and other factors that influence/cause/relate to the PCAM results. I think Dr. Galvin's idea of creating an SDH risk factors ontology is a wonderful next step, though it's likely to transcend LOINC codes.

The PCAM doesn't provide insights into the why what and how questions which is why it's important to consider the context in which it is used. The nurse completing the assessment is not reading the questions and recording a response, she/he is embedding the discussion in an outreach phone call following a discharge. The RN's skill in interviewing and in understanding the environmental and psychosocial factors that influence recovery is critical. Futhermore, the discussion of what to do in serious cases is carried on in the context of the entire interprofessional team. It's why I think that having a professional nurse or social worker participate in the calls is essential. The nurse is expert in developing a comprehensive care plan that includes the team, including the patient and caregiver.