Conditions in the places where people live, learn, work, and play affect a wide range of health risks and outcomes; these conditions are known as social determinants of health (SDH). Everyone knows the core areas of taking care of oneself: eat well, stay active, don’t smoke, etc. However, just as important, health and health outcomes are also determined by access to and quality of social and economic opportunities in people's homes, neighborhoods, schools, places of work, and communities at large. It is known that SDH interplay with biological factors to impact health and health outcomes in a myriad of ways and that those populations with negative SDH are vulnerable to poorer health and health outcomes as a result. The SDH Americans face explain why Americans are healthier than other nations, but also why Americans are not as healthy as they could be or the remaining nations. In the United States, a standardized approach for incorporating, measuring, and addressing SDH in health records is lacking. There have been efforts to address this concern, but they have been ad hoc and have produced specialized tools for measuring and addressing SDH, leading to a lack of understanding for best practices. Customization should not be ignored, but a largely standardized approach could help to streamline incorporation, increase adoption, and yield usable data to lead towards improvements more quickly in succinctly addressing SDH.
Thank you for the rich discussion. As we wind down, I wanted to pose one final question. What should the priorities be for this emerging SDOH interventional field both from a research and public policy point of view?
Thank you again for sharing your expertise with us.
One of the many goals for collecting SDH related data is to understand and address differences in health equity. Because of the diversity of factors involved in SDH and the way they affect different populations, it seems like it would be important to have a quantifiable measure when we talking about addressing SDH to improve health equity. Does that measure exist?
Great question! My short answer is yes, some measures exist, most of which were developed within social work. I imagine many healthcare-centric measures will be developed soon as the healthcare systems begin to integrate unmet social needs into care models. However, I think the more difficult question is when this measure is developed and applied, how can it be used in a way that honors patient preference?
On the one hand, a quantitative metric could be useful to quickly understand the degree to which individual patients are experiencing unmet social need. Paired with other existing data and used on a large scale this could have serious implications for allocating funds, clinic workflow, and research. For example, within my program, we use a tool called the Self-Sufficiency Matrix, a tool developed by social workers that measures progress along 19 social domains on a scale of 1 (“In Crisis”) to 5 (“Empowered”) which allows us to assess at baseline then track changes over time.
On the other hand, while convenient it’s also reductive in that it completely neglects patient preference. Meaning, if a provider or provider team were to only see numerical scores they wouldn’t understand which domain the patient actually prefers to address. Is it possible that a quantitative metric handed to the care team could take the place of a patient conversation with a member of the care team? And if that is the case and one day becomes essential (bearing in mind clinic workflow), should it be done in a way that includes patient preference? If so, how?
Thank you, Kate for sharing this wonderful tool. I look forward to studying it. Who in the clinic is best to capture the data for the Self-Sufficiency Model? How often do you revisit the tool with patients? And, it looks like it could take some time to complete. How do you "sell" it to the staff?
I completely agree with the importance of any composite measure reflecting patient preference and prioritization. I also agree with the value of measures that reflect patient status, as opposed to capturing or reflecting social/environmental/or behavioral factors in isolation. The notion of self-sufficiency seems closely related to our embraced measures of resiliency, concrete support in times of need, social connections, etc., as promoted by the Protective Factors Framework of Strengthening Families initiative of the Center for the Study of Social Policy.
What do you see as the primary purpose for collecting SDoH data? Is it to modify the care that the individual patient receives, intervene and address adverse SDoH, risk factor stratification, addressing financial reimbursement by payers particularly in an ACO value-based model, accessing more on site community resources for health care systems, or to get a larger lens view of population health for the health care system? And if its all or some of the above, is there a silver bullet in SDoH data collection that can address all of these various possibilities? If not, what do see as the potential trade offs for current SDoH instruments and models of care?
That is a great question Arvin. As one would expect, it depends on who is answering the question. As an MD, I care most about individual level data to help understand my patient's life and interventions that may or may not work (e.g .asking a diabetic to exercise more in an area with low walkability). I also see SDH as opportunities for patients to get introduced to community services, and for organizations to find out which service orgs are most effective. For a leader in a healthcare system, SDH certainly are important for ACO-related programs, and for better understanding population health. I don't see a silver bullet because I think it's important to both collect individual data and know about community-level data, with the understanding that the "status" of a community may not reflect a particular patients' circumstances. I would refer the group to your article on unintended consequences (ncbi.nlm.nih.gov/pubmed/?term=...) as well.
Nice question, Arvin!
Seems to me that the primary purpose for collecting SDH data is to use those data for all the use cases you described (and possibly others)..
The silver bullet would be a flexible, modular SDH assessment tool that can:
• Build a vast library of data elements (DEs) used to collect data relevant to all use cases, existing and future
• Organize the library's DEs into logical categories/domains such that all relevant domains exist for all the use cases
• Add, remove, and modify the each DE so that they are all reliable, valid, and useful for their respective use case
• Create implementable rules and maps by which the selection of a use case automatically identifies the DEs associated with the use case
• Perform the data analytics relevant to the use case, and present the resulting information in appropriate visualization models.
This can all be done with a sophisticated spreadsheet-based app or other tools having the requisite technical capabilities.
The biggest challenge, imo, is defining and mapping the relationships between the use cases and their valid/reliable DEs, analytic algorithms, and presentation models, which would likely include AI processes.
I agree that the answer is all of the above, depending on the specific priorities of the data collector. Clinicians will value the opportunity to leverage data to initiate a discussion with patients. At the population level, such data is so helpful in guiding and informing community-level interventions and policy considerations. A relevant example is the use of the Early Development Instrument (EDI) in Vancouver, BC. Population-level data on the developmental status of children is matched with community-based programs and services to enable a determination of community needs, gaps, and capacity issues that must be addressed to improve developmental trajectories of children. Resource planning and allocation is greatly facilitated by this data.
During pregnancy and in pediatrics, I see the study of SDoH as a way to enhance the primary care system. Primary care is an ideal and underutilized platform for addressing challenges related to SDoH and population-level reach early in the life course. Advantages include: 1) frequent prenatal and pediatric visits, which are widely attended, even among high-risk families; 2) use of existing infrastructure to lower cost and decrease need for additional transportation; and 3) ability to build on preexisting provider relationships. Primary prevention programs that aim to prevent the adverse effects of poverty-related SDoH before they occur could be integrated into primary care. Co-located services that are fully integrated into primary care increase the likelihood that services are obtained.
Although I am a fierce advocate of the need to include SDH in all health encounters even in the face of many challenges, there is another. I was dismayed to read of commercial development of SDH data with the intent to develop "scores", holding the potential for abuse. The prospect of a formerly homeless individual being denied insurance, for instance, came to mind.
What are some thoughts about this issue?
I share your dismay. We are impresses with the extent to which the implications and interpretations of screening tools and scores are misinterpreted and misapplied. As we know, screening "merely" is meant to identify elevated risk. Furthermore, conditions such as SoDH do not ideally fulfill the criteria by which conditions are judged amenable to the screening process, nor do the SoDH tools ideally fulfill the properties expected of screening tests. This does not undermine the value of such tools, but rather emphasize the importance of results not being interpreted in isolation, but rather in the context of all that is know of the population. The results of screening for SoDH raise important issues and inform key questions. Merely citing a screening score is particularly perilous. We have seen a similar rationale in pediatrics applied to screening for adverse childhood experiences, in which experts have recommended viewing the ACE score, rather than the specifics contributing to the score, as the relevant data. This is a potentially perilous application/interpretation of otherwise important and useful information.
Good morning everyone. Thank you for the excellent participation thus far. I am excited to be joining the panel. As we continue forward with the discussion let’s acknowledge the importance of data that is being collected.
What data have you found to be most important to collect in order for organizations to improve outcomes? How can it be assured that the SDH data being collected is of high-quality and useful?
In terms of SDH intervention, we’re constantly reminded of how important it is to fully understand transportation-related resources and barriers. The program I work on spans urban and rural areas, includes ESL speakers, and serves many individuals with physical and intellectual disabilities. Client transportation barriers are unique and require great attention from program staff. Without working closely with clients to find solutions to transportation barriers, referrals and appointments will surely be missed. Securing transportation is absolutely essential to accessing SDH resources and ultimately improving outcomes.
In terms of SDH screening, to screen for unmet transportation needs we opted for a colloquial item, “How do you get around?” This item is entered into our database with a multi-select response and followed by a free text box for notes. It is intended to serve as a starting point for a discussion of transportation resources and barriers. (Our program is home-based and average visit time is approximately 45 minutes, allowing for ample discussion.)
In my (limited) experience, housing is the most important question when it comes to outcomes at least in adult clinics. It may be totally different in pediatrics, where childcare or baby supplies may be in higher demand.
Regarding data quality, I think that using standards on the backend of screening tools is the important first step. If an institution has resources like what Kate describes above, then the quality of data may not matter that much on the clinical side; once someone screens positive and moves into an effective program for linking them to necessary resources, the fact that they screened positive for housing or educational needs is not as important as the fact that they got the right level of attention and help in finding resources. On the research and policy sides, the results of screening and referrals are important, so using electronic, standardized tools and making sure the screening process is patient -friendly (not in a setting where they may be intimidated or embarrassed to state their needs) will help.
Thanks for the great questions, Dr. Garg. Anecdotally, I can't comment on the specific social determinant that has been most important to improve outcomes, but I can say that addressing social needs/SDH in general was shown to improve parent-rated child health and decrease social needs when studied at one of the hospitals at which I work, Zuckerberg San Francisco General Hospital (see jamanetwork.com/journals/jamap...). A recent review by Laura Gottlieb, Holly Wing, and Nancy Adler highlights other studies of interventions on patients' social needs. Their study demonstrates that there is a heterogeneity in outcomes measured, with most outcomes being "process measures" (such as number screened or number referred). Other measures include change in social determinant status (such as going from being unemployed to being employed), change in health outcomes, and impact on health care cost/utilization (see sciencedirect.com/science/arti...). Regarding quality assurance of the data, anecdotally I have found that the most accurate data is collected in-person with SDH-specific questionnaires that have been studied in prior research (such as Dr. Garg's WE CARE screening tool). While this data is generally incredible rich, it poses challenges for mining existing data to try to identify SDH. ICD codes exist for some SDH, but they are rarely documented by clinicians. Natural language processing, a method involving programs mining clinician notes for keywords (e.g. food insecurity) increase detection of SDH (see ncbi.nlm.nih.gov/pubmed/282952...), but are likely not a great substitute for prospective identification. Lastly I will say that the way the information is collected likely matters, as a recent study found that sensitive information was endorsed more commonly when collected in electronic format vs. face-to-face (ncbi.nlm.nih.gov/).
So, is the following a valid conclusion (and if not, how should it be modified)?
The quality and usefulness of SDH data can be determined if analysis indicates a statistically significant positive relationship between a people’s health (as measured by objective wellness and subjective well-being) and a safe, stable, and supportive living environment characterized by readily available and affordable healthy food, easy access to high quality and affordable healthcare that is delivered in a culturally mindful and rewarding manner that fosters positive patient-provider relations and engagement, good education that promotes health literacy and effective self-management, and the absence of neglect/abuse.
If it is a valid conclusion, then wouldn’t this be a reasonable way to judge the efficacy of any SDH assessment’s content and structure, as well as the process by which the data are collected?
In our previous study of Social and Behavioral Determinants of Health (SBHD) data in EHRs, we learned that sensitive SDBH data were being collected by persons in a number of different roles, such as admitting clerks. The exception was when the nurse's comprehensive assessment was documented using a standardized terminology (in this case, the Omaha System). Our re-use of such data for research has shown known patterns in SBDH associated with care outcomes, validating the quality of nurse-documented SBDH data.
We are doing assessments left and right (I know, I was recently a patient) and we need to be able to standardize our assessment data to enable patient centered care, ensure communication across the healthcare team, and have real data to help generate new knowledge and show care quality and outcomes.
In my experience the only alternative to structured documentation by a clinician for SBDH is patient-documented answers to instruments proposed by the IOM (2014a & b).
In our experience, the most valuable insights are gained through open forum approaches with residents. Processes that afford a population the opportunity to identify its needs and priorities are powerful vehicles to strengthen population health. While demographic data is certainly important, as evidenced by such tools as Harvard's Opportunity Atlas, the extent to which zip codes indicate well being, and the utility of such population-based measures of children's developmental status as the Early Development Instrument (EDI), discussions with residents that identify factors that may either strengthen or undermine such critical determinants of health and well being as resiliency, concrete support in times of need, and social connections may ultimately prove most useful in planning interventions and strengthening programs and services at the community level.
One problem with discussions of this topic is terminology:
• I agree with Karen's comments about Social and Behavioral Determinants of Health (SBDH), which I firmly believe is far superior to the constraints imposed by many definitions of SDH that inadequately address behavioral health issues/influences on overall health.
• Wellness, Well-being, and Quality of Life should more be clearly defined in a way that differentiates each one and describes how they overlap, e.g.,
linkedin.com/feed/update/urn:l...
In pediatrics, the mechanism through which poverty-related social determinants of health have negative consequences for early child health is largely through exposure to parental stress. While screening for SDH is key to opening conversations and identifying risks, screening for parental stress may be just as important. While families should be connected to services that aim to directly decrease household material hardships, interventions to help families cope with living in the context of these hardships are also needed. Providing interventions that buffer the adverse effects of living with material hardships by decreasing parental stress and improving positive parenting practices in the context of living with poverty-related SDH, is critical to decreasing the toxic nature of these stressors.
Thank you all for your thoughtful responses. As we kick off this discussion and examine the ways in which SDH data can be collected and assessed, acknowledging what data is applicable could be useful. Would it be helpful if a standardized set of SDH data were recognized, and if so, how could that data be determined as standard? Is the recognition of a standard set of data points for SDH dependent on the population being screened?
Since there is a somewhat limited set of SDH (going by PRAPARE, accountable communities, Healthleads, or any of the other groups that have published screening tools), I would frame this differently. The better approach would be a set of standardized representations of SDH-related concepts (e.g. housing situation, education level, home safety) that you could pull into screening tools, either those that been published or ones that are being created by different organizations. Community-level SDH are potentially much bigger so in that case I would start with the relevant ACS categories and look at resources like HealthLandscape. The standard set of data points definitely is population- specific. Different data points are important even in pediatric vs adult populations.
There are a variety of SDH screeners that have become more popularized recently, such as PRAPARE, which Dr. Cantor mentions above. A fairly large list of screening resources can be found on the UCSF SIREN website: sirenetwork.ucsf.edu/tools-res.... Some of the screeners were designed for adults, and some for children, which echoes Dr. Cantor's point about different data being more relevant depending on the population. While I agree that this is true, and that you should really understand the community that you serve before deciding what SDH information to collect, there are some SDH that are more universally applicable to broad populations. The 2015 National Academy of Medicine (NAM) report entitled "Recommended Social and Behavioral Domains and Measures for Electronic Health Records" identified 12 social and behavioral determinants of health that were most strongly linked to outcomes, and that had consistent questions that could be used for assessment and incorporation into EHRs, with their associated questionnaires. Optimally, EHRs will have the flexibility to contain a standard, broadly applicable set of SDH questions. By "standard", I mean questions that have withstood the rigor that the NAM committee employed when determining which SDH factors were most consistently strongly shown to be related to health, and also have standardized questions. The benefit to having some standard questions is that we could track our efforts to address SDH more easily, as well as compare to other organizations doing the same thing, breeding innovation and collaboration. At the same time, this should be in conjunction with - and not to the exclusion of - adding more population specific SDH screening items that could be used at more local levels.
There are at least two critical issue that must be addressed before SDH standards can be determined:
1) Decide on the domains to include. Not only is there disagreement in the number of domains an SDH assessment tool should have, but there’s even a reason to change the terms to “social and behavioral determinants of health (SBDH)” as the Institute of Medicine recommended bit.ly/1iwMKAK.
2) Decide on the precise wording and response scale for each data element (question-response set) that is used to measure each domain.
I suggest that these decisions be based on sound research tasked with determining the most valid, reliable, and useful assessment items/measures.
Validity and reliability should be determined by evaluating an assessment tool’s ability to accurately and dependably measure the key SDH/SBDH factors that influence a population and an individual. This may require somewhat different versions of a tool based on a person’s age, gender, culture, etc.
Usefulness, otoh, refers to a tool’s ability, e.g., to support decisions about clinical care and referrals, as well as provide insights regarding the effect of SDH/SBDH on a person’s wellness, well-being, and quality of life outcomes. bit.ly/2NYJe6i
Anyone care to comment on the value of adding 9-digit zip and dual-eligible status to a standardized minimum data set around SDoH?
9-digit zip is obviously useful for community-level determinants, and , perhaps more importantly, for finding local resources. The main issue with 9 digit zips in research is privacy, since in many areas it's the equivalent of giving a full home address. Dual-eligible seems less important. If SDH are relevant , it probably doesnt matter what insurance they have, and the person doing the referring (assuming it's a healthcare setting) would know the status. Dual-eligible status may help for community risk stratification but that's more of a policy than clinical issue.
I love the idea of including a 9 digit-zip from a research perspective. With respect to Dr. Beller's comments, I completely agree with the steps he laid out and the challenges he highlighted. Given that there has been a NAM report to address the domains, I feel like that is a good place to start, particularly given that the domains identified have been linked to health outcomes in research contexts. However, I still think there is disagreement about how strongly some of these measures are related to specific health outcomes. As such, perhaps assessment of these domains in certain settings may be more critical than in others- such as a safety net hospital. Then there is a question of value added from the health system's perspective. Does screening and possibly intervening on these SDOHs translate to $$ saved? I would love to hear some commentary on this.
Dr. Prather's comment about the importance of knowing how SDH screens and interventions impact specific clinical and economic outcomes in different setting is spot on! It not only is this very relevant to value-based care, but it could provide a path toward SDH standardization. Does such research exist? Who could fund and manage (additional) research?
I appreciate the challenges of attempting to identify those factors most relevant to the outcomes of optimal health, development, and well being that we seek for children and families. While we certainly recognize the overwhelming influence of social and environmental factors on such outcomes, and are mindful of the many sectors within which such factors "reside" (e.g., transportation, housing, neighborhood health and safety, early care and education, food & nutrition, economic development, arts & culture, workforce development & employment, etc), we are challenged to both capture the myriad of potential drivers of health and well being and attempt to prioritize their importance for intervention. In fact, prioritization is dependent on the needs of the population we serve. Screening may identify potential issues demanding our attention, but only if so viewed by the recipients of such services. Data must not be viewed in isolation, but rather in the context of the needs and priorities of the population to be served. I suggest that screening for social determinants may be strengthened by including a focus on such critical protective factors as families' perceptions of concrete support in times of need, social connections, and resiliency, while we seek to identify the specific issues/gaps that may undermine such protective factors and contribute to adverse outcomes.
There is a lot to be learned from zip code. However, there is also some really interesting geographical work around place-based shopping habits and mobility showing that people tend to spend a lot of time, participate in activities, and shop outside their own neighborhoods. This doesn't mean we shouldn't consider zipcode when we're thinking about SDH, rather that it should be considered, similar to what Dr. Dworkin said, in the broader context of individuals' actual lives and behaviors.
Welcome, and thank you for joining IHMI's discussion, "Screening and Assessment: Considering Tools for SDH Data and Outcomes." Here we will be discussing the ways in which health care engages with social determinants of health (SDH) data, whether through screening, assessing, or collecting, and the tools that we have to assist us. If we could begin with panelists providing the group with some background on their work within SDH and how data plays a role.
Thanks Nathan. My work in SDH is in 2 principal areas. We created FACETS , a set of publicly available, community-level data for New York City, categorized by census tract. Most of the data is from the American Community Survey ,but we also have data from the CDC, EPA, and local NYC agencies. The 2nd area (more interest than work) is around standardizing individual-level SDH data in EHRs. This would involve mapping tools like PRAPARE or the Accountable Communities screening tools into a standard like LOINC, as well as expanding LOINC to be able to represent the questions and concepts in those questionnaires directly. Expanded access to local data (like NYC opendata) and standardizing data points related to individual SDH are my goals for these areas.
Hi Nathan, thanks for the question. My work in the SDH space has largely focused on getting the appropriate SDH measures into the electronic health record (EHR). I work closely with Dr. Nancy Adler here at UCSF who co-chaired (along with Bill Stead) the National Academy of Medicine report focused on identifying the appropriate social and behavioral measures to be incorporated into the EHR. They provided the blueprint but now it is time to actually put their plan into action. The first question was whether individuals could reliably report on the recommended questions and how long it would take. We found that YES individuals can reliably report on the specific domains and can do so in less than 5 minutes. Next the question was whether these measures were related to health metrics. Again the answer was YES- at least when it came to self-reported mental and physical health. We are now embarking on analyses testing whether these measures prospectively predict clinical health outcomes in some existing data. Preliminary data again suggest that inclusion of the SDH measures provided an important value added. Our next steps are to figure out the various levers can be pushed and pulled to get these SDH measures routinely included in the EHR, identifying champions across departments to push for this, and working within existing workflows to make assessing SDH second nature.
I currently work on the Interprofessional Care Access Network (I-CAN) program at OHSU. I-CAN aims to teach health professional students about integration of SDH into healthcare and test the effectiveness of a student-administered SDH intervention. We are currently exploring two novel methods to measure programmatic success. First, we are carefully considering client goal-setting and client preference as communicated with the program by conducting a qualitative analysis of client goals set within the program. Preliminary results have shown that clients primarily seek assistance with navigation of the medical system, help in and around the home, and assistance with housing insecurity. Second, we are currently analyzing results of the Self-Sufficiency Matrix, a 15-item social domain screening tool used to measure social change at multiple time-points measured on a scale of 1 (In Crisis) to 2 (Empowered). Next, we plan to develop a data sharing plan and eventually a cost effectiveness analysis in cooperation with partner Coordinated Care Organizations using claims data.
I also led efforts to adapt and adopt a custom-built SDH-centric EHR for I-CAN. I now manage all EHR training and actively seek and integrate user feedback for quality improvement.
Hello! Thanks for the question, Nathan. I am an assistant professor at the University of California, San Francisco (UCSF), and am a pediatrician by training. I see patients at UCSF Benioff Children's Hospital, Zuckerberg San Francisco General Hospital, and Washington Hospital. I conduct research with the UCSF Center on Health and Community, the UCSF Social Interventions Research and Evaluations Network (SIREN), and the UCSF Preterm Birth Initiative. My work focuses on the utility of incorporating information on the social determinants of health into clinical decision making, the analysis of large datasets that contain SDH information, and biological manifestations of the social determinants of health. I am interested in how we can combine SDH data with traditional clinical data to help enhance risk prediction models. For example, we have shown that social isolation is an equally strong predictor of mortality as compared to traditional clinical risk factors such as high blood pressure and smoking, even when controlling for health status. Currently we are working on projects that mine health records for SDH data so that we can combine this information with other clinical data to predict outcomes. Additionally, through my work with SIREN, I am interested in how addressing social needs can impact both patients and providers.
weve done some work getting LOINC codes for the PCAM -- patient centered assessment method for both the questions and responses, and for the scoring of the results and identification of health concerns that can be included as part of an interoperable shared care plan.
As a general pediatrician and a clinical research investigator at the New York University School of Medicine and Bellevue Hospital Center, my work has focused on identifying the early antecedents of poverty-related disparities in the rates of early child obesity. I am particularly interested in understanding how social determinants of health impact early infant feeding and growth, as well as parent engagement in preventive services. Given the known negative outcomes associated with poverty, the American Academy of Pediatrics (AAP) and the Academic Pediatric Association (APA) independently issued statements addressing the growing problem of childhood poverty and its implications for children’s health. These statements urged pediatricians to screen for poverty-related social determinants of health, to develop interventions to reduce the adverse effects of poverty, and to advocate for programs and policies aimed at eliminating childhood poverty. In the context of early child obesity prevention in prenatal and pediatric primary care, we have incorporated screening for social determinants of health, in particular material hardships and psychosocial stressors, and referral to needed services.
Our work has long focused on the answer to the question, how can we best strengthen child health services to promote children's optimal health, development, and well being. The new millennium has brought our expanded knowledge of the critical importance of the "biology of adversity"-adverse childhood experiences, toxic stress, social determinants of health, health equity. We also recognize that 90% of the outcomes that we seek are driven by social, environmental, behavioral, and genetic/epigenetic factors. We must therefore take a "systems approach" and engage all sectors critical to such outcomes. Our longstanding interests in screening and early detection, heavily informed by our work in early detection of developmental and behavioral issues, has also led us to view such screening-including that of screening for SDH as important, but always interpreted in the context of families' priorities and needs-and never in isolation. We have come to appreciate the importance of viewing data in the context of what it tells us about families' protective factors and adverse influences, and incorporating proximate measures of our more distal outcomes, such as measures of family resilience and access to concrete supports, as important. In short, measure of SDH are incredibly valuable, but must always be viewed and interpreted in context and never in isolation.
Pending
Research and policy must not focus on screening and screening tools in isolation, but rather in the context of a robust, integrated approach to addressing patients needs and priorities through early detection, referral & linkage. Research and/or policy on the isolated use of screening tools and/or detection without measuring the impact of referral and linkage is counterproductive.
Pending
I agree with Paul- there is not a lot of evidence (yet) about the actual effects of SDH screening and referral programs in adults. Measuring their impact should be a priority. The second area should be figuring out ways to integrate all the players- healthcare systems, medical providers, public health agencies, community organizations, etc. - to help people the most.