CDS Development , Equity , Underserved populations
While most patients have little to no experience with the development and implementation of clinical decision support (CDS), many are familiar with and often engaged in collaborative decision-making with their caregivers and direct-care clinicians. Patient and caregiver trust that their individual circumstances are considered in the shared decision-making process is key to meaningful communication and follow-through of the agreed upon plan of care, helping patients to be more informed and active partners. To accomplish this, CDS development and implementation should rely on a foundation of evidence-based content that includes consideration of the impact that unique patient circumstances and abilities have on the patient’s care and ultimate health outcomes.
Health equity implies a system designed and built for inclusion and equal access to the best health outcomes for all. When considering the components of health equity, the healthcare industry often refers to socio-economic status, which incorporates education, income and occupation. In fact, health equity is more expansive, and refers to inclusion and best health outcomes for all no matter their physical environment, personal behaviors and abilities, or social circumstances (gender identity, military services, sexual orientation, citizenship status, social status, history of incarceration, culture and tradition, social connectedness, work conditions, early childhood education and development).
Professional and lay people often confuse the concepts and meaning of diversity, equality, and equity. In the past, I have used them interchangeably, but they’re not the same. Let’s consider the differences by applying the definitions to a common class of preventive health guidelines that can be supported by CDS content and tools – sexually transmitted infections (STIs).
I appreciate the diversity of people with their varied spiritual backgrounds, gender orientation, culture, attitudes, skills, abilities, communities. Does the STI research leading to evidence-based guidelines include the relevant diversity of patient populations? Is research conducted with sufficient samples of men having sex with men, heterosexual sex, and women having sex with women; people exposed to licit and illicit intravenous drugs; those with concurrent complex conditions; young adults and the elderly; the poor and the well-off, and those with cognitive, developmental, and physical challenges and those without?
Equality means that all people are treated exactly the same, no matter what their circumstances. Equality issues for STI treatment could be requiring young adults to obtain parent permission for testing or treatment or treating individuals with limited access to health care exactly the same as those that have unlimited access.
Equity aims to promote fairness, by providing people with the things they need to achieve the desired health outcomes of the CDS content and tools - doing more for those who need it to achieve the same outcomes. Do the CDS researchers, authors, and designers recognize that some patients and caregivers are disadvantaged by their circumstances? Do they consider the social circumstances, behaviors, and physical environment in the CDS design? Do people with positive STI results have access to treatment and medications? Can the individual afford treatment? Do they have transportation to services? Can they find primary and specialty care for ongoing health issues resulting from STIs?
Equity in CDS Development and Implementation
Developing and implementing CDS can be challenging enough without considering equity. Yet if CDS designers and implementers hope that the use of their CDS content and tools will assist in improving health outcomes for all patients, no matter their circumstances, they would benefit from planning and designing for equity as early in the development and design stage as possible. Let’s examine some of the steps and decisions that might be made in the process of translating STI clinical guidelines and evidence into STI CDS tools.
One way to picture this is to visualize a clinician turning the screen to face the patient or caregiver during a clinical encounter to share information provided by the CDS. Another option might involve the patient or caregiver viewing the CDS intervention on their smart phone or personal computer. Imagine what’s going through the patient’s or caregiver’s head when they view the information generated by the CDS. What questions will they ask or perhaps just think, without asking? Does the design of the CDS content or tool anticipate those questions?
Let me give you a sneak preview of the bottom line before we proceed: Successful consideration of equity can not occur without direct care clinicians, patients, and caregivers at the table from the beginning of CDS content and tool research and design through testing and use. While CDS professionals may have expertise in research, design, development and implementation, they need the expertise of direct care clinicians in identifying the relevant populations they serve, the best methods and barriers of teaching and follow-through for those under-served or disadvantaged groups, and the work flows of their practices. Patients and caregivers are expert in their lives: behaviors, abilities, social circumstances, physical environments.
Existing frameworks can be leveraged to aid in guideline development[i] and implementation. We draw from two models in use, the GRADE system[ii], which categorizes the quality of the evidence for a specific evidence-based resource, and the Shiffman model[iii], which defines a series of steps in the CDS development and implementation process. Below, we ask some questions of the people conducting clinical guideline research and designing CDS content and tools to address the incorporation of equity. We then provide some health equity examples relevant to STIs.
Evidence-Based Guideline Selection
- Does the clinical guideline research identify and reference relevant underserved or disadvantaged communities (those with known bias and economic, health, social, and environmental burdens)? Some examples:
- Gender inequality in education and restricted social autonomy for women can impact choice and availability of treatment for STIs and limit identification and treatment of sexual partners. Do studies identify and examine cohorts where education and autonomy are a factor?
- Health data for Los Angeles County states that chlamydia rates are highest among African-American females[iv]. Do studies related to chlamydia include African-American females as a separate cohort?
- Some STI studies and guidelines specify certain populations for inclusion or exclusion, such as men having sex with men or antibiotic use for infants. Are relevant populations such as these considered?
- Does the clinical guideline research include inequities and barriers present in underserved or disadvantaged communities and report on the effectiveness of the recommended care when provided in those communities?
- Does the research include studies of incarcerated or food insecure persons with STIs?
- Does the research identify health outcomes for a disadvantaged or under-served population related to STI treatment (e.g., a reduction in the STI incidence, a reduction in the transmission of disease to newborns, an increase in the proportion of patients that take the full cycle of antibiotics)?
CDS Development and Integration with Health IT, Clinician Workflow and Patient Life Flow
- Does the CDS content and/or tool link to local social services for recommended interventions or choices?
- Does the resulting CDS action automatically send prescriptions to the patient’s pharmacy or does the tool interface with a local social services database for referral?
- Are patient, caregiver, and direct care clinicians with expertise in relevant disadvantaged communities included in CDS development and implementation?
- Are Patient Experts included as co-principle or associate investigators?
- Do the study or design teams have stakeholder advisors[v]?
- Has the CDS been tested in “real life” settings (clinical and non-clinical) that include disadvantaged communities?
- Is the CDS content and/or tool tested in a variety of primary care and specialty clinics with patients from varied backgrounds and circumstances and as well as in locations that provide computer access to those without home computer or smart phone access?
- Does the CDS output and resulting guidance address issues the direct care clinician and patient might face together while using the CDS content or tool?
- Does evidence-based guidance consider out-of-pocket cost of antibiotic treatment?
Call to Action
- Ensure the selected evidence-based guideline addresses relevant disadvantaged patient populations.
- Budget for and recruit patient, caregiver, and direct care clinicians with disadvantaged community experience to participate in CDS development and implementation teams. This includes all aspects of the process (e.g., governance, planning, design, operations, learning).
- Examine and address inclusions and exclusions within the studies.
- Design strategies and interventions to overcome identified barriers in all CDS actions.
- Invest in CDS pilot testing for small, underserved populations[vi].
The earlier that health equity is considered when identifying evidence-based guidelines and developing and implementing CDS content and tools, the more impact the CDS can have on all patients, regardless of their circumstances. In order to achieve this goal, CDS developers and implementers must consider budgeting for, recruiting, and using patient, caregiver, and direct care clinicians who can contribute to ensuring that inclusion and access are considered during all aspects of the CDS lifecycle.
[i] Kavanagh BP. The GRADE system for rating clinical guidelines. PLoS Med. 2009;6(9):e1000094. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2735782/
[ii] Considering health equity when moving from evidence-based guideline recommendations to implementation: a case study from an upper-middle income country on the GRADE approach Javier Eslava-Schmalbach, Paola Mosquera, Juan Pablo Alzate, Kevin Pottie, Vivian Welch, Elie A Akl, Janet Jull, Eddy Lang, Srinivasa Vittal Katikireddi, Rachel Morton. Health Policy and Planning, Volume 32, Issue 10, 1 December 2017, Pages 1484–1490, https://doi.org/10.1093/heapol/czx126
[iii] Tso, G. J., Tu, S. W., Oshiro, C., Martins, S., Ashcraft, M., Yuen, K. W., Wang, D., Robinson, A., Heidenreich, P. A., … Goldstein, M. K. (2017). Automating Guidelines for Clinical Decision Support: Knowledge Engineering and Implementation. AMIA ... Annual Symposium proceedings. AMIA Symposium, 2016, 1189-1198. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5333329/
[iv] Can acknowledging racism address high STD rates? Nicole Knight. Center for Health Journalism Fellowships Blog. March 16, 2018. https://www.centerforhealthjournalism.org/2018/03/13/can-acknowledging-racism-address-high-std-rates
[v] Patient stakeholder engagement in research: A narrative review to describe foundational principles and best practice activities. Harrison, et.al. Health Expectations. 2019; 1-10. https://onlinelibrary.wiley.com/doi/epdf/10.1111/hex.12873
[vi] Big Data Science: Opportunities and Challenges to Address Minority Health and Health Disparities in the 21st Century. Xinzhi Zhang, MD, et al. Ethnicity and Disease. Volume 27, Number 2, Spring 2017. https://www.researchgate.net/publication/316291032