Primary Care Physician Perspectives on the Influence of Patient Values, Health Priorities, and Preferences on Clinical Decision-Making for Complex Patients with Multimorbidity: A Qualitative Study.

Introduction: The prevalence of patients with multimorbidity (ie, multiple chronic conditions) is increasing. Clinical decision-making guided by patients' values, health priorities and goals, and treatment preferences is particularly important in the context of interacting diseases and psychosocial needs. Physicians face challenges incorporating patient perspectives into care plans. We examined primary care physician (PCP) views on the influence of patients' values, health priorities and goals, and preferences on clinical decisions for patients with multimorbidity and increased psychosocial complexity. Methods: We conducted semi-structured telephone interviews with 23 PCPs within patient-centered medical home teams in a nationally integrated health system in the United States between May and July 2020. Data were analyzed via thematic analysis with deductive and inductive coding. Results: Three major themes emerged: 1. Patient personal values were rarely explicitly discussed in routine clinical encounters but informed more commonly discussed concepts of patient priorities, goals, and preferences; 2. Patient values, health priorities and goals, and preferences were sources of divergent views about care plans between healthcare teams, patients, and families; 3. Physicians used explicit strategies to communicate and negotiate about patient values, health priorities and goals, and preferences when developing care plans, including trust-building; devoting extra effort to individualizing care; connecting patient values to healthcare recommendations; deliberate elicitation and acknowledgement of patient concerns; providing "space" for patient perspectives; incorporating family into care planning; pairing physician to patient priorities; and collaborative teamwork. Conclusion: Primary care physicians perceive patient values, health priorities and goals, and preferences as influential during clinical decision-making for complex patients with multimorbidity. Participants used concrete strategies to negotiate alignment of these aspects when physician-patient divergence occurred. While rarely discussed directly in clinical encounters, personal values affected patient health priorities, goals, and preferences during care planning, suggesting a clinical role for more deliberate elicitation and discussion of patient values for this population.

Factors affecting primary care physician decision-making for patients with complex multimorbidity: a qualitative interview study.

BACKGROUND: Patients with multiple chronic conditions (multimorbidity) and additional psychosocial complexity are at higher risk of adverse outcomes. Establishing treatment or care plans for these patients must account for their disease interactions, finite self-management abilities, and even conflicting treatment recommendations from clinical practice guidelines. Despite existing insight into how primary care physicians (PCPs) approach care decisions for their patients in general, less is known about how PCPs make care planning decisions for more complex populations particularly within a medical home setting. We therefore sought to describe factors affecting physician decision-making when care planning for complex patients with multimorbidity within the team-based, patient-centered medical home setting in the integrated healthcare system of the U.S. Department of Veterans Affairs, the Veterans Health Administration (VHA). METHODS: This was a qualitative study involving semi-structured telephone interviews with PCPs working > 40% time in VHA clinics. Interviews were conducted from April to July, 2020. Content was analyzed with deductive and inductive thematic analysis. RESULTS: 23 physicians participated in interviews; most were MDs (n = 21) and worked in hospital-affiliated clinics (n = 14) across all regions of the VHA's national clinic network. We found internal, external, and relationship-based factors, with developed subthemes describing factors affecting decision-making for complex patients with multimorbidity. Physicians described tailoring decisions to individual patients; making decisions in keeping with an underlying internal style or habit; working towards an overarching goal for care; considering impacts from patient access and resources on care plans; deciding within boundaries provided by organizational structures; collaborating on care plans with their care team; and impacts on decisions from their own emotions and relationship with patient. CONCLUSIONS: PCPs described internal, external, and relationship-based factors that affected their care planning for high-risk and complex patients with multimorbidity in the VHA. Findings offer useful strategies employed by physicians to effectively conduct care planning for complex patients in a medical home setting, such as delegation of follow-up within multidisciplinary care teams, optimizing visit time vs frequency, and deliberate investment in patient-centered relationship building to gain buy-in to care plans.

Does machine learning improve prediction of VA primary care reliance?

OBJECTIVES: The Veterans Affairs (VA) Health Care System is among the largest integrated health systems in the United States. Many VA enrollees are dual users of Medicare, and little research has examined methods to most accurately predict which veterans will be mostly reliant on VA services in the future. This study examined whether machine learning methods can better predict future reliance on VA primary care compared with traditional statistical methods. STUDY DESIGN: Observational study of 83,143 VA patients dually enrolled in fee-for-service Medicare using VA and Medicare administrative databases and the 2012 Survey of Healthcare Experiences of Patients. METHODS: The primary outcome was a dichotomous measure denoting whether patients obtained more than 50% of all primary care visits (VA + Medicare) from VA. We compared the performance of 6 candidate models-logistic regression, elastic net regression, decision trees, random forest, gradient boosting machine, and neural network-in predicting 2013 reliance as a function of 61 patient characteristics observed in 2012. We measured performance using the cross-validated area under the receiver operating characteristic (AUROC) metric. RESULTS: Overall, 72.9% and 74.5% of veterans were mostly VA reliant in 2012 and 2013, respectively. All models had similar average AUROCs, ranging from 0.873 to 0.892. The best-performing model used gradient boosting machine, which exhibited modestly higher AUROC and similar variance compared with standard logistic regression. CONCLUSIONS: The modest gains in performance from the best-performing model, gradient boosting machine, are unlikely to outweigh inherent drawbacks, including computational complexity and limited interpretability compared with traditional logistic regression.

Protocol-driven primary care and community linkages to improve population health in rural Zambia: the Better Health Outcomes through Mentoring and Assessment (BHOMA) project.

INTRODUCTION: Zambia's under-resourced public health system will not be able to deliver on its health-related Millennium Development Goals without a substantial acceleration in mortality reduction. Reducing mortality will depend not only upon increasing access to health care but also upon improving the quality of care that is delivered. Our project proposes to improve the quality of clinical care and to improve utilization of that care, through a targeted quality improvement (QI) intervention delivered at the facility and community level. DESCRIPTION OF IMPLEMENTATION: The project is being carried out 42 primary health care facilities that serve a largely rural population of more than 450,000 in Zambia's Lusaka Province. We have deployed six QI teams to implement consensus clinical protocols, forms, and systems at each site. The QI teams define new clinical quality expectations and provide tools needed to deliver on those expectations. They also monitor the care that is provided and mentor facility staff to improve care quality. We also engage community health workers to actively refer and follow up patients. EVALUATION DESIGN: Project implementation occurs over a period of four years in a stepped expansion to six randomly selected new facilities every three months. Three annual household surveys will determine population estimates of age-standardized mortality and under-5 mortality in each community before, during, and after implementation. Surveys will also provide measures of childhood vaccine coverage, pregnancy care utilization, and general adult health. Health facility surveys will assess coverage of primary health interventions and measures of health system effectiveness. DISCUSSION: The patient-provider interaction is an important interface where the community and the health system meet. Our project aims to reduce population mortality by substantially improving this interaction. Our success will hinge upon the ability of mentoring and continuous QI to improve clinical service delivery. It will also be critical that once the quality of services improves, increasing proportions of the population will recognize their value and begin to utilize them.