Population Health Management最新文献

In recent decades, the integration of artificial intelligence (AI) into health care has revolutionized diagnostics, treatment customization, and delivery. In low-resource settings, AI offers significant potential to address health care disparities exacerbated by shortages of medical professionals and other resources. However, implementing AI effectively and responsibly in these settings requires careful consideration of context-specific needs and barriers to equitable care. This article explores the practical deployment of AI in low-resource environments through a review of existing literature and interviews with experts, ranging from health care providers and administrators to AI tool developers and government consultants. The authors highlight 4 critical areas for effective AI deployment: infrastructure requirements, deployment and data management, education and training, and responsible AI practices. By addressing these aspects, the proposed framework aims to guide sustainable AI integration, minimizing risk, and enhancing health care access in underserved regions.

Population health research has long demonstrated that where someone lives is highly correlated with health outcomes and quality of life. This study explored if this relationship remained between zip code socioeconomic deprivation index (SDI) and member-reported healthy days among commercially insured adults interacting with virtual care and navigation services between May 1, 2023 and May 31, 2024, offered by Included Health. Members received an SMS-based survey that included the 4 Centers for Disease Control and Prevention Healthy Days questions after interacting with the digital health application. The proportion of members who reported frequent (14 or more) physically unhealthy, mentally unhealthy, and total unhealthy days during the past 30 days was calculated. The adjusted odds ratio for reporting frequent total unhealthy days was estimated by members' zip code SDI quartile, accounting for member demographic characteristics. Of the 6692 survey respondents, 13.7% reported frequent physically unhealthy days, 20.8% reported frequent mentally unhealthy days, and 29.2% reported frequent total unhealthy days. After adjusting for covariates, members in the highest SDI quartile were 1.2 times more likely to report frequent unhealthy days (P = 0.047) than those in the lowest SDI quartile. The results demonstrate the importance of geographic indices, in the absence of other data, to assist employers in identifying members with potentially higher need of digital health services. It also highlights the feasibility of collecting quality of life measures to identify members who could benefit from timely intervention.

Total hip arthroplasty (THA) is a widely performed surgical procedure in the United States, but disparities in THA outcomes related to hospital-level factors, such as safety-net burden, are underexplored. This study expands on previous research by analyzing multicenter, multistate data from 2015 to 2020 to investigate the impact of hospital safety-net burden-defined as the proportion of services billed to Medicaid and uninsured patients-on THA outcomes. This study is a retrospective analysis using data from the State Inpatient Databases for Florida, Kentucky, Maryland, New York, Washington, New Jersey, and North Carolina. The study cohort included 543,814 inpatient primary THA admissions, with patient demographics, comorbidities, and hospital characteristics analyzed across 3 categories of hospital safety-net burden (low, medium, and high). Generalized linear mixed models assessed the association between safety-net burden and in-hospital mortality and postoperative complications, whereas multilevel negative binomial regression evaluated the impact on hospital length of stay. The study findings indicate that patients undergoing THA at hospitals with high safety-net burden had significantly higher odds of in-hospital mortality (adjusted odds ratio [aOR]: 1.20, 95% confidence interval [CI]: 1.02-1.42), postoperative complications (aOR 1.33, 95% CI 1.20-1.48), and longer hospital stays (adjusted incidence rate ratio 1.15, 95% CI 1.10-1.21) compared with those at low-burden hospitals. These results suggest that hospitals with higher safety-net burden, often serving more vulnerable populations, may have suboptimal perioperative processes and protocols, leading to poorer outcomes. The study underscores the need for targeted interventions to improve THA outcomes in these hospitals.

A payvider organization provides both payer and provider services and has been linked to administrative and patient cost reduction by improving right-sized utilization of patient care services. A geriatric-focused transitional program was offered to patients covered under a value-based care risk contract formed by the payvider relationship of an integrated health system. This article describes a prospective study comparing utilization and cost metrics of patients enrolled in the transitional care program with the goal of analyzing utilization of services to better understand patient behavior patterns and care needs after hospital admission and consecutive enrollment in the program. Patients enrolled in the program incurred lower costs in all categories but home health care than the reference population. The cost avoidance achieved during the study period was estimated to be over $1.1 million. Individuals participating in the program had similar emergency department visit rates during the 90- and 180-days following the hospital as the reference population but had significantly lower inpatient readmissions (7.8% vs. 15.4%) even with a higher average readmission risk score (66.8 vs. 65.5). The implementation of the transitional care program led to reduced costs and more efficient utilization of services than those not enrolled in the program. The payvider relationship allows systems to think proactively about new initiatives and programs that will better serve their communities, especially when identifying groups with high projected costs and service utilization. Patients benefit from the assurance that the services they are receiving are covered by their insurer and their trusted organization.

In response to rising costs associated with providing health care services to Americans over 65 years old, policymakers have called for the expansion of care coordination programs to reduce total spending while improving patient outcomes and provider efficiency. This study uses a Markov Chain model to estimate financial impacts associated with the implementation of a care coordination program across the state of Iowa. Estimates revealed an association between the implementation of the Iowa Return to Community (IRTC) and a reduction in health care service use, which yielded per capita cost savings of $7,920.24 over a 5-year span. Subgroup analysis showed that inclusion of informal care partners enhances these savings, as they contributed to reduced inpatient hospital use and deferred nursing home admissions. The continued expansion of the IRTC appears as a viable strategy to curtail aggregate health care spending while supporting older adults stay at home.

Lung cancer screening (LCS) rates are low, and lung cancer mortality is high in the United States. This report describes a strategy that health systems can use to identify LCS areas of need and engage associated primary care providers and patients in screening. A research team from Jefferson Health (JH), a large, urban health system, used geocoded standardized lung cancer mortality rates (SMRs) to identify zip codes in Philadelphia where lung cancer mortality is high. In addition, health system electronic medical record data were used to identify primary care practices serving these areas. The study also developed an online program to train providers in shared decision making (SDM) about LCS. Finally, primary care leaders were interviewed to learn about training obstacles and opportunities. The JH research team identified 8 high-SMR zip codes and 8 practices with patients from those areas. Working with the American College of Chest Physicians and the National Lung Cancer Round Table, the authors developed a free, online, accredited course to train providers in patient education, values elicitation, and decision support for LCS. Interview analyses with practice leaders encouraged the health system to incentivize provider training and use of SDM tools in practice. Health systems can implement a systematic approach to identify LCS areas of need and train primary care providers to engage patients in SDM about LCS. Research is needed to implement such an approach and evaluate the program's impact on patient engagement, screening, and related outcomes among patients' diverse populations.

Behavioral health integration (BHI) is increasingly implemented to expand capacity to address behavioral health conditions within primary care. Survey and claims data from the evaluation of the Public Hospital Redesign and Incentives in Medi-Cal program were used to examine the relationship between BHI and alcohol-related outcomes among Medicaid patients within 17 public hospitals in California. Key informant survey data measured hospital-level BHI at 3 levels (overall composite, infrastructure, and process domains, 10 themes). Multilevel logistic regression models estimated the relationship between BHI and outcomes indicating receipt of appropriate alcohol-related care (any primary care visit, any detoxification, timely initiation, timely engagement) and acute care (any emergency department [ED] visit or hospitalization, classified as alcohol-related or all-cause) in the year following an alcohol-related index encounter. Of 6196 patients, some had an alcohol-related primary care visit (33%), detoxification (16%), timely initiation (14%), or engagement in treatment (7%). ED visits resulting in discharge were more common (40% alcohol-related, 64% all-cause) than hospitalizations (15% alcohol-related, 26% all-cause). Controlling for patient-level characteristics, no significant relationships between overall BHI and these outcomes were observed. However, greater BHI infrastructure was associated with alcohol-related (odds ratio [OR] 1.86, 95% confidence interval [CI] 1.14-3.05) and all-cause hospitalization (OR 1.25, 95% CI 1.01-1.55). Associations emerged between BHI themes (eg, related to support of providers) and greater likelihood of alcohol-related detoxification, primary care visit, timely initiation, and acute care utilization. Findings suggest that implementing specific BHI components may improve receipt of alcohol-related treatment, and warrant future research into these relationships.