Explore the vast range of titles available.

Coronavirus disease 2019 (COVID-19) spurred a rapid rise in telemedicine, but it is unclear how use has varied by clinical and patient factors during the pandemic. We examined the variation in total outpatient visits and telemedicine use across patient demographics, specialties, and conditions in a database of 16.7 million commercially insured and Medicare Advantage enrollees from January to June 2020. During the pandemic, 30.1 percent of all visits were provided via telemedicine, and the weekly number of visits increased twenty-three-fold compared with the prepandemic period. Telemedicine use was lower in communities with higher rates of poverty (31.9 percent versus 27.9 percent for the lowest and highest quartiles of poverty rate, respectively). Across specialties, the use of any telemedicine during the pandemic ranged from 68 percent of endocrinologists to 9 percent of ophthalmologists. Across common conditions, the percentage of visits provided during the pandemic via telemedicine ranged from 53 percent for depression to 3 percent for glaucoma. Higher rates of telemedicine use for common conditions were associated with smaller decreases in total weekly visits during the pandemic.

Objectives. To examine whether growing use of telemental health (TMH) has reduced the rural–urban gap in specialty mental health care use in the United States. Methods. Using 2010–2017 Medicare data, we analyzed trends in the rural–urban difference in rates of specialty visits (in-person and TMH). Results. Among rural beneficiaries diagnosed with schizophrenia or bipolar disorder, TMH use grew by 425% over the 8 years and, in higher-use rural areas, accounted for one quarter of all specialty mental health visits in 2017. Among patients with schizophrenia or bipolar disorder, TMH visits differentially grew in rural areas by 0.14 visits from 2010 to 2017. This growth partially offset the 0.42-visit differential decline in in-person visits in rural areas. In net, the gap between rural and urban patients in specialty visits was larger by 2017. Conclusions. TMH has improved access to specialty care in rural areas, particularly for individuals diagnosed with schizophrenia or bipolar disorder. While growth in TMH use has been insufficient to eliminate the overall rural–urban difference in specialty care use, this difference may have been larger if not for TMH. Public Health Implications. Targeted policy to extend TMH to underserved areas may help offset declines in in-person specialty care.

To compare rates and analyze health facility determinants of emergency department visits and hospitalizations for ambulatory care-sensitive conditions (ACSCs) among Medicaid patients by geographical location. Retrospective cross-sectional analysis of 48.3 million patients receiving Medicaid and their acute care visits across 34 states and the District of Columbia in 2019. Descriptive analyses of county-level variations in emergency department visits and hospitalizations (acute care) for ACSCs, and multivariate regressions of proximity to and density of health facility infrastructure as correlates to utilization and spending. Regression models were adjusted for county-level poverty rates, chronic disease rates, and state fixed effects. Among the studied patient population receiving Medicaid, nearly 40% of acute care visits were for ACSCs, with variations across and within states. Rates ranged from 17.8 per 1000 member-months in Vermont to 39.0 in Mississippi, and from 5.9 to 77.9 between counties within states. Longer distances to the nearest urgent care center and primary care shortage area designation correlated to higher acute care visits for ACSCs (+4.3 per 1000 member-months for every 100 miles; 95% CI, 2.9-5.7; P < .001; +1.5 per 1000 member-months if shortage area; 95% CI, 0.4-2.6; P = .006). Counties with more rural health clinics had fewer acute care visits for ACSCs (-3.4 fewer visits per rural clinic per 1000 population; 95% CI, -4.6 to -2.2; P < .001). Among 6 states with additional spending data, 4.2% of total Medicaid spending was attributable to acute care visits for ACSCs. Our evaluation revealed more than 13-fold variation in acute care utilization for ACSCs between Medicaid counties within the same state. Proximity to urgent care facilities and density of rural health clinics were major explanatory variables for these variations, underscoring the importance of local health infrastructure in reducing acute care utilization for ACSCs.

Patients receiving Medicaid often experience social risk factors for poor health and limited access to primary care, leading to high utilization of emergency departments and hospitals (acute care) for non-emergent conditions. As programs proactively outreach Medicaid patients to offer primary care, they rely on risk models historically limited by poor-quality data. Following initiatives to improve data quality and collect data on social risk, we tested alternative widely-debated strategies to improve Medicaid risk models. Among a sample of 10 million patients receiving Medicaid from 26 states and Washington DC, the best-performing model tripled the probability of prospectively identifying at-risk patients versus a standard model (sensitivity 11.3% [95% CI 10.5, 12.1%] vs 3.4% [95% CI 3.0, 4.0%]), without increasing “false positives” that reduce efficiency of outreach (specificity 99.8% [95% CI 99.6, 99.9%] vs 99.5% [95% CI 99.4, 99.7%]), and with a ~ tenfold improved coefficient of determination when predicting costs (R 2 : 0.195–0.412 among population subgroups vs 0.022–0.050). Our best-performing model also reversed the lower sensitivity of risk prediction for Black versus White patients, a bias present in the standard cost-based model. Our results demonstrate a modeling approach to substantially improve risk prediction performance and equity for patients receiving Medicaid.

Low-income populations have disproportionately low completion of recommended healthcare services, from missed vaccinations to cancer screenings. While machine learning models help identify high-risk patients for targeted treatment, they have rarely been evaluated for quality measure gap completion—or among low-income populations underrepresented in typical datasets. Analyzing 14.2 million Medicaid recipients—including those excluded from electronic health records and without prior utilization—we developed models to predict gaps in nine nationally adopted quality measures, including preventive care and chronic disease management. Using clinical data to prioritize outreach, the clinical-only model improved accuracy by 32.5 percentage points (pp) over non-predictive methods such as alphabetical calling or birthday reminders (AUROC: 0.88, F1-score: 0.69). Incorporating social determinants of health data further improved performance by 2.0pp in accuracy (to 84.5%) and increased F1-score by 5.0pp (to 0.74), with no change in AUROC (area under the receiver operating characteristic curve). Compared to the clinical-only model, the SDoH model also reduced pre-existing Black–White disparities in prediction accuracy. Model performance was especially sensitive to SDoH factors like healthcare workforce and facility availability.

Population health initiatives often rely on cold outreach to close gaps in preventive care, such as overdue screenings or immunizations. Tailoring messages to diverse patient populations remains challenging, as traditional A/B testing requires large sample sizes to test only two alternative messages. With increasing availability of large language models (LLMs), programs can utilize tiered testing among both LLM and manual human agents, presenting the dilemma of identifying which patients need different levels of human support to cost-effectively engage large populations. Using microsimulations, we compared both the statistical power and false positive rates of A/B testing and Sequential Multiple Assignment Randomized Trials (SMART) for developing personalized communications across multiple effect sizes and sample sizes. SMART showed better cost-effectiveness and net benefit across all scenarios, but superior power for detecting heterogeneous treatment effects (HTEs) only in later randomization stages, when populations were more homogeneous and subtle differences drove engagement differences.

ObjectiveTo examine the proportion of healthcare visits are delivered by nurse practitioners and physician assistants versus physicians and how this has changed over time and by clinical setting, diagnosis, and patient demographics.DesignCross-sectional time series study.SettingNational data from the traditional Medicare insurance program in the USA.ParticipantsOf people using Medicare (ie, those older than 65 years, permanently disabled, and people with end stage renal disease), a 20% random sample was taken.Main outcome measuresThe proportion of physician, nurse practitioner, and physician assistant visits in the outpatient and skilled nursing facility settings delivered by physicians, nurse practitioners, and physician assistants, and how this proportion varies by type of visit and diagnosis.ResultsFrom 1 January 2013 to 31 December 2019, 276 million visits were included in the sample. The proportion of all visits delivered by nurse practitioners and physician assistants in a year increased from 14.0% (95% confidence interval 14.0% to 14.0%) to 25.6% (25.6% to 25.6%). In 2019, the proportion of visits delivered by a nurse practitioner or physician assistant varied across conditions, ranging from 13.2% for eye disorders and 20.4% for hypertension to 36.7% for anxiety disorders and 41.5% for respiratory infections. Among all patients with at least one visit in 2019, 41.9% had one or more nurse practitioner or physician assistant visits. Compared with patients who had no visits from a nurse practitioner or physician assistant, the likelihood of receiving any care was greatest among patients who were lower income (2.9% greater), rural residents (19.7%), and disabled (5.6%).ConclusionThe proportion of visits delivered by nurse practitioners and physician assistants in the USA is increasing rapidly and now accounts for a quarter of all healthcare visits.

This commentary notes the superiority of targeting rising-risk patients rather than high-cost claimants for Medicaid cost containment based on analysis of 13.1 million beneficiaries across 15 states. In 2019, spending for rising-risk patients (13.6% of sample) increased by 98.5% whereas spending for high-cost claimants (0.64%) decreased by 41.6%. Significantly, 54% of high-cost claimants in the first half of 2019 fell below the cost threshold in the second half of the year, and 50% of new high-cost claimants were previously identified as rising risk. Our findings reveal the limitations of focusing solely on high-cost claimants, whose costs naturally decrease due to regression to the mean. We argue that Medicaid programs should shift from reactive, cost-management interventions to proactive, prevention-oriented outreach, particularly as new predictive algorithms become more sensitive and specific. Early identification of and intervention for rising-risk patients is a more effective way to prevent the progression of chronic conditions and manage associated costs than attempting to reduce extreme utilization, which tends to decrease naturally over time.

Nurse practitioners (NPs) and physician assistants (PAs) represent a growing share of the health care workforce, but much of the care they provide cannot be observed in claims data because of indirect (or incident to) billing, a practice in which visits provided by an NP or PA are billed by a supervising physician. If NPs and PAs bill directly for a visit, Medicare and many private payers pay 85 percent of what is paid to a physician for the same service. Some policy makers have proposed eliminating indirect billing, but the possible impact of such a change is unknown. Using a novel approach that relies on prescriptions to identify indirectly billed visits, we estimated that the number of all NP or PA visits in fee-for-service Medicare data billed indirectly was 10.9 million in 2010 and 30.6 million in 2018. Indirect billing was more common in states with laws restricting NPs scope of practice. Eliminating indirect billing would have saved Medicare roughly $194 million in 2018, with the greatest decrease in revenue seen among smaller primary care practices, which are more likely to use this form of billing.