30-day hospital readmission is a long standing medical problem that affects patients' morbidity and mortality and costs billions of dollars annually. Recently, machine learning models have been created to predict risk of inpatient readmission for patients with specific diseases, however no model exists to predict this risk across all patients. We developed a bi-directional Long Short Term Memory (LSTM) Network that is able to use readily available insurance data (inpatient visits, outpatient visits, and drug prescriptions) to predict 30 day re-admission for any admitted patient, regardless of reason. The top-performing model achieved an ROC AUC of 0.763 (0.011) when using historical, inpatient, and post-discharge data. The LSTM model significantly outperformed a baseline random forest classifier, indicating that understanding the sequence of events is important for model prediction. Incorporation of 30-days of historical data also significantly improved model performance compared to inpatient data alone, indicating that a patients clinical history prior to admission, including outpatient visits and pharmacy data is a strong contributor to readmission. Our results demonstrate that a machine learning model is able to predict risk of inpatient readmission with reasonable accuracy for all patients using structured insurance billing data. Because billing data or equivalent surrogates can be extracted from sites, such a model could be deployed to identify patients at risk for readmission before they are discharged, or to assign more robust follow up (closer follow up, home health, mailed medications) to at-risk patients after discharge.

Importance: To improve health care price transparency and promote cost-conscious selection of health care organizations and practitioners, the Centers for Medicare & Medicaid Services (CMS) required that hospitals share payer-specific negotiated prices for selected shoppable health services by January 2021. While this regulation improves price transparency, it is unclear whether disclosed prices reflect total costs of care, since many hospital-based services are delivered and billed separately by independent practitioners or other health care entities. Objective: To assess the extent to which prices disclosed under the new hospital price transparency regulation are correlated with total costs of care among commercially insured individuals. Design, Setting, and Participants: This cross-sectional study used a large database of commercial claims from 2018 to analyze encounters at US hospitals for shoppable health care services for which price disclosure is required by CMS. Data were analyzed from November 2020 to February 2021. Exposures: Whether the service was billed by the hospital or another entity. Main Outcomes and Measures: Outcomes of interest were the percentage of encounters with at least 1 service billed by an entity other than the hospital providing care, number of billing entities, amounts billed by nonhospital entities, and the correlation between hospital and nonhospital reimbursements. Results: The study analyzed 4545809 encounters for shoppable care. Independent health care entities were involved in 7.6% (95% CI, 6.7% to 8.4%) to 42.4% (95% CI, 39.1% to 45.6%) of evaluation and management encounters, 15.9% (95% CI, 15.8% to 16%) to 22.2% (95% CI, 22% to 22.4%) of laboratory and pathology services, 64.9% (95% CI, 64.2% to 65.7%) to 87.2% (95% CI, 87.1% to 87.3%) of radiology services, and more than 80% of most medicine and surgery services. The median (IQR) reimbursement of independent practitioners ranged from $61 ($52-$102) to $412 ($331-$466) for evaluation and management, $5 ($4-$6) to $7 ($4-$12) for laboratory and pathology, $26 ($20-$32) to $210 ($170-$268) for radiology, and $47 ($21-$103) to $9545 ($7750-$18277) for medicine and surgery. The reimbursement for services billed by the hospital was not strongly correlated with the reimbursement of independent clinicians, ranging from r = -0.11 (95% CI, -0.69 to 0.56) to r = 0.53 (95% CI, 0.13 to 0.78). Conclusions and Relevance: This cross-sectional study found that independent practitioners were frequently involved in the delivery of shoppable hospital-based care, and their reimbursement may have represented a substantial portion of total costs of care. These findings suggest that disclosed hospital reimbursement was usually not correlated with total cost of care, limiting the potential benefits of the hospital price transparency rule for improving consumer decision-making..

This cross-sectional study assesses how much monthly cost-sharing limits, as opposed to annual limits, could reduce out-of-pocket costs for commercially insured patients in the US.

Background: Recent emphasis on value based care and population management, such as Accountable Care Organizations in the United States, promote patient navigation to improve the quality of care and reduce costs. Evidence supporting the efficacy of patient navigation for chronic disease care is limited. The objective of this study was to evaluate the effect of a patient navigation program on medical and administrative outcomes among patients with diabetes in an urban, safety-net hospital clinic setting. Methods: A retrospective cohort study with pre- and post-intervention periods was conducted. Eligible patients were those with A1C ≥ 8.5% and at least one appointment no-show in the previous 12 months. The intervention and reference groups were balanced on observed characteristics and baseline outcome levels using propensity score matching. The effect of patient navigation was isolated using the difference-in-differences approach. Primary outcomes were A1C, low-density lipoprotein cholesterol, triglycerides, random urine microalbumin, the number of scheduled appointments, clinic visits, emergency visits, and inpatient stays, and the percentage of arrivals, cancellations, and no-shows to scheduled appointments. Results: Of 797 eligible patients, 328 entered the navigation program. Matching reduced the sample size to 392 individuals (196 in each group). Patient navigation resulted in improved A1C (-1.1 percentage points; p < .001), more scheduled appointments (+ 5.3 per year; p < .001), more clinic visits (+6.4 per year; p < .001), more arrivals to scheduled appointments (+7.4 percentage points; p =.009) and fewer no-shows (-9.8 percentage points; p < .001). Conclusions: Navigation was associated with improved glycemic control and better clinic engagement among patients with diabetes. Further research is important to identify what features of navigation in diabetes care are critical to achieving success and to understand navigators' role in other settings.

OBJECTIVE: The aim of this study was to investigate the value of coformulated Tenofovir disoproxil fumarate and emtricitabine (TDF/FTC) for preexposure prophylaxis (PrEP) for conception in the U.S. and to identify scenarios in which 'Undetectable = Untransmittable' (U = U) may not be adequate, and rather, PrEP or assisted reproduction would improve outcomes. DESIGN: We developed a Markov cohort simulation model to estimate the incremental benefits and cost-effectiveness of PrEP compared with alternative safer conception strategies, including combination antiretroviral therapy (cART) alone for the HIV-infected partner and assisted reproductive technologies. We modelled various scenarios in which HIV RNA suppression in the male partner was less than perfect. SETTING: U.S. healthcare sector perspective. PARTICIPANTS: Serodiscordant couples in the U.S. was composed of an HIV-infected male and HIV-uninfected female seeking conception. INTERVENTION: Economic analysis. MAIN OUTCOME MEASURE(S): Cumulative risks of HIV transmission to women and babies, maternal life expectancy, discounted quality-adjusted life years (QALY), discounted lifetime medical costs and incremental cost-effectiveness ratios. RESULTS: cART with condomless intercourse limited to ovulation was the preferred HIV prevention strategy among women seeking to conceive with an HIV-infected partner who is HIV-suppressed. PrEP was not cost-effective for women who had partners who were virologically suppressed. When the probability of male partner HIV suppression was low and we assumed generic pricing of PrEP, PrEP was cost-effective, and sometimes even cost-saving compared with cART alone. CONCLUSION: From a U.S. healthcare sector perspective, when the male partner was not reliably suppressed, PrEP became economically attractive, and in some cases, cost-saving.