Explore the vast range of titles available.

Background: Sickle cell disease (SCD) is one of the most common inherited diseases in the U.S. and disproportionately affects African Americans and Hispanic Americans. People with SCD suffer acute pain crises and chronic pain throughout the lifespan and since disease cure is often unattainable, most turn to intensive and costly disease management, often with particularly high emergency care use. Hydroxyurea (HU), an inexpensive therapeutic drug, has proven to be effective at reducing pain crises. However, HU is underutilized for a variety of reasons including lack of insurance coverage, misperceptions and negative beliefs, and lack of provider awareness of guidelines. This paper aims to quantify how healthcare utilization, spending, and pain crises in SCD patients change before and after prescription of hydroxyurea, as well as estimate the potential averted burden if those not prescribed hydroxyurea were to be placed on the treatment regimen.Methods: MarketScan Commercial Claims and Encounters databases person-specific information on healthcare utilization, healthcare spending, prescriptions, and diagnoses from 2016 to 2019 was aggregated into 90-day periods. 2-tailed t-tests were used to compare their mean outcomes in care utilization, spending, and pain crises per 90 days, pre and post receipt of that prescription and negative binomial regressions were run to estimate mean outcomes over 5, 90-day time periods. Control SCD patients who had never received a prescription for hydroxyurea were matched 1:1 using propensity score matching to the previously established treated group on age, sex and genotype. T-tests and regressions were repeated to compare mean outcomes between control and treated groups. The ratio of change in outcome pre and post HU prescription among treated patients was then applied to the control group to estimate the potential averted care utilization, pain, and spending if the control received the treatment.Results: 7,793 patients between the ages of 0 and 65 with a diagnosis of SCD from ICD-10 with a minimum of 1 year data coverage were selected, of whom 25.7% (n =2008) received a prescription for hydroxyurea at any point. 631 patients met the inclusion criteria for at least 90 days prior to initial prescription of hydroxyurea and 360 days post prescription. Prior to initiating treatment, hydroxyurea-treated patients had worse outcomes per 90 days with an estimated total of 5.4 (95% confidence interval 5.0-5.9) visits,$15,691 (13,088 – 18,812) net pay, and 1.2 (1.1-1.4) pain crises, compared to matched controls who in cross section had 1.7 (1.6-1.8) total visits, $ 4,319 (3,750 – 4,975) total net pay, and 0.3 (0.3-0.3) pain crises. Though outcomes among those initiating hydroxyurea remained significantly higher than controls, across care utilization, spending, and pain crises, patient 90-day outcomes were significantly lower post hydroxyurea initiation, in all types of care except outpatient visits and outpatient net pay, and the greatest declines were in inpatient settings.Interpretation: SCD patients have high care utilization, spending, and pain outcomes. Those treated with hydroxyurea may experience greater disease severity, but this study indicates that if hydroxyurea were more widely utilized, their outcomes may be improved, and a large health and financial burden could be averted.

Little is known about health care spending variation across the US for recent years. To estimate health spending by state and payer, we combined data from the government's State Health Expenditure Accounts, which have estimates through 2014, with other primary data on spending. In 2019 state-specific per person spending ranged from$7,250 to $ 14,500. After adjustment for inflation, annualized per person spending growth for each state ranged from 1.0 percent in Washington, D.C., to 4.2 percent in South Dakota between 2013 and 2019. The factors that explained the most variation across states were incomes (25.3 percent) and consumer prices (21.7 percent). Medicaid expansion was associated with increases in total spending per person, although the median of spending in expansion states showed slower growth in out-of-pocket spending than the median in nonexpansion states. Contemporary estimates of state health spending are valuable for tracking divergent expenditure trajectories in the US and assessing the associated factors.

Under-reporting of count data poses a major roadblock for prediction and inference. In this paper, we focus on the Pogit model, which deconvolves the generating Poisson process from the censuring process controlling under-reporting using a generalized linear modeling framework. We highlight the limitations of the Pogit model and address them by adding constraints to the estimation framework. We also develop uncertainty quantification techniques that are robust to model mis-specification. Our approach is evaluated using synthetic data and applied to real healthcare datasets, where we treat in-patient data as `reported' counts and use held-out total injuries to validate the results. The methods make it possible to separate the Poisson process from the under-reporting process, given sufficient expert information. Codes to implement the approach are available via an open source Python package.