Abstract: Objectives To describe hospital utilization and costs associated with preterm or low birth weight births (preterm/LBW) by payer prior to implementation of the Affordable Care Act and to identify areas for improvement in the quality of care received among preterm/LBW infants. Methods Hospital utilization—defined as mean length of stay (LOS, days), secondary diagnoses for birth hospitalizations, primary diagnoses for rehospitalizations, and transfer status—and costs were described among preterm/LBW infants using the 2009 Nationwide Inpatient Sample. Results Approximately 9.1 % of included hospitalizations (n = 4,167,900) were births among preterm/LBW infants; however, these birth hospitalizations accounted for 43.4 % of total costs. Rehospitalizations of all infants occurred at a rate of 5.9 % overall, but accounted for 22.6 % of total costs. This pattern was observed across all payer types. The prevalence of rehospitalizations was nearly twice as high among preterm/LBW infants covered by Medicaid (7.6 %) compared to commercially-insured infants (4.3 %). Neonatal transfers were more common among preterm/LBW infants whose deliveries and hospitalizations were covered by Medicaid (7.3 %) versus commercial insurance (6.5 %). Uninsured/self-pay preterm and LBW infants died in-hospital during the first year of life at a rate of 91 per 1000 discharges—nearly three times higher than preterm and LBW infants covered by either Medicaid (37 per 1000) or commercial insurance (32 per 1000). Conclusions When comparing preterm/LBW infants whose births were covered by Medicaid and commercial insurance, there were few differences in length of hospital stays and costs. However, opportunities for improvement within Medicaid and CHIP exist with regard to reducing rehospitalizations and neonatal transfers.

INTRODUCTION:

We examined the cost-effectiveness of smoking cessation integrated with treatment for post-traumatic stress disorder (PTSD).

METHODS:

Smoking veterans receiving care for PTSD (N = 943) were randomized to care integrated with smoking cessation versus referral to a smoking cessation clinic. Smoking cessation services, health care cost and utilization, quality of life, and biochemically-verified abstinence from cigarettes were assessed over 18-months of follow-up. Clinical outcomes were combined with literature on changes in smoking status and the effect of smoking on health care cost, mortality, and quality of life in a Markov model of cost-effectiveness over a lifetime horizon. We discounted cost and outcomes at 3% per year and report costs in 2010 US dollars.

RESULTS:

The mean of smoking cessation services cost was $1286 in those randomized to integrated care and $551 in those receiving standard care (P < .001). There were no significant differences in the cost of mental health services or other care. After 12 months, prolonged biochemically verified abstinence was observed in 8.9% of those randomized to integrated care and 4.5% of those randomized to standard care (P = .004). The model projected that Integrated Care added $836 in lifetime cost and generated 0.0259 quality adjusted life years (QALYs), an incremental cost-effectiveness ratio of $32 257 per QALY. It was 86.0% likely to be cost-effective compared to a threshold of $100 000/QALY.

CONCLUSIONS:

Smoking cessation integrated with treatment for PTSD was cost-effective, within a broad confidence region, but less cost-effective than most other smoking cessation programs reported in the literature.

Abstract: Community hospital stays in 12 states during 2008–2009 were analyzed to determine predictors of 12-month hospital readmission and emergency department (EDs) revisits among persons with a mental health or substance abuse diagnosis. Probabilities of hospital readmission and of ED revisits were modeled as functions of patient demographics, insurance type, number of prior-year hospital stays, diagnoses and other characteristics of the initial stay, and hospital characteristics. Alcohol or drug dependence, dementias, psychotic disorders, autism, impulse control disorders, and personality disorders were most strongly associated with future inpatient admission or ED revisits within 12 months of initial encounter. Insurance type, including uninsured status, were highly significant (p < .01) predictors of both readmission and ED revisits.

Outpatient treatment for substance use disorders promotes recovery and reduces readmission (1,2). We measured the annual percentage of individuals aged 13–64 who were covered under employer health plans and who received outpatient treatment for a substance use disorder within 30 days of a related inpatient stay.

All data (2001–2012) were from the Truven Health MarketScan Commercial Claims and Encounters Database, which includes claims of employed individuals and dependents from all U.S. Census divisions. Data were weighted to be nationally representative of individuals with employer-sponsored insurance, 48% of the U.S. population in 2012 (3). The denominator was number of claims that met these criteria: discharge from an inpatient acute or residential facility, age 13–64, primary ICD-9-CM discharge diagnosis of drug or alcohol abuse or dependence, discharge date between January 1 and December 1 (to allow for 30 days of December follow-up), discharge status was not death or transfer to another facility, and continued plan enrollment for 30 days after discharge. The numerator was number of denominator stays for which there was an outpatient visit related to a substance use disorder within 30 days, defined by a primary ICD-9-CM diagnosis or code pertaining to mental health or substance abuse. We captured both types of diagnoses because substance use disorders are often recorded as mental disorders.

Since 2003 the rate at which individuals have received outpatient treatment for a substance use disorder within 30 days of a related inpatient stay has alternated between plateaus and periods of increase (Figure 1). The rate has grown at least a little every year since 2005, reaching 66.1% in 2012.

A single measure in the Healthcare Effectiveness Data and Information Set (HEDIS) tracks the rate of outpatient follow-up within seven or 30 days of an inpatient psychiatric discharge. There is no exact analog for substance use disorders. The most similar HEDIS measure, “engagement” in substance use disorder treatment, requires two outpatient visits related to a substance use disorder within 30 days of “initiation,” which does not require a hospital stay. The dissimilarity of the two follow-up measures makes it difficult to compare outcomes for persons with substance use disorders and those with other mental disorders. A 30-day follow-up measure for substance use disorders would promote care delivery, research, policy making, and advocacy. The figure presented here represents a first step toward that goal.

Abstract:

I take advantage of regulatory and pricing dynamics in Medicare Part D to explore interactions among adverse selection, inertia, and regulation. I first document novel evidence of adverse selection and switching frictions within Part D using detailed administrative data. I then estimate a contract choice and pricing model that quantifies the importance of inertia for risk-sorting. I find that in Part D switching costs help sustain an adversely-selected equilibrium. I also estimate that active decision-making in the existing policy environment could lead to a substantial gain in annual consumer surplus of on average $400-$600 per capita—20%-30% of average annual spending.

How health technology interacts with underlying cultural norms may have unanticipated consequences for development. Recent policy debates on closing the gender gap in developing countries have focused on affirmative action and economic growth, but there are reasons to expect that policies targeting the health of young children might also be effective. I investigate a national vaccination campaign targeting under-fi…ve children in Turkey, and document gains in human capital among age-eligible children of both sexes but spillover effects that accrue exclusively to older girl siblings. These …findings are consistent with predictions from a standard intrahousehold model of time allocation in the presence of gender norms regarding the division of household labor and suggest technologies and policies that improve the health of young children may have the added benefi…t of improving educational outcomes for their older sisters.