Tracking Substance Use Complications: A Collaborative Analysis of Public Health and Academic Medical Center Records on Drug Use-Associated Infective Endocarditis.

BACKGROUND: Healthcare systems and public health agencies use different methods to measure the impact of substance use (SU) on population health. We studied the ability of systems to accurately capture data on drug use-associated infective endocarditis (DUA-IE). METHODS: We conducted a retrospective analysis of patients with IE discharge diagnosis from an academic medical center, 2011-2017, comparing data from hospital Electronic Health Record (EHR) to State Uniform Hospital Discharge Data Set (UHDDS). To identify SU we developed a composite measure. RESULTS: EHR identified 472 IE discharges (430 of these were captured in UHDDS); 406 (86.0%) were correctly coded based on chart review. IE discharges increased from 57 to 92 (62%) from 2012 to 2017. Hospitalizations for the subset of DUA-IE identified by any measure of SU increased from 10 to 54 (440%). Discharge diagnosis coding identified 128 (60.7%) of total DUA-IE hospitalizations. The composite measure identified an additional 65 (30.8%) DUA-IE hospitalizations and chart review an additional 18 (8.5%). CONCLUSIONS: The failure of discharge diagnosis coding to identify DUA-IE in 40% of hospitalizations demonstrates the need for better systems to capture the impact of SU. Collaborative data sharing could help improve surveillance responsiveness to address an emerging public health crises.

Differences in utilization of Intracytoplasmic sperm injection (ICSI) within human services (HHS) regions and metropolitan megaregions in the U.S.

BACKGROUND: Anecdotal evidence suggests that US practice patterns for ART differ by geographical region. The purpose of this study was to determine whether use of ICSI differs by region and to evaluate whether these rates are correlated with differences in live birth rates. METHODS: Public data for 2012 were obtained from the Centers for Disease Control and Prevention. Clinics with ≥100 fresh, non-donor cycles were grouped by 10 nationally recognized Department of Health & Human Services regions and 11 metropolitan Megaregions and were compared for use of ICSI, frequency of male factor infertility, and live birth rate in women <35 years. RESULTS: There were 274 clinics in the Health & Human Services regions and 247 in the Megaregions. ICSI utilization rates in Health & Human Services groups ranged between 52.5-78.2% (P < 0.0001). Live birth rates per cycle in women <35 years differed (34.1-47.6%; P < 0.0001) but did not correlate with rates of ICSI (R2 = 0.2096; P = 0.18) per cycle. For Megaregions, rates of ICSI per cycle differed (63.4%-93.5%, P < 0.0001) as did live birth rates per cycle for women <35 (36.0%-59.0%, P = 0.001) but there was only minimal correlation between them (R2 = 0.5347; P = 0.01). Highest rates of ICSI occurred in Front Range (93.5%) and Gulf Coast (83.1%) Megaregions. Lowest rates occurred in the Northeast (63.4%) and Florida (64.8%) Megaregions. Male factor infertility rates did not differ across regions. CONCLUSIONS: ICSI utilization and live birth rates per cycle for each clinic group were significantly different across geographical regions of the U.S. However, higher ICSI utilization rate was not associated with higher rates of male factor infertility nor were they strongly correlated with higher live birth rates per cycle. Studies are needed to understand factors that may influence ICSI overutilization in the U.S.

Engagement with health agencies on twitter.

OBJECTIVE: To investigate factors associated with engagement of U.S. Federal Health Agencies via Twitter. Our specific goals are to study factors related to a) numbers of retweets, b) time between the agency tweet and first retweet and c) time between the agency tweet and last retweet. METHODS: We collect 164,104 tweets from 25 Federal Health Agencies and their 130 accounts. We use negative binomial hurdle regression models and Cox proportional hazards models to explore the influence of 26 factors on agency engagement. Account features include network centrality, tweet count, numbers of friends, followers, and favorites. Tweet features include age, the use of hashtags, user-mentions, URLs, sentiment measured using Sentistrength, and tweet content represented by fifteen semantic groups. RESULTS: A third of the tweets (53,556) had zero retweets. Less than 1% (613) had more than 100 retweets (mean  = 284). The hurdle analysis shows that hashtags, URLs and user-mentions are positively associated with retweets; sentiment has no association with retweets; and tweet count has a negative association with retweets. Almost all semantic groups, except for geographic areas, occupations and organizations, are positively associated with retweeting. The survival analyses indicate that engagement is positively associated with tweet age and the follower count. CONCLUSIONS: Some of the factors associated with higher levels of Twitter engagement cannot be changed by the agencies, but others can be modified (e.g., use of hashtags, URLs). Our findings provide the background for future controlled experiments to increase public health engagement via Twitter.

Who benefits most from Head Start? Using latent class moderation to examine differential treatment effects.

Head Start (HS) is the largest federally funded preschool program for disadvantaged children. Research has shown relatively small impacts on cognitive and social skills; therefore, some have questioned its effectiveness. Using data from the Head Start Impact Study (3-year-old cohort; N = 2,449), latent class analysis was used to (a) identify subgroups of children defined by baseline characteristics of their home environment and caregiver and (b) test whether the effects of HS on cognitive, and behavioral and relationship skills over 2 years differed across subgroups. The results suggest that the effectiveness of HS varies quite substantially. For some children there appears to be a significant, and in some cases, long-term, positive impact. For others there is little to no effect.

Association of a higher density of specialist neuroscience providers with fewer deaths from stroke in the United States population.

OBJECT: Stroke is a leading cause of death and disability. Given that neurologists and neurosurgeons have special expertise in this area, the authors hypothesized that the density of neuroscience providers is associated with reduced mortality rates from stroke across US counties. METHODS: This is a retrospective review of the Area Resource File 2009-2010, a national county-level health information database maintained by the US Department of Health and Human Services. The primary outcome variable was the 3-year (2004-2006) average in cerebrovascular disease deaths per million population for each county. The primary independent variable was the combined density of neurosurgeons and neurologists per million population in the year 2006. Multiple regression analysis was performed, adjusting for density of general practitioners (GPs), urbanicity of the county, and socioeconomic status of the residents of the county. RESULTS: In the 3141 counties analyzed, the median number of annual stroke deaths was 586 (interquartile range [IQR] 449-754), the median number of neuroscience providers was 0 (IQR 0-26), and the median number of GPs was 274 (IQR 175-410) per million population. On multivariate adjusted analysis, each increase of 1 neuroscience provider was associated with 0.38 fewer deaths from stroke per year (p < 0.001) per million population. Rural location (p < 0.001) and increased density of GPs (p < 0.001) were associated with increases in stroke-related mortality. CONCLUSIONS: Higher density of specialist neuroscience providers is associated with fewer deaths from stroke. This suggests that the availability of specialists is an important factor in survival after stroke, and underlines the importance of promoting specialist education and practice throughout the country.

How should risk adjustment data be collected?

Risk adjustment has broad general application and is a key part of the Patient Protection and Affordable Care Act (ACA). Yet, little has been written on how data required to support risk adjustment should be collected. This paper offers analytical support for a distributed approach, in which insurers retain possession of claims but pass on summary statistics to the risk adjustment authority as needed. It shows that distributed approaches function as well as or better than centralized ones-where insurers submit raw claims data to the risk adjustment authority-in terms of the goals of risk adjustment. In particular, it shows how distributed data analysis can be used to calibrate risk adjustment models and calculate payments, both in theory and in practice--drawing on the experience of distributed models in other contexts. In addition, it explains how distributed methods support other goals of the ACA, and can support projects requiring data aggregation more generally. It concludes that states should seriously consider distributed methods to implement their risk adjustment programs.

Taking up or turning down: new estimates of household demand for employer-sponsored health insurance.

This study provides new estimates of demand for employer-sponsored health insurance, using the 1997-2001 linked Household Component-Insurance Component of the Medical Expenditure Panel Survey (MEPS). Our focus is on households' decisions to take up coverage through a worker's employer. We found a significant inverse relationship between the out-of-pocket premium and the probability of taking up coverage, with the price effect considerably larger when we used instrumental variables methods to account for endogenous out-of-pocket premiums. Additionally, workers in families with more children eligible for Medicaid were less likely to take up coverage.

National trends in spinal arteriovenous malformations.

OBJECT: Spinal arteriovenous malformations (AVMs) are rare and understudied vascular lesions that cause neurological insult by mass effect, venous obstruction, and vascular steal. These lesions are challenging entities to treat because of their complicated anatomy and physiology. Current management options include open microsurgery, endovascular embolization, and stereotactic radiosurgery. METHODS: Our study used the National Inpatient Sample database to analyze outcome data for spinal AVMs treated nationwide over an 11-year period from 1995 through 2006. Trends in procedural management, hospital course, and epidemiology of spinal AVMs are investigated. RESULTS: Annually, an average of 300 patients presented with spinal AVMs requiring hospital treatment. The average length of hospital stay for this treatment has declined from more than 9 days in 1995 to 6 days in 2006. However, the average cost of a hospital stay has increased from < $30,000 to nearly $70,000. Whereas one-half of spinal AVMs were treated operatively in 1995, one-third were managed operatively in 2006. CONCLUSIONS: Spinal AVMs are being increasingly treated by endovascular, radiosurgical, or combined means. A discussion of modern strategies to treat these disorders is presented.

Billions for biodefense: federal agency biodefense funding, FY2001-FY2005.

Over the past several years, the United States government has spent substantial resources on preparing the nation against a bioterrorist attack. This article analyzes the civilian biodefense funding by the federal government from fiscal years 2001 through 2005, specifically analyzing the budgets and allocations for biodefense at the Department of Health and Human Services, the Department of Homeland Security, the Department of Defense, the Department of Agriculture, the Environmental Protection Agency, the National Science Foundation, and the Department of State. In total, approximately $14.5 billion has been funded for civilian biodefense through FY2004, with an additional $7.6 billion in the President's budget request for FY2005.