Faculty Bibliography

The health of people ever incarcerated in New York City (NYC) jails during 2001 to 2005 was characterized by matching jail, shelter, mortality, sexually transmitted infection, HIV, and tuberculosis (TB) data from the NYC Departments of Health and Mental Hygiene, Correction, and Homeless Services. Compared with nonincarcerated people and those living in the lowest income NYC neighborhoods, those ever incarcerated had higher HIV prevalence and HIV case rates. Ever-incarcerated females also had higher rates of gonorrhea and syphilis than nonincarcerated females. Ever-incarcerated people who used the single adult homeless shelter system had higher HIV, gonorrhea, and TB case rates and all-cause mortality rates than ever-incarcerated people without shelter use, when adjusting for other variables. People ever incarcerated in NYC jails are at risk for conditions of public health importance. Sex-specific jail- and community-based interventions are needed.

We review recent findings related to the neurobiology of infant attachment, emphasizing the role of parenting quality in attachment formation and emotional development. Current findings suggest that the development of brain structures important for emotional expression and regulation (amygdala, prefrontal cortex, hippocampus) is deeply associated with the quality of care received in infancy, with sensitive caregiving providing regulation vital for programming these structures, ultimately shaping the development of emotion into adulthood. Evidence indicates that without sensitive caregiving, infants fail to develop mechanisms needed for later-life emotion and emotion regulation. Research suggests that a sensitive period exists in early life for parental shaping of emotional development, although further cross-species research is needed to discern its age limits, and thus inform interventions.

There is now an unprecedented opportunity to improve the care of the over 5 million people who are living with Alzheimer's disease and related dementias and many more with cognitive impairment due to brain injury, systemic diseases, and other causes. The introduction of a new Medicare care planning benefit-long sought openly by advocacy organizations and clinicians and badly needed by patients and families-could greatly improve health care quality, but only if widely and fully implemented. We describe the components of this new benefit and its promise of better clinical care, as well as its potential to create a new platform for clinical and health outcomes research. We highlight external factors-and some that are internal to the benefit structure itself-that challenge the full realization of its value, and we call for broad public and professional engagement to ensure that it will not fail.

Study Objectives: To help address the growing opioid overdose epidemic and help teach a core toxicological emergency, the authors taught the use of naloxone as an antidote to an opioid overdose, for all first-year medical students as a part of first responder training using cardiopulmonary resuscitation, as an online and in-person training over three years. Previously we demonstrated that in-person opioid overdose prevention training as an adjunct to BLST improves knowledge and preparedness. To compare the educational outcomes; knowledge, preparedness, and attitudes, for online vs in-person opioid overdose prevention training. Methods: Opioid overdose prevention trainings were conducted in person in 2014 and 2015, and online in 2016. First year students completed pre-and post-training surveys covering three measures: knowledge (11-point scale), attitudes (66-point scale) towards patients with opioid use disorders, and self-reported preparedness (60-point scale) to respond to an opioid overdose. Online and in-person scores across all three measures were compared using analysis of covariance (ANCOVA) methods across two years of trainings. Results: After controlling for pre-test scores, there were very small and not meaningful differences in attitude and knowledge scores between in-person training and online training. The estimated difference for knowledge was-0.06 (95% CI-0.48-0.35) and for attitudes was 0.64 (95% CI-0.22-1.50). The average scores related to preparedness were higher for the students who took the course online, estimated at 2.10 points (95% CI 0.97-3.22). Feedback was generally positive, with 96% of the in-person group saying future classes should receive the training and 95% of the online group saying all medical schools should provide the training. Conclusions: Online training has become a more common method of medical education due to its many advantages including standardization, scalability and flexibility to accommodate asynchronous learning. However, few studies have performed analyses of online training vs in-person training for relative effectiveness. The authors have demonstrated that for training medical students to administer naloxone as an antidote to an opioid overdose, online training is comparable to in-person training. These results support the use of online training for adding training on administering naloxone

OBJECTIVE: Injury related morbidity and mortality is an important emergency medicine and public health challenge in the United States (US). Here we describe the epidemiology of traumatic injury presenting to US emergency departments, define changes in types and causes of injury among the elderly and the young, characterize the role of trauma centers and teaching hospitals in providing emergency trauma care, and estimate the overall economic burden of treating such injuries. METHODS: We conducted a secondary retrospective, repeated cross-sectional study of the Nationwide Emergency Department Data Sample (NEDS), the largest all-payer emergency department survey database in the US. Main outcomes and measures were survey-adjusted counts, proportions, means, and rates with associated standard errors, and 95% confidence intervals. We plotted annual age-stratified emergency department discharge rates for traumatic injury and present tables of proportions of common injuries and external causes. We modeled the association of Level 1 or 2 trauma center care with injury fatality using a multi-variable survey-adjusted logistic regression analysis that controlled for age, gender, injury severity, comorbid diagnoses, and teaching hospital status. RESULTS: There were 181,194,431 (standard error, se = 4234) traumatic injury discharges from US emergency departments between 2006 and 2012. There was an average year-to-year decrease of 143 (95% CI -184.3, -68.5) visits per 100,000 US population during the study period. The all-age, all-cause case-fatality rate for traumatic injuries across US emergency departments during the study period was 0.17% (se = 0.001). The case-fatality rate for the most severely injured averaged 4.8% (se = 0.001), and severely injured patients were nearly four times as likely to be seen in Level 1 or 2 trauma centers (relative risk = 3.9 (95% CI 3.7, 4.1)). The unadjusted risk ratio, based on group counts, for the association of Level 1 or 2 trauma centers with mortality was RR = 4.9 (95% CI 4.5, 5.3), however, after accounting for gender, age, injury severity and comorbidities, Level 1 or 2 trauma centers were not associated with an increased risk of fatality (odds ratio = 0.96 (0.79, 1.18)). There were notable changes at the extremes of age in types and causes of emergency department discharges for traumatic injury between 2009 and 2012. Age-stratified rates of diagnoses of traumatic brain injury increased 29.5% (se = 2.6) for adults older than 85, and increased 44.9% (se = 1.3) for children younger than 18. Firearm related injuries increased 31.7% (se = 0.2) in children five years and younger. The total inflation-adjusted cost of emergency department injury care in the US between 2006 and 2012 was $99.75 billion (se = 0.03). CONCLUSIONS: Emergency departments are a sensitive barometer of the continuing impact of traumatic injury as an important cause of morbidity and mortality in the US. Level 1 or 2 trauma centers remain a bulwark against the tide of severe trauma in the US. But, the types and causes of traumatic injury in the US are changing in consequential ways, particularly at the extremes of age, with traumatic brain injuries and firearm-related trauma presenting increased challenges

BACKGROUND: Data on both known and unknown drug use in the electronic dance music (EDM) scene is important to inform prevention and harm reduction. While surveys are the most common method of querying drug use, additional biological data can help validate use and detect unknown/unintentional use of drugs such as new psychoactive substances (NPS). We sought to determine the extent of both known and unknown use of various substances in this high-risk scene. METHODS: We hair-tested 90 self-reported past-year ecstasy/MDMA/Molly users attending EDM parties in New York City during the summer of 2016 using UHPLC-MS/MS. Results were compared to self-reported past-year use. RESULTS: Three quarters (74.4%) tested positive for MDMA, a third (33.3%) tested positive for an NPS, and 27.8% tested positive specifically for one or more synthetic cathinones (e.g., butylone, ethylone, pentylone, methylone, alpha-PVP). Half (51.1%) of participants tested positive for a drug not self-reported, with most testing positive for synthetic cathinones (72.0%), methamphetamine (69.0%), other NPS stimulants (e.g., 4-FA, 5/6-APB; 66.7%), or new dissociatives (e.g., methoxetamine, diphenidine; 60.0%). Attending parties every other week or more often, reporting higher-frequency ecstasy pill use, having tested one's ecstasy, and having found out one's ecstasy was adulterated, were risk factors for testing positive for synthetic cathinones and NPS in general. CONCLUSION: Hair testing appears to be a valuable addition to drug epidemiology studies. Many EDM party attendees-even those who test their ecstasy-are unknowingly using NPS and/or other drugs. Prevention information and harm reduction may help reduce unknown/unintentional use.

Microarray studies generate a large number of p-values from many gene expression comparisons. The estimate of the proportion of the p-values sampled from the null hypothesis draws broad interest. The two-component mixture model is often used to estimate this proportion. If the data are generated under the null hypothesis, the p-values follow the uniform distribution. What is the distribution of p-values when data are sampled from the alternative hypothesis? The distribution is derived for the chi-squared test. Then this distribution is used to estimate the proportion of p-values sampled from the null hypothesis in a parametric framework. Simulation studies are conducted to evaluate its performance in comparison with five recent methods. Even in scenarios with clusters of correlated p-values and a multicomponent mixture or a continuous mixture in the alternative, the new method performs robustly. The methods are demonstrated through an analysis of a real microarray dataset.

Living in communities with persistent gun violence is associated with negative social, behavioral, and health outcomes, analogous to those of a natural disaster. Taking a disaster-preparedness approach may identify targets for community-based action to respond to on-going gun violence. We assessed the relevance of adapting a disaster-preparedness approach to gun violence and, specifically, the relationship between perceived collective efficacy, its subscales of social cohesion and informal social control, and exposure to gun violence. In 2014, we conducted a cross-sectional study using a community-based participatory research approach in two neighborhoods in New Haven, CT, with high violent crime rates. Participants were ≥18 years of age and English speaking. We measured exposure to gun violence by adapting the Project on Human Development in Chicago Neighborhoods Exposure to Violence Scale. We examined the association between perceived collective efficacy, measured by the Sampson Collective Efficacy Scale, and exposure to gun violence using multivariate modeling. We obtained 153 surveys (51% response rate, 14% refusal rate, and 35% non-response rate). Ninety-five percent reported hearing gunfire, 58% had friend or family member killed by gun violence, and 33% were physically present during a shooting. In the fully adjusted model, one standard deviation higher perceived collective efficacy was associated with lower reported exposure to gun violence (β = -0.91, p < 0.001). We demonstrated that it is possible to activate community members and local officials to engage in gun violence research. A novel, community-based approach adapted from disaster-preparedness literature may be an effective framework for mitigating exposure to gun violence in communities with persistent gun violence.

Hospital patients with serious mental illness (SMI) have high rates of smoking. There are few post-discharge treatment models available for this population and limited research on their treatment uptake following discharge. This study is a secondary analysis of an RCT that compared multi-session intensive telephone counseling versus referral to state quitline counseling at two safety net hospitals in New York City. For this analysis, we selected all trial participants with a history of schizophrenia, schizoaffective disorder or bipolar disorder (N = 384) and used multivariable logistic regression to compare groups on self-reported 30-day abstinence at 6 months and to identify patient factors associated with use of tobacco treatment. Analyses found no significant group differences in abstinence 6 months (28% quitline vs. 29% intervention, p > 0.05), use of cessation medications (42% quitline vs. 47% intervention, p > 0.05) or receipt of at least one counseling call (47% quitline vs. 42% intervention, p > 0.05). Patients with hazardous drinking (p = 0.04) or perceived good health (p = 0.03) were less likely to use cessation medications. Homeless patients were less likely to use counseling (p = 0.02). Most patients did not use cessation treatment after discharge, and the intensive intervention did not improve abstinence rates over quitline referral. Interventions are needed to improve use of cessation treatment and long-term abstinence in patients with SMI.