The finding: The hospitalization rate of all behavioral health-related ED visits across 16 hospitals was found to be 15%. Together, enabling factors (health insurance status, ED LOS, community-based behavioral service availability, and mode of arrival to the ED) accounted for more of the variability in hospitalization than predisposing (demographic) or need-based (behavioral health diagnosis, alcohol and substance use disorders) factors. Older age, increased ED LOS, arrival by ambulance, co-occuring suicidal ideation and behavior, and diagnoses of schizophrenia or other psychotic disorder, affective disorder or personality disorder all increased the likelihood of hospitalization following an ED visit. Factors that decreased the odds of hospitalization included African American and Hispanic ethnicity, uninsured status, availability of community-based behavioral health services within 5 miles of the ED, and the diagnosis of an anxiety disorder.

Strength and weaknesses: This large multi-site retrospective cohort study adapted the existing Anderson Healthcare Utilization Model to examine predictors of inpatient psychiatric hospitalization among a population presenting for behavioral health visits to a network of EDs within a county in Texas. This is a commonly used framework for systematically categorizing an array of health indicators but has its limitations, such as that it cannot determine causation or detect potential interactions between factors. The study identified and expanded upon a set of factors from the literature, placed them in a framework to predict the likelihood of hospitalization, and hypothesized how these findings could be used to improve the delivery of behavioral health care.

Relevance: With the rise of behavioral health-related ED visits, and the ED being the predominant point of entry for psychiatric hospitalizations, further understanding the factors associated with inpatient admissions can lead to strategies for improving care for these patients as well as resource utilization for emergency psychiatric services.

The finding: Pain intensity scores were significantly reduced when the pregabalin–duloxetine combination approach was used when compared to pregabalin alone or placebo. The combination approach had a favorable outcome compared to duloxetine monotherapy, but failed to demonstrate statistical significance. The study also validated duloxetine monotherapy as a superior treatment to pregabalin monotherapy when compared directly.

Strength and weaknesses: This RCT was well designed with multiple standardized outcome measures and a rigorous crossover design, allowing comparisons of each treatment approach with each individual subject. The main limitation of a crossover trial is the potential carryover effects of between serial treatment round, which was minimized by the extended taper and a drug-free day.

Relevance: Combination therapy for fibromyalgia is common. This is the first RCT comparing a combination of an antidepressant and anticonvulant with monotherapy of each for the treatment of fibromyalgia.