Acute Trauma Factor Associations With Suicidality Across the First 5 Years After Traumatic Brain Injury.

OBJECTIVE: To determine whether severity of head and extracranial injuries (ECI) is associated with suicidal ideation (SI) or suicide attempt (SA) after traumatic brain injury (TBI). DESIGN: Factors associated with SI and SA were assessed in this inception cohort study using data collected 1, 2, and 5 years post-TBI from the National Trauma Data Bank and Traumatic Brain Injury Model Systems (TBIMS) databases. SETTING: Level I trauma centers, inpatient rehabilitation centers, and the community. PARTICIPANTS: Participants with TBI from 15 TBIMS Centers with linked National Trauma Data Bank trauma data (N=3575). INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: SI was measured via the Patient Health Questionnaire 9 (question 9). SA in the last year was assessed via interview. ECI was measured by the Injury Severity Scale (nonhead) and categorized as none, mild, moderate, or severe. RESULTS: There were 293 (8.2%) participants who had SI without SA and 109 (3.0%) who had SA at least once in the first 5 years postinjury. Random effects logit modeling showed a higher likelihood of SI when ECI was severe (odds ratio=2.73; 95% confidence interval, 1.55-4.82; P=.001). Drug use at time of injury was also associated with SI (odds ratio=1.69; 95% confidence interval, 1.11-2.86; P=.015). Severity of ECI was not associated with SA. CONCLUSIONS: Severe ECI carried a nearly 3-fold increase in the odds of SI after TBI, but it was not related to SA. Head injury severity and less severe ECI were not associated with SI or SA. These findings warrant additional work to identify factors associated with severe ECI that make individuals more susceptible to SI after TBI.

Differential adoption of pharmacotherapy recommendations for type 2 diabetes by generalists and specialists.

Newer, multimedication (novel) regimens provide better glycemic control for many type 2 diabetics when sulfonylurea monotherapy (traditional) becomes ineffective. Because better glycemic control is associated with decreased likelihood of complications and lower utilization and cost of care, the authors examined change in prescribing patterns for veterans with type 2 diabetes between FY 97 and 99. They classified medication regimens as traditional and novel based on the combination of diabetes medications patients received at the end of each year. Multivariate logistic regression analyses controlling for disease severity indicated that patients were more likely to receive novel regimens over time, but those seen only in primary care were less likely to receive novel regimens than those previously seen by a specialist. Geographic differences and differences in how recommendations were implemented by generalists and specialists suggest that diffusion of innovations theory may help explain variations in practice and guide interventions designed to translate research into practice.