An Interrupted Time Series Analysis of the Dissemination of a Sickle Cell Vaso-Occlusive Episode Treatment Algorithm and a Case Management Referral Form for Individuals With Sickle Cell Disease in the Emergency Department.

BACKGROUND: Sickle cell disease is associated with frequent vaso-occlusive episode and emergency department visits. Our group developed (1) a vaso-occlusive episode treatment algorithm based on the National Heart, Lung, and Blood Institute recommendations, and (2) a case management referral form to identify social behavioral health needs of patients with sickle cell disease in the emergency department. The aims of this project were to (1) disseminate the vaso-occlusive episode algorithm and case management referral form, and (2) to evaluate the individual provider-reported awareness, use, and preferred method of access to each tool among emergency department providers in North Carolina. METHODS: An interrupted time series analysis was used to study the impact that an educational effort had on the awareness of a sickle cell vaso-occlusive episode treatment algorithm and a case management referral form. A targeted list was developed to identify the providers working in emergency departments with the largest number of sickle cell disease patient visits. In-service education was provided to targeted emergency departments in North Carolina over a period of 3 years. The vaso-occlusive episode algorithm and case management referral form were put up on the websites of professional organizations. Surveys were provided to emergency department providers at these targeted emergency departments with a baseline and post dissemination at 20 and 32 months for assessing the provider awareness, use, and preferred method of access of the tools. Additional feedback could be given by the participants through telephone interviews. Descriptive statistics were obtained, and content analysis was performed on interviews. RESULTS: We received survey responses for the following periods: baseline (T1, n = 190), post dissemination at 20 (T2, n = 142), and 32 months (T3, n = 93). Awareness of the tools was between 42% (baseline) and 41% (post dissemination at T2, T3) for the vaso-occlusive episodes algorithm and 25% (baseline) and 29% (post dissemination at T2, T3) for the case management referral form. However, use of these tools was found to be low as only 19% of the emergency department providers used the vaso-occlusive episode algorithm at T1 and 13% T2, while 5% of emergency department providers used the case management referral form at T1 and 6% at T2. With further education about the tools, an increase in the usage of the tools was observed T3, which was 29% for the vaso-occlusive episodes algorithm and 9% for the case management referral form. Lack of incorporation of the decision support tools into emergency department processes was observed to be an overarching barrier to dissemination of the tools identified in interviews (n = 8). CONCLUSIONS: This study can be used to inform future strategies on dissemination of evidence-based tools to emergency department providers.

Multicenter Validation of an Emergency Department-Based Screening Tool to Identify Elder Abuse.

STUDY OBJECTIVE: Emergency department (ED) visits provide an important opportunity for elder abuse identification. Our objective was to assess the accuracy of the ED Senior Abuse Identification (ED Senior AID) tool for the identification of elder abuse. METHODS: We conducted a study of the ED Senior AID tool in 3 US EDs. Participants were English-speaking patients 65 years old and older who provided consent or for whom a legally authorized representative provided consent. Research nurses administered the screening tool, which includes a brief mental status assessment, questions about elder abuse, and a physical examination for patients who lack the ability to report abuse or for whom the presence or absence of abuse was uncertain. The reference standard was based on the majority opinion of a longitudinal, expert, all data (LEAD) panel following review and discussion of medical records, clinical social worker notes, and a structured social and behavioral evaluation. For the reference standard, LEAD panel members were blinded to the results of the screening tool. RESULTS: Of 916 enrolled patients, 33 (3.6%) screened positive for elder abuse. The LEAD panel reviewed 125 cases: all 33 with positive screen results and a 10% random sample of negative screen results. Of these, the panel identified 17 cases as positive for elder abuse, including 16 of the 33 cases that screened positive. The ED Senior AID tool had a sensitivity of 94.1% (95% confidence interval [CI] 71.3% to 99.9%) and specificity of 84.3% (95% CI 76.0% to 90.6%). CONCLUSION: This multicenter study found the ED Senior AID tool to have a high sensitivity and specificity as a screening tool for elder abuse, albeit with wide CIs.