Variations in Provider Responses to Automated Decision Support and Impact on Missed Opportunities for Vaccine Adolescent Administration

Vaccine coverage rates are far lower for adolescents than they are for younger children. Providers are more likely to deliver recommended vaccines to adolescents at standard preventive visits and less likely to vaccinate older adolescents and problem-focused visits. An automated clinical decision support tool provides actionable and up-to-date information at the point of care to reduce missed opportunities for vaccination of adolescents regardless of when they engage in care. The purpose of this study is to determine the degree to which vaccine delivery following a vaccine computerized clinical decision support system (CCDSS) varied by patient characteristics (age, gender, and number of vaccines due) and visit characteristics (preventive vs. problem focused). We implemented a CCDSS in a single urban teaching clinic where residents, adolescent medicine fellows, and nurse practitioners provide pediatric primary care, adolescent medicine and HIV specialty care. This tool provided an alert to providers, using the clinic's electronic medical platform, when adolescents ages 11-21 were due for any recommended vaccine, based on standard Center for Disease Control & Prevention schedules. These alerts were triggered for all types of visits, including acute visits. We examined factors associated with vaccination including age (using following age categories: 11-12, 13-17, and =18); gender; type and number of vaccines due; and visit type using chi-squared and multiple logistic regression analyses. In a 15-month period, 5943 alerts for vaccines due were sent during 3,672 visits among 1,936 unique patients. The number of alerts per visit ranged from 1-5 vaccines (Mean 1.6, SD 0.89). Alerts were most common for HPV (47.7%) and MCV (27.2%). Hepatitis B (2.2%), IPV (2.0%) and MMR (1.7%) were less common. Overall, 45.8% of reminders resulted in vaccine administration at that visit. Response rates were highest for younger adolescents (55.3% for ages 11-12, compared to 47.1% for ages 13-17 and 34.1% for those =18, p=<0.001), and higher for males than females (51.4% vs. 40.5%, p=<0.001). Response rates were highest when patients were due for 3 vaccines (57.6%) and lowest when patients were due for only 1 vaccine (42.5%) (p < 0.001). In the final adjusted model, males had a 46% greater odds to receive vaccines in response to alerts than females (OR 1.46, 95% CI 1.28-1.67), while older adolescents were less likely to receive vaccinations as compared to 11-12 year olds (13-17yo: OR 0.79 (95% CI 0.67-0.94) & =18yo: OR 0.46 (95% CI 0.37-0.55). Patients requiring 2 or 3 vaccinations were more likely to receive them than those requiring only 1 (OR 1.40, (95% CI 1.17-1.67) and OR 1.58 (95% CI 1.29-1.92) respectively). Provider response to CCDSS did not differ by visit type in final model. The use of an electronic alert system for adolescents has potential for improving vaccine administration by promoting vaccination during early adolescence and targeting certain groups that would otherwise be missed by vaccines targeting only females. Additional efforts may need to be used to promote vaccination in older adolescents.