A strategy to improve the utilization of pneumococcal vaccine

To evaluate the effectiveness of a computerized reminder for pneumococcal vaccination at hospital discharge and to determine patient and physician characteristics associated with increased use of the vaccine.

Pre- and postintervention study.

All medical services in a university teaching hospital.

All patients with at least one indication for pneumococcal vaccination discharged from the hospital during one of two three-month time periods; resident and faculty physicians caring for the same patients.

Incorporation of a predischarge reminder for pneumococcal vaccination in the hospital information system.

Of 539 eligible patients discharged during the three months after the intervention, 244 (45%) received the vaccine compared with 16 of 474 (3.4%) before the intervention (p less than 0.0001). Following the intervention, patients discharged with a diagnosis of alcoholism were more likely to receive the vaccine than were those without that diagnosis (58.1% vs. 42.7%, p less than 0.05), while patients with a diagnosis of cancer were less likely to get the vaccine (42 of 130, or 32.3%) than were those without cancer (202 of 409, 49.3%) (p less than 0.01). Patients whose attending physicians specialized in hematology-oncology or cardiology were also less likely to receive the vaccine than were all other patients. With the intervention in place, physicians were more likely to vaccinate patients with more than one indication for pneumococcal vaccine.

1) A predischarge reminder is an inexpensive, effective method to improve physicians' utilization of pneumococcal vaccine in high-risk patients; 2) additional improvements in pneumococcal vaccine utilization will require selective components directed toward specific diagnoses or attending physician subspecialities.