Computer-generated recall letters for underimmunized children: how cost-effective?

OBJECTIVE

To evaluate the effectiveness and cost effectiveness of computer-generated recall letters to parents of children overdue for immunizations.

METHODS

This randomized controlled trial included children of two facilities in a regional health maintenance organization. Parents of 20-month-olds who had not yet received a measles-mumps-rubella (MMR) immunization were identified via a computerized immunization tracking system. One half were mailed personalized letters that included the recommended immunization schedule and a request to call for an appointment; the other half served as a control group. Receipt of the MMR between 20 and 24 months of age was evaluated with the computerized tracking system. A telephone survey was conducted with parents whose children had not received the MMR by 24 months. Decision analysis was used to project the theoretical outcomes and costs of a recall letter policy for other populations.

RESULTS

Among 20-month-old children 10% had not received the MMR; 289 families were included in the analysis. Of families who were mailed letters, 54% (82 of 153) received the MMR by 24 months of age, compared with 35% (47 of 136) of those in the control group (P = 0.001). The telephone survey was completed with 110 parents of children who still did not appear on the health plan computer as having received the MMR by 24 months. Fifteen percent said the child had received an immunization at an outside provider, and of the rest 62% said they had not been aware that an immunization was due. In the cost effectiveness analysis it was projected that recall letters would increase the immunization rate for the regional population of approximately 30000 children from 86% to 90% at a total cost of $5031 annually. The cost per additional child appropriately immunized was $4.04. In sensitivity analyses this cost effectiveness ratio varied depending on the baseline population coverage rate as well as the estimated effectiveness of recall letters.

CONCLUSIONS

Computer-generated letters to recall children overdue for immunizations resulted in a higher proportion of children appropriately immunized in this setting. However, the strategy was not as cost-effective as intuition might suggest. Further studies in health maintenance organization (HMO) settings should compare the cost effectiveness of letters with other low cost strategies including automated telephone reminders.