Measuring the Use of Health Services by Household Interviews: A Comparison of Procedures Used in Three Child Health Surveys

Publisher Website
Google Scholar
Abstract
Four different survey methods for estimating the volume of physician visits in a child population were employed in three different surveys: the standard 2-week-frequency recall question, a supplementary 12-month-frequency recall question, a 4-week health calendar and a telephone follow-up interview. Comparison of the various estimates with each other and known figures about the actual supply of services permitted an assessment of their overall error. For selected cases a validity check against medical records and insurance payment files was possible. Though these checks rarely permitted study designs that would detect under-reporting and over-reporting equally well, or that would permit generalizations for all patients or providers, they gave insights in the amount and type of error involved. The health calendar provided the best utilization data, but at high costs. As the most efficient procedure, especially for small community surveys, we recommend a combined use of the standard 2-week question with a supplementary 12-month question. The telephone follow-up covering about 12 months had the highest error. A combination of household surveys, provider statistics and selected validity checks has in our view the greatest payoff for monitoring a health services market to aid regional planning.

This publication has 0 references indexed in Scilit: