RESUMO
A patient-education program model with 18 criteria was refined into a survey instrument and sent to 217 Michigan hospitals. Of these, 181 hospitals and one free-standing clinic responded, with 47 indicating they had no applicable programs and 135 providing data on 281 existing patient-education programs. A mean (+/- SD) of 12.1 +/- 3.9 criteria were met. The range included two programs that met only one criterion and 15 that met all 18. After it was established that the criteria operated independently of each other, six factors were analyzed to observe their impact on the number of criteria met by the 281 programs. Analysis of variation (ANOVA) showed that presence and type of program coordinator accounted for 17% of explained variance, subject for 8%, hospital size for 4.4%, existence of a defined budget for 4%, and geographic location for 3%. Because of unequal numbers, the full-time/part-time coordinator factor was not included in the ANOVA. More than one third of the variance affecting numbers of criteria met by patient-education programs was explained by these factors.