[Patient satisfaction after inpatient rehabilitation: different results at different follow-up times?].

UNLABELLED: Different Results at Different Times of Questioning? In 1994 a quality assurance programme for the German medical rehabilitation system was introduced by the national association of pension funds. One of its five parts focussed on patients' satisfaction and health effects. Both domains were to be assessed by postal questionnaires. The first section of the instrument enquired about patients' perceptions and evaluations of both rehabilitation structures and processes. It is well known that satisfaction surveys tend to underestimate the true level of criticism due to social desirability bias. This is especially problematic in the context of quality assurance programmes where criticism can be seen as one driving force of improvement. The magnitude of the bias depends on many factors. In this study we tested the influence of different time intervals between discharge from rehabilitation clinic and first postal contact.

DESIGN: We conducted an experimentally varied satisfaction survey among four groups of rehabilitands. If necessary, the first postal questionnaire was followed by two reminders.

SETTING: One Bavarian pension fund for blue collar workers.

PATIENTS: 520 rehabilitands were recruited from various rehabilitation clinics specialized in muskuloskeletal disorders. About 70% suffered from nonspecific back pain. They were randomly allocated to one of the following time intervals between discharge and postal questionnaire: one, four, eight and twelve weeks. INSTRUMENTS AND TARGET VARIABLES: All patients received an identical questionnaire with 67 items and several subscales. We compared the extent of dissatisfaction between the four groups.

RESULTS: The response rate (after two reminders) showed a linear decrease from 97% (postal questionnaire after one week) to 83% (12 weeks). The extent of dissatisfaction did not change systematically though there was a (statistically nonsignificant) tendency towards higher levels of criticism at week twelve.

CONCLUSIONS: An experimental variation of time intervals between discharge from rehabilitation and first postal contact does not seem to influence the extent of reported dissatisfaction though it clearly influences the response rate: The later the sending out the lower the response rate. Taking into account additional administrative and organisational aspects of the programme we propose to send out satisfaction questionnaires between 8 and 10 weeks after rehabilitation.