Influence of follow-up methodology and completeness on apparent clinical outcome of fundoplication

Abstract

Background: Outcomes after surgery can be determined by face-to-face interview, interviewing patients by telephone, or by mailing out questionnaires. For reasons of convenience, many clinical research studies use mail survey methods. Mail-out responses, however, are rarely complete, and patients returning mail-out questionnaires might be different from patients who do not return questionnaires.

Methods: We identified a group of 75 patients who were all sent a mail-out questionnaire 5 or more years after a laparoscopic fundoplication. Some of these patients returned questionnaires and some did not. All patients were subsequently interviewed by telephone. The responses to the two different data collection methods were compared as well as the interview responses of patients who returned questionnaires with those of patients who did not to determine the influence of follow-up methodology on apparent clinical outcome.

Results: Of the 75 patients interviewed, 49 patients (65%) had previously returned a mail-out questionnaire and 26 patients had not. The mean time difference between mail-out response and telephone interview was 1 month. When mail-out responses were compared with interview responses of the patients who had returned questionnaires, equivalent answers were obtained for all symptoms except for dysphagia, for which a higher incidence and greater scores were reported by patients in their mail-out questionnaire responses. When mail-out nonresponders were compared with responders, the nonresponders had a lower level of overall satisfaction with the outcome of surgery, a higher incidence of dysphagia, and greater levels of heartburn.

Conclusions: The measurement of surgical outcomes using a mail-out questionnaire may not be an accurate method of assessing patient symptoms and satisfaction after surgery, particularly if follow-up is incomplete. In this study, patients not responding to mail-out questionnaires were more likely to have adverse symptoms and a lower level of satisfaction than patients returning questionnaires. This dissatisfaction could make patients less likely to return mail-out questionnaires, and for this reason studies with incomplete follow-up that rely on mail-out questionnaire responses may report erroneously higher surgical success rates.