Low endoscopy completion rates are a major problem in the VA, causing delay or failure to receive essential care, increased clinic wait times, lost capacity, increased costs, thus limiting endoscopic screening for colorectal cancer. This study tests whether an Interactive Voice Response (IVR) messaging system is equally effective in promoting the completion of flexible sigmoidoscopy and colonoscopy as usual clinical care practices that include phone calls from nurses to patients prior to preparation and procedures. Previous studies have examined the role of scheduling facilitation or patient adherence on endoscopy completion and the use of IVR technology to enhance patient adherence in other medical contexts. This is the first study, however, to evaluate the use of IVR for endoscopy completion and the first to compare it to the effectiveness of phone calls from nurses prior to an endoscopy appointment.

Background:

Low endoscopy completion rates are a major problem nationwide and in the VA. For clinics, delays or failure to complete exams can cause clinic inefficiencies, such as increased wait times for needed procedures, lost capacity, and increased costs. For patients, delays reduce the chance for recommended timely screening consistent with practice guidelines and for diagnostic tests, can cause significant anxiety, delayed treatment and possibly poorer prognosis. This study tests whether an Interactive Voice Response (IVR) messaging system is equally effective in motivating patients to complete a flexible sigmoidoscopy or colonoscopy as usual clinical care practices, which include reminder phone calls from clinic nurses. This is the first study to evaluate the use of IVR for endoscopy completion and the first to compare it to the effectiveness of phone calls from nurses prior to an endoscopy appointment.

Objectives:

The primary set of objectives was to test whether IVR messaging was equivalent to clinic usual care (UC) practices in motivating patients to attend a scheduled flexible sigmoidoscopy or colonoscopy appointment and to adequately prepare for the exam. Secondary objectives included comparing patient satisfaction with UC and IVR phone calls and assessing if IVR or UC was more effective for sub-groups that may have more difficulty with preparation, including those with poor physical and mental functioning, health literacy, social support and trust in physicians and those with spinal cord injury, paraplegia, PTSD, or with little intention to be tested for colorectal cancer in the future.

Methods:

This was a stratified 3-arm randomized controlled trial among patients with upcoming flexible sigmoidoscopy or colonoscopy appointments. All patients who had a colonoscopy or flexible sigmoidoscopy appointment scheduled from August 20, 2007 through October 31, 2008 were assessed for inclusion in this study. Patients were not considered eligible if, based on a medical record review prior to randomization, they had unreliable means of receiving the intervention or the intervention would have provided inappropriate or inaccurate information. The three study arms included: 1) UC (nurse phone call 7 days prior to the procedure); 2) IVR7 (call from IVR system 7 days prior to procedure); and, 3) IVR3 (call from IVR system 3 days prior to procedure). One week after the initial appointment self-administered surveys were sent to all participants to assess satisfaction with reminder/motivation calls. Appointment and gastrointestinal (GI) procedure data were extracted from medical record files to assess study outcomes. The principal outcome measures were (1) attendance at the scheduled endoscopy appointment; (2) adequate preparation for the exam; (3) patient satisfaction with reminder/motivation telephone calls.

Status:

Complete

Inclusion Criteria:

All patients with either flexible sigmoidoscopy or colonoscopy appointments scheduled greater than 7 days before their appointment in the GI endoscopy clinic from August 20, 2007 through October 31, 2008.

Exclusion Criteria:

Patients were not considered eligible for inclusion in the study if, based on a medical record review prior to randomization, they had unreliable means of receiving the intervention or the intervention would have provided inappropriate or inaccurate information. These patients included those who lived in a nursing or group home or homeless shelter; had no listed telephone number; scheduled the appointment less than 8 days in advance; or, had Type 1 diabetes, dementia or Alzheimer's or multiple GI procedures on the same day (such as those with both upper and lower GI procedures). All patients excluded from the study were assigned to usual care.