There is little understanding of the consistency and quality of treatment given to inflammatory bowel disease (IBD) patients in the community setting. Practice guidelines have been published, but it is unknown whether they are appropriate for and adhered to in community practice, or whether there are barriers to their implementation. We hypothesize relatively high variation in drug therapy for maintenance of remission in Crohn's disease (CD) and ulcerative colitis (DC). If variation is high for reasons that cannot bejustified, then there maybe opportunities to improve care by changing the forces that drive treatment decision-making. The study will describe the natural history of CD and DC, and whether care is consistent with practice guidelines. It will assess predictors of therapies, visits, and procedures, including predictors at the level of the patient, the disease, the provider, and the clinic. If possible, we will also link predictors and therapies to flare. Patient and disease characteristics will include: (a) the type and severity of IBD, (b) age and demographics, and (c) co-morbidity. Provider characteristics will include (a) specialty, (b) gender and years of experience, (c) IBD patient volume, and (d) actual average use of the procedures and therapies across their IBDpatient panel. Therapies will include: (a) 5ASA medications for maintenance of remission in UC, (b) 6MP/azathioprine for maintenance of remission in patients in CD and UC, (c) steroids, both short- and long-term, (d) infliximab, and (e) patient adherence to medications. Outcomes will include: (a) utilization of visits, procedures, and drugs, and (b) disease flare. In addition, we will qualitatively assess organizational factors, operating at the level of the clinic, that affect therapy decision-making, including staffing, information, guidelines, peer review, and specialist expertise, among others. The study will be conducted at Kaiser Permanente, Northern California. It will include more than 7,000 IBD patients who were enrolled for at least one year as of January 1, 2001 as well 75 gastroenterologists and primary care providers working at 20 clinics across Northern California. Patients will be observedfrom 1995 through 2006, with predictors being assessed during the first half of the 12-year observation period (1995- 2000), utilization during the second half (2001-2006), and drug therapy and disease activity throughout the entire 12-year period. Provider information will be obtainedfrom the computerized provider file with IBD patient volume and actual averagedrug use in their IBD patient panel determined from utilization data. All analyses will be conducted separatelyfor CD and UC. Analytical models that take into account the non- independence of patients within providers and clinics will be used to simultaneously evaluate all key variables.
Showing the most recent 10 out of 11 publications
