A Cost Analysis of Colonoscopy using Microcosting and Time-and-motion Techniques

Abstract

Background

The cost of an individual colonoscopy is an important determinant of the overall cost and cost-effectiveness of colorectal cancer screening. Published cost estimates vary widely and typically report institutional costs derived from gross-costing methods.

Objective

Perform a cost analysis of colonoscopy using micro-costing and time-and-motion techniques to determine the total societal cost of colonoscopy, which includes direct health care costs as well as direct non-health care costs and costs related to patients’ time. The design is prospective cohort. The participants were 276 contacted, eligible patients who underwent colonoscopy between July 2001 and June 2002, at either a Veterans’ Affairs Medical Center or a University Hospital in the Southeastern United States.

Major results

The median direct health care cost for colonoscopy was $379 (25%, 75%; $343, $433). The median direct non-health care and patient time costs were $226 (25%, 75%; $187, $323) and $274 (25%, 75%; $186, $368), respectively. The median total societal cost of colonoscopy was $923 (25%, 75%; $805, $1047). The median direct health care, direct non-health care, patient time costs, and total costs at the VA were $391, $288, $274, and $958, respectively; analogous costs at the University Hospital were $376, $189, $368, and $905, respectively.

Conclusion

Microcosting techniques and time-and-motion studies can produce accurate, detailed cost estimates for complex medical interventions. Cost estimates that inform health policy decisions or cost-effectiveness analyses should use total costs from the societal perspective. Societal cost estimates, which include patient and caregiver time costs, may affect colonoscopy screening rates.

Appendix

Additional details of Microcosting and Time-and-Motion Techniques

Resource Identification

Resources consumed during each transaction included all personnel, equipment, and medications associated with performing colonoscopy as well as resources administered outside the endoscopy unit such as overhead (e.g. utilities), building costs, and administrative costs. Resources consumed during particular transactions were measured through real-time observation. The cost of complications associated with colonoscopy was not addressed in this study.

Direct Observation of Resource Utilization

The same nurse physically accompanied every participant during his or her colonoscopy from the patient’s leaving the waiting room through his or her postprocedure recovery. Using a watch, this nurse recorded the time that all health care personnel arrived at and left a patient’s bedside on a standardized data collection tool that allowed calculation of the total number of minutes that each type of health care personnel (e.g. nurse, resident) spent providing care to each patient. The nurse also directly observed and recorded the medication and equipment used for each procedure.

Using a watch, one author (SGH) personally observed a convenience sample of at least 20 instances of each coordinating transaction at both hospitals and recorded the start and stop times for each coordinating transaction on a standardized data collection instrument that allowed calculation of the mean duration of each coordinating transaction to the nearest minute. For transactions performed by several different types of health care personnel (e.g. nurses and GI technicians), separate sets of observations were made for each personnel type. The mean time required to verify insurance approval at the University Hospital was derived from both observation and interviews with insurance coordinators, who obtained approvals for multiple patients’ procedures in batches rather than processing each patient’s approval separately.

Monetary Valuation of Resources

To estimate operating utilization, we divided the estimated number of hours spent performing procedures during FY 2002 by the total number of room-hours (number of rooms per procedure unit multiplied by the total hours of operation during FY 2002) at each hospital during FY 2002. One author (SGH) observed a convenience sample of at least 20 of each type of noncolonoscopy endoscopic procedure performed at each hospital to determine mean procedure length. These values and the mean duration for colonoscopy at each site were used to calculate the total number of hours spent performing procedures in FY 2002. The capacity utilization was calculated to be 59% for the University Hospital and 48% for the VA Medical Center. After adjusting the per-minute wages for administrative overhead and capacity utilization (see Table 1 notes), labor costs were calculated by multiplying the employee time for each procedure by the appropriate per-minute rate. For transactions that were conducted by more than one type of health care worker, the per-minute compensation rate used to determine cost was a weighted average based on the fraction of total observed transactions completed by each type of health care worker.