Patients treated at the Segal Cancer Centre proceed through a standard route when they arrive for an appointment with their oncologist or for receipt of their treatment. In this study we mapped out that route, determined the length of time staff spend attending to the patient at each step and what that translates into with regards to personnel costs. We also assessed the cost of usage of the oncology clinic and treatment area which encompasses two floors of one pavilion in the hospital.
Cost analysis studies from different countries comparing XELOX and FOLFOX have routinely shown the former to be less expensive when taking into account the cost of purchasing and administering the drugs, the management of serious adverse events and other medical resource utilization costs [5–8]. In contrast, our study focused on the contribution of staff time and facility usage to the cost per patient visit and demonstrated that it was only approximately 12% lower with XELOX. Thus it appears that the self-administered oral medication, capecitabine, has no significant financial advantage with regards to staff time and facility usage because of the preparation and administration of drugs requiring intravenous access and infusion (oxaliplatin and bevacizumab).
The 2011 drug costs of FOLFOX/bevacizumab (modified FOLFOX6, bevacizumab, calcium and magnesium) and XELOX/bevacizumab (capecitabine, bevacizumab, oxaliplatin, calcium and magnesium) were $3,324/cycle given every two weeks and $5,603/cycle given every three weeks, respectively. The physician fee per follow-up visit, which usually occurs every second treatment, was $41.90. Taking into account the personnel and O&M costs determined here for each treatment ($136.06/cycle and $119.94/cycle respectively), in a 12-week time span the cost per patient of administering the treatment is $20,886 for FOLFOX/bevacizumab and $22,976 for XELOX/bevacizumab.
One striking outcome of this analysis is the demonstration that for the treatment of metastatic colorectal cancer, personnel and O&M costs are minimal, with drug costs contributing >95% of the total cost. This finding holds true across the entire calculated confidence intervals, which reflect the sensitivity of time results. Indeed, because all costs in the model were based on mean values of collected data on actual time rather than deterministic fixed parameters, the uncertainty in the model is captured through confidence intervals rather than sensitivity analyses. Given that the patient flow and work steps are similar across cancer types at the Segal Cancer Centre, this trend may well be the same for other cancers with some minor variations depending on the cancer type and the cost of treatments.
There were several limitations to this study. Indirect steps were determined separately from the direct steps and therefore were not linked to a particular patient visit. Furthermore, some indirect steps were performed in batches (e.g. blood samples sent to laboratory, preparation of calcium/magnesium) rather than individually for each patient visit. Although every effort was made to time all direct work steps of each patient visit, due to unforeseen circumstances, for some visits the timing could not be completed for particular steps, (e.g. registration, blood taking). However, given the consistency of the timings that had been obtained, the missed ones would likely not have significantly altered the outcome. Our time and motion protocol did not include assessing the time spent for unscheduled visits to the Centre for management of serious adverse events and symptom control issues. Moreover, during the treatment step the RA was only present for medication changeovers, which may have resulted in an underestimation of these work times. However, it is not likely that additional costs associated with these limitations would have significantly increased the impact of the personnel costs as compared with the cost of the drugs.
This time and motion study also provided a detailed description of the patient trajectory through our cancer centre, giving a glimpse of the complexity of care of these patients. The inefficiencies in patient flow, particularly specific points with exaggerated wait times, correspond directly with dissatisfaction consistently identified in two patient satisfaction surveys conducted by the Jewish General Hospital’s Quality Program. Furthermore, simple observation of the use of the common space (corridors, waiting areas) also aligns with the flow and waiting time data documented in this study. For example, the lengthy appointment booking time resulted in over-crowding of the common space and undue stress on patients as they waited to book their appointments. Process improvements have now been implemented to facilitate appointment booking and decrease the time of that work step. Given the relatively low proportion of personnel and facility usage cost in the total costs, improving efficiencies in the patient flow are not likely to result in major cost savings to the system. However, improving efficiency is of the utmost importance in order to be well-equipped to handle the increasing number of patients going through the cancer trajectory, and is a process that has been implemented at other cancer facilities [20–22].