Results of a randomized controlled trial analyzing telemedically supported case management in the first year after living donor kidney transplantation - a budget impact analysis from the healthcare perspective

A prospective, controlled, randomized and open project-study was realized as follows [7]: In September 2011, 50 patients who were scheduled for a living-donor kidney transplantation at the Transplantation Center Freiburg between October 2011 and March 2012 were randomly allocated to two groups receiving different aftercare programs during the first postoperative year.

The first group was offered standard aftercare in combination with telemedically supported case management. Daily, the patients completed a pre-defined medical questionnaire about their physical condition presented to them in their homes via an interactive web-based telemonitor. These data were monitored by medical staff with a special qualification for work with patients after kidney transplantation. If anomalous values occured, the medical staff contacted the patient by phone or video conference to discuss the following treatment process. The control group was offered standard aftercare without telemedically supported case management. Of the 50 randomized patients, two patients in each group dropped out, resulting in 46 available valid datasets.

At three points (3, 6 and 12 months after transplantation), utilization of health care resources is recorded for every patient. For outpatient care visits at general practitioners (N = 194) and or medical specialists (N = 1829), standardized unit costs from Bock et al [2014] [17] were applied, representing costs from a societal perspective. For scheduled (N = 39) and unscheduled (N = 68) episodes of hospitalization, actual amounts of reimbursement were collected from the respective hospitals. Ambulatory visits at the outpatient department were priced according to the respective reimbursement (€270 per visit). All prices reflect 2015 values. Overall, six cost figures (costs of general practitioner out-patient care visits, costs of medical specialist out-patient care visits, costs of scheduled in-hospital care, costs of unscheduled in-hospital care, costs of ambulatory visits at the outpatient department and total costs of care) were calculated for each observation period and patient.

Within the statutory health insurance in Germany, telemedically supported case management is not an element of standard care. A reimbursement by sickness funds is possible by using individual case requests, only. A hypothetical number of 20 patients receiving telemedically supported case management for a one year period is assumed to require an additional 50% nurse position (€28,500) with a suitably equipped work place (€500) and internal server provision (€1,024). In addition, patient-variable costs for a Touch-Screen-PC (€600) and software licenses (€300 per year) are claimed. To these fixed (€30,024 for 20 hypothetical patients) and variable (€900 per patient) costs of the one year telemedical support, an extra 25% for infrastructure expenses was added.

Skewed data is a major issue in statistical models of healthcare costs [18‐20]. Beside the fact that the six cost figures were positively skewed, most of them were equal to zero during a considerable number of observation periods because patients did not see a physician and/or were not hospitalized in the respective time period. In order to accommodate these characteristics of the data, a two-part model approach was chosen for the regression analyses [21‐24]. In two part models, a binary choice model is estimated for the probability of observing a zero versus positive outcome. Then, conditional on a positive outcome, an appropriate regression model is estimated for the positive outcome [25]. For part one of the applied models a logistic regression analysis was chosen to predict whether or not patients would utilize resources related to the respective costs figure. As recommended in the literature [20, 21, 24, 26‐28], a generalized linear model (GLM) with the log link and gamma distribution was chosen for the second part. In order to first analyze the temporal development of the cost estimates between the two treatment groups, an interaction term between treatment group and observation period was used. The cluster option was used to address the fact that multiple monthly cost estimates are included in the dataset for the same patient. Then, the respective resource uses were summarized over the one-year period for each patient in order to analyze the treatment-related cost differences across the entire one-year period. Finally, a budget impact analysis [29] from the healthcare perspective is performed by combining the costs of a hypothetical number of 20 patients receiving telemedically supported case management with the actual results of summarized resource uses in the two groups. One-way sensitivity analyses are carried out to assess the impact that changes in a certain parameter will have on the analyses’ results. Marginal means from all regression analyses are shown on the raw scale (€ per observation period). All analyses were performed using Stata 14 (Stata Corp., Texas. USA).

Table 1 provides an overview of baseline patient characteristics. The medical results of the RCT this study is a part of, including more detailed statistics, have been published elsewhere [8].

Table 2 provides estimates regarding the different cost figures over the three observational periods. Between month 0 and month 3 after transplantation, the number of out-patient care visits is comparable between the two groups. Between month 3 and month 6, however, the group of patients with telemedically supported case management required fewer medical specialist consultations, which is associated with average cost savings of € 293.54 per patient (p = 0.036) for this period. Between month 6 and month 12 after transplantation, these cost savings increase to € 419.12 (p = 0.089). Over the entire one-year period, patients with telemedically supported case management substantially less frequently visited medical specialists, resulting in total cost savings of € 761.37 per patient (p = 0.048). As far as general practitioner visits are concerned, on the other hand, the costs are relatively identical between both groups (p = 0.879).

As a result of early diagnoses and short delay between the first onset of symptoms and initiation of treatment, telemedically supported case management is associated with substantial cost savings related to unscheduled hospitalizations. The associated cost savings are highest between month 3 and month 6 after transplantation and add up to savings of € 3417.46 (p = 0.003) for the entire one-year period. The costs of scheduled in-hospital care, in contrast, are less affected by telemedically supported case management (cost savings of € 964.31 over the entire one-year period, p = 0.073). On the other hand, patients with telemedically supported case management more often applied for ambulatory visits at the university outpatient department, resulting in additional costs of € 117.39 (p = 0.035) for the entire one-year period. Overall, telemedically supported case management is clearly associated with cost savings due to less frequent and/or less intense medical service utilization during the one-year period (see Fig. 1). These savings are highest between month 3 and month 6 after transplantation (€ 2166.36, p < 0.001) and add up to total savings of € 4945.07 (p < 0.001) for the entire one-year period. Overall, patients in the standard aftercare group are associated, on average, with €10,449.28 of on-year medical service utilization costs while patients with telemedically supported case management are associated with €5,504.21 of on-year medical service utilization costs.

As a first sensitivity analysis, we calculate the net savings of our telemedically supported case management program depending on the number of patients participating. Due to the effect of fixed cost degression the program becomes cost-neutral at around ten participating patients, and was beneficial at 15 patients (see Fig. 2). After an initial rapid increase, the increase of savings begins to level off, increasing from €67 at a program size of ten patients to €1318 at 15 patients, but only from €2318 to €2569 from 25 to 30 patients. The results are projected to become statistically significant (p < 0.05) at a program size of 30 patients. Please note that the results for more than 20 patients imply the assumption that an additional 50% nurse position may can provide telemedically supported case management to more than 20 patients.

As a second sensitivity analysis, we kept the share of personnel expenses (an additional 50% nurse position for 20 patients) flexible. We assumed a 2.5% nurse position necessary for every additional patient with telemedically supported case management (which is equal to €1,781.25). As shown in Fig. 2, the program now becomes beneficial even from a very low number of patients. Please note that these results imply the assumption that the nurse position may be easily increased and decreased relative to the number of patients with telemedically supported case management. All results of the sensitivity analyses are shown in Table 3.

Finally we transfer our results to the German healthcare system. We recommend that the program should be implemented in bigger transplant centers like Freiburg, with more than 30 living donor kidney transplantations a year. The program is highly feasible and, if implemented in this way, would result in significant savings of upwards of €3,000 per patient and year of telemedically supported case management. According to EuroTransplant, a total of 645 living kidney transplantations were conducted in 38 German transplant centers in 2015 [30]. Of these, 12 centers conducted more than 20 transplantations (n ~ 407 in total). According to our results, routine implementation of telemedically supported case management in these 12 centers would result in annual savings of €791,033 according to the procedure numbers of 2015 and the results of our base case scenario.