Losartan Reduces the Costs Associated With Diabetic End-Stage Renal Disease

RESEARCH DESIGN AND METHODS

Study design

The RENAAL study design and results were reported by Brenner et al. (3). Briefly, the RENAAL study was an investigator-initiated multinational double-blind randomized placebo-controlled clinical trial designed to evaluate the renoprotective effects of losartan in 1,513 patients with type 2 diabetes and nephropathy. Patients were randomized to losartan or placebo on a background of non-ACE inhibitor conventional antihypertensive therapy (e.g., diuretics, calcium-channel antagonists, α- or β-blockers, centrally acting agents, or some combination of these types of medications). Patients enrolled had type 2 diabetes and a urinary albumin:creatinine ratio of at least 300 mg/g on a first morning specimen and serum creatinine between 1.3 to 3.0 mg/dl. Of the patients, 97% were either receiving antihypertensive therapy or were noted as having had hypertension but were not receiving antihypertensive therapy at baseline. Baseline characteristics of the study population are summarized in Table 1. The study protocol was approved by the Institutional Review Board of each center, and all patients gave written informed consent. The primary efficacy end point was a composite of the time to first event of doubling of serum creatinine, ESRD, or death.

Economic evaluation

The objective of the economic evaluation was to evaluate the effect of losartan and CT compared with placebo and CT on health care resource use and costs. The prespecified primary hypothesis was that therapy with losartan compared with placebo would reduce within-trial cumulative ESRD-related costs after 3.5 years of follow-up. As supportive analyses, we examined the effect of therapy on total cost.

Health economic measures

ESRD-related costs.

The cost associated with ESRD was calculated for each patient by combining the number of days the patient experienced ESRD with the cost of ESRD over time. To adjust for both differential length of follow-up and death, we estimated the number of days with ESRD by subtracting the area under the Kaplan-Meier survival curve for time to the minimum of ESRD or all-cause death from the area under the Kaplan-Meier survival curve for all-cause death. Although there were a small number of kidney transplantations performed (three losartan, five placebo), for the purpose of the cost analysis, we assumed that all individuals with ESRD were treated with hemodialysis. Given that the cost of ESRD is greater at the onset of ESRD (5) and greater among individuals with diabetes (1), we estimated the longitudinal cost of ESRD for patients with diabetes using 1997 and 1998 data from the U.S. Renal Data System (1). The average daily cost attributable to ESRD for individuals with diabetes was estimated to be $267 per day the first 90 days after onset of ESRD and $147 per day thereafter.

Total cost.

Total cost was defined as the sum of the cost attributable to ESRD and the cost of losartan therapy. The cost of losartan was estimated based on the average wholesale price multiplied by the days on therapy by dose. The 2001 average U.S. wholesale price for losartan at 25-, 50-, and 100-mg doses was $1.43, $1.43, and $1.95 per tablet, respectively. We assumed that patients who discontinued study therapy incurred no additional medication costs. We did not include the costs associated with monitoring serum creatinine and potassium because this monitoring would be performed routinely in individuals with diabetes and renal disease (4). Given that there were no differences in the incidence of side effects between treatment groups (3), we assumed that the difference in the cost of side effects between treatment groups was zero. We also conservatively assumed that there were no differences in the cost of nonstudy medications between treatment groups because there was a small but not significantly greater use of nonstudy medications in the placebo group (3).

Analyses

To estimate costs, we adopted the perspective of a health care system responsible for all direct medical costs. All randomized participants were included in the analysis on an intention-to-treat basis. The null hypothesis was that there is no difference between treatment groups in mean ESRD-related cost at 3.5 years of follow-up. We compared the 3.5-year mean ESRD-related cost between treatment groups using a regression-based method (6). This method accounts for administrative censoring brought about by staggered entry into the trial and involves two stages: 1) estimation of the mean relationship between cost and survival time and 2) weighting of this mean relationship by the Kaplan-Meier probabilities of survival. The bootstrap method (7) was used to construct a 95% CI on the treatment difference (placebo-losartan). All costs were discounted at an annual rate of 3% and are reported in 2001 U.S. dollars. Sensitivity analyses were performed on the cost of ESRD and the cost of losartan.

RESULTS

Table 2 and Fig. 1A show the mean number of days with ESRD by treatment group and follow-up time. Among individuals with type 2 diabetes and nephropathy, losartan compared with placebo on a background of non-ACE inhibitor/non-angiotensin II receptor agonist (AIIA) antihypertensive therapy reduced the number of days with ESRD by 33.6 days (95% CI 10.9–56.3) or 31% per patient over 3.5 years. Over 4 years, the number of ESRD days saved increased to 46.9 per patient (19.1–74.7). Table 3 and Fig. 1B show the mean ESRD-related cost by treatment group and follow-up time. The reduction in ESRD days resulted in a decrease in cost associated with ESRD of $5,144 (P = 0.003) per patient over 3.5 years. The reduction in cost associated with ESRD increased to $7,058 (P = 0.002) per patient over 4 years. Figure 1C shows the ESRD-related cost savings per patient by follow-up time. Table 4 and Fig. 1D show the estimated net cost savings per patient by follow-up time. After accounting for the cost of losartan, the reduction in ESRD days resulted in a net savings of $3,522 (95% CI $143 to $6,900) per patient over 3.5 years. This saving increased to $5,298 ($954 to $9,643) per patient over 4.0 years.

Sensitivity analyses indicate that treatment with losartan saves on cost under a variety of scenarios. For example, treatment with losartan would still save on costs after 3.5 years if the cost of ESRD decreased by 68% or if the cost of losartan increased threefold.

CONCLUSIONS

The RENAAL economic evaluation has demonstrated that, in individuals with type 2 diabetes and nephropathy, treatment with losartan resulted in substantial cost savings from the perspective of a health care system responsible for all direct medical costs. These results are meaningful from many perspectives.

From a managed care perspective in the U.S., these savings are notable because they accrue during the 30-month period when the managed care organization is responsible for the health care costs of a person developing ESRD before that person is eligible for Medicare coverage (8). For example, we found that for every dollar spent treating patients with type 2 diabetes and nephropathy, $3.17 in ESRD-related cost savings would be realized over 3.5 years. For the Medicare program in the U.S. where ESRD accounts for 5% of the total Medicare budget, treatment with losartan could potentially reduce a portion of these expenditures in patients with characteristics similar to the RENAAL study population. This could also be the case outside the U.S., where ESRD represents a significant line item on many national health care system budgets (9,10).

From the perspective of the public health system in the U.S., the results of the RENAAL study are consistent with the goals of Healthy People 2010 (11), where objective 4.1 is aimed at “reducing the rate of new cases of end-stage renal disease.” In particular, the addition of losartan to the treatment regimens of 100 patients with type 2 diabetes, nephropathy, and characteristics similar to those of the RENAAL patients could be expected to lead to a reduction of 6.3 cases of ESRD and 9.2 person-years with ESRD over 3.5 years. This reduction in years with ESRD would translate into a $514,000 reduction in the cost of ESRD and a $352,000 net savings over 3.5 years for these 100 patients. From a societal perspective, these savings represent an underestimate, given that the costs do not include patient out-of-pocket costs or productivity losses.

These savings are noteworthy for a number of other reasons. First, few medical interventions in health care actually save on costs (12). However, many health programs aimed at reducing ESRD have been shown to be cost-effective and even cost-saving. For example, Rodby et al. (13) demonstrated that treatment of individuals with type 1 diabetes and nephropathy using the ACE inhibitor captopril saves on costs. These results are consistent with the findings of this study given both populations studied were at high risk for a high-cost outcome (i.e., ESRD). Second, and unlike the Rodby economic evaluation (13), the effect of therapy on net costs was statistically significant (P = 0.041). This finding is unique to losartan given no other antihypertensive or angiotensin II antagonist agent has demonstrated a significant reduction in the incidence of ESRD in individuals with type 2 diabetes and nephropathy. Third, sensitivity analyses indicate this net cost savings is robust across a wide range of values for the cost of ESRD and losartan.

A reasonable question is whether these within-trial savings would be sustained over the long-run. This question is beyond the scope of this analysis, as this analysis would require developing a model that projects the long-term consequences of therapy. However, the data shown in Fig. 1D demonstrate that the cost savings increase over time between 2.5 and 4.0 years. Moreover, given that the competing mortality rate is so high in this population, it is likely that these savings would persist because individuals would likely die from other causes before reaching ESRD. Regardless, for many decision makers, these near-term savings are meaningful.

In summary, losartan reduced the estimated number of days with ESRD for patients with type 2 diabetes and nephropathy by 33.6 days over 3.5 years. This reduction in ESRD days resulted in a decrease in cumulative ESRD-related cost of $5,144 per patient after 3.5 years. After accounting for the cost of losartan, the reduction in ESRD days resulted in a net savings of $3,522 per patient after 3.5 years. These savings increased with greater duration of follow-up during the trial. These findings demonstrate that treatment with losartan in patients with type 2 diabetes and nephropathy not only reduced the incidence of ESRD, but also resulted in substantial cost savings.