The influence of economic variables on the design of

Peritoneal Dialysis International, Vol. 27, pp. 359 363 Printed in Canada. All rights reserved. 0896-8608/07 $3.00 +.00 Copyright 2007 International Society for Peritoneal Dialysis COST/UTILITY STUDY OF PERITONEAL DIALYSIS AND HEMODIALYSIS IN CHILE Alejandro Pacheco, 1 Antonio Saffie, 1 Rubén Torres, 1 Cristian Tortella, 2 Cristian Llanos, 2 Daniel Vargas, 2 and Vito Sciaraffia 2 Nephrology Section, 1 Department of Medicine, University of Chile Clinical Hospital; Health Administration Institute, 2 Faculty of Economy and Business, University of Chile, Santiago, Chile In Chile the reimbursement/patient/year for chronic peritoneal dialysis (PD) is US$14,654 and for chronic hemodialysis (HD) US$10,909. However, no study comparing global (direct plus indirect) costs has been performed in our country. Our research objective was to compare global costs and quality of life between the two therapies. Patients (n = 159) from five selected dialysis units in Chile [57 patients on PD (50 on automated PD) and 102 on standard HD (3 4 hours weekly)] were retrospectively studied. No patient had previously received the alternate therapy. Items analyzed were quality of life, customer satisfaction, direct and indirect costs, annual global costs, and cost/utility index. Mean age on HD was 54.14 ± 16.01 years and on PD 49.76 ± 18.88 years (p > 0.05). No differences in the distribution of diabetic patients between the therapies were found. Hemodialysis and PD groups did not have differences in the quality of life index, although there was better customer satisfaction with PD than with HD. Direct and indirect costs were calculated. We found significant differences in favor of PD in erythropoietin consumption (2.24 ± 1.57 vials/week on HD and 1.35 ± 0.85 vials/week on PD, p < 0.05) and working time (31.0 ± 13.3 hours/week on HD and 38.5 ± 12.2 hours/week on PD, p < 0.05). The quality life index (Health-Related SF-36 Health Survey) was 65.75 on HD and 66.88 on PD. Annual global costs were US$20,803 for HD and US$20,742 for PD. The cost/utility index was 3.16 for HD and 3.10 for PD. Patients on PD have an advantage related to erythropoietin consumption and working capacity compared with HD patients. Addition of related indirect costs to reimbursements gives a more accurate insight into treatment costs. Considering all these parameters, we did not find significant differences between HD and PD in quality life index, cost/utility index, or annual global cost in this Chilean end-stage renal disease population. Perit Dial Int 2007; 27:359 363 www.pdiconnect.com Correspondence to: A. Pacheco, Nephrology Section, University of Chile Clinical Hospital, Santos Dumont 999, 4 piso sector E, Independencia, Santiago, 838-0456 Chile. apacheco@redclinicauchile.cl Received 8 January 2007; accepted 23 February 2007. KEY WORDS: Hemodialysis; quality of life; cost-effectiveness; cost/utility; health economics. The influence of economic variables on the design of clinical management guidelines and health policy decisions with the view of optimizing limited healthcare monetary resources is increasing, particularly in Latin America. In general, new therapies and medical interventions offer potential health benefits but involve increasing costs. Studies on cost-effectiveness (comparing the costs of an intervention versus a measurable biomedical or clinical variable such as life expectancy or prevented years of sickness ) and cost/utility (comparing the costs of a health intervention with quality of life or, better yet, Quality-Adjusted Life Years) evaluate the balance between additional health benefits for a medical action (economic, social, or other) and costs associated with said action (1). Economy health evaluations frequently consider only so-called direct costs, which are those generated by the medical intervention per se and usually include medicines, professional fees, value of the procedure, research, complications derived from side effects, etc. However, they do not consider the so-called indirect costs, a concept that covers items such as losses due to low labor productivity, absenteeism, premature death, and early retirement (1). A more complete vision of this matter may be obtained by including both variables (cost-effectiveness and cost/utility) in what we define as global costs. International literature on the cost-effectiveness of interventions in end-stage renal disease (ESRD) exists, but literature that considers quality-of-life parameters (i.e., cost/utility) is scarce. Most studies approach the problem from the perspective of direct costs (2 6). Dialysis therapies have a high direct cost, produce quality-of-life changes, and their results with respect to morbidity and mortality are of a long-term nature. It is not easy to extrapolate these costs from one country to 359

PACHECO et al. MAY 2007 VOL. 27, NO. 3 PDI another due to large social, cultural, and provisional differences. Therefore, studies that evaluate local realities and include global costs are necessary in order to direct appropriate economic decisions referring to the dialytic management of ESRD. In Chile in 2005, there were 10 693 patients on hemodialysis [HD; 685 patients per million population (pmp)], with a rate of increase of some 10% during the past 5 years (7). Historically, chronic peritoneal dialysis (PD) has had a later development than HD. Its access is not universal or completely elective, being conditional on certain medical indications (no vascular access, hepatitis B-positive patient) or social circumstances (extreme distances from the HD center, especially in rural areas). In 2005, 581 adult PD patients were on record (37.2 pmp), however, with a 25.5% rate of increase between 2001 and 2004 (8). The principal portion of direct costs for these two interventions is well known, as their value (FONASA reimbursement) is set by law and is universal. For 2005 it was US$14,654/patient/year for patients on PD and US$10,909/patient/year for patients on HD (9). In our country, there are no publications studying the global costs of both therapies. The purpose of the present study was to compare the global costs and the quality of life of patients treated with HD and PD in Chile. PATIENTS AND METHODS Cost and quality-of-life information were obtained from official sources by 2005 and the information given by patients through surveys and interviews performed in the last 2 months of that year. Five dialysis centers that practiced both techniques and had more than 50 patients were selected. Three of these are in Santiago, Chile, one is in La Serena (northern Chile), and one is in Valdivia (southern Chile). The total number of patients in these five centers was 709, with 404 on HD and 305 on PD. The inclusion criteria used were as follows: patients between 25 and 70 years of age, in treatment longer than 3 months, and not coming from another renal replacement therapy (i.e., be native to PD or HD). Patients with mental or neurological disorders that might make their survey questionable, or who would refuse to cooperate therewith, were excluded. According to these inclusion/exclusion criteria, 230 patients were selected at random. All patients were given an Informed Consent form explaining the trial in which they were invited to participate, as well as its usefulness, confidentiality, and voluntary nature. Finally, 159 patients agreed to join the study, 57 on PD [including 50 on automated peritoneal dialysis (APD)] and 102 on conventional HD (3 4-hour sessions per 360 week). The trial used a survey that covered different aspects according to the proposed objectives: quality of life, user satisfaction, direct costs (reimbursement, hospitalizations, and medications associated with the treatment), indirect costs (losses in production or income, time spent in transportation and waiting, etc.), and global costs. To measure quality of life in a standard fashion, our survey included a Form SF-36 Health Survey that made it possible to include the concept of Health-Related Quality of Life (HRQL). The SF-36 questionnaire, a validated instrument for measuring health-related quality of life in adult populations, includes questions capable of evaluating a wide spectrum of health conditions of an individual. This questionnaire contains eight health dimensions to be evaluated, which in turn are summarized in two components: a Physical Health Component (PHC) and a Mental Health Component (MHC). For the purposes of the SF-36 questionnaire used in this study, the data successfully met the validity and reliability tests required for this kind of instrument. The survey was conducted by external surveyors duly trained by the researchers. The Spanish version of the SF-36 manual was used for the purposes of variable transformation and tabulation. The SPSS program v.10.0 (SPSS Inc., Chicago, Illinois, USA) was used for the analysis of results. The cases actually considered were the questionnaires with at least 80% of actual answers given. Results are shown as mean and standard deviation. Differences between groups were analyzed by a t-test of independent samples and Fisher s test. Values of p less than 0.05 were considered significant. RESULTS CHARACTERISTICS OF THE HD AND PD GROUPS A total of 159 patients completed the study: 126 from Santiago dialysis centers and 33 from other provinces (La Serena and Valdivia); 102 patients were on HD and 57 patients were on PD. Most patients on PD (50) were on APD performed during the night. Mean age of the HD group (55 men, 47 women) was 54.14 ± 16.01 years; duration of therapy was 55.4 ± 52.6 months. Mean age of the PD group (33 men, 24 women) was 49.76 ± 18.88 years; duration on PD was 26.1 ± 17.3 months. There were no significant differences between the HD and PD groups in relation to age and gender distribution, but there were differences in relation to time on therapy. As for schooling, 99% of the patients on HD had completed their basic education (mandatory 8 years in Chile) and 18.6% had attended college. Patients on PD had 100% completion of basic schooling and 31.5% had

PDI MAY 2007 VOL. 27, NO. 3 COST/UTILITY STUDY OF PD AND HD IN CHILE attended college (of borderline statistical significance: p = 0.06). QUALITY OF LIFE 29.8% PD 3.5% 1.8% HD 3.9% 1.0% 35.3% Quality of life, as measured according to the SF-36 questionnaire, showed a favorable trend for PD in seven of the eight dimensions analyzed (Figure 1). There were no significant differences found between the treatments in relation to gender or age. Analysis of the eight scales of the SF-36 was performed to estimate the PHC and MHC components. Upon studying the PHC and MHC by years of treatment, no significant differences between HD and PD were found in the number of years gained. When using the HRQL indicator, estimations of differences between the groups in relation to quality of life did not include Quality-Adjusted Life Years (quality of life years of life gained). In this way, the HRQL indicator was calculated as 65.75 for HD and 66.88 for PD (no significant difference). USER SATISFACTION Most (64.9%) patients on PD assessed their therapy as excellent, versus 35.3% of those on HD (p < 0.0005) (Figure 2). The percentage of dissatisfaction was very low for both therapies. DIRECT COSTS Four elements were considered in calculating direct costs: First, the FONASA schedule, or annual reimbursement, in 2005 was US$14,654/patient/year for patients on PD and US$10,909/patient/year for patients on HD (9). The second element was the use of intravenous iron: 64.9% 59.8% Excellent Good Regular Bad Figure 2 User satisfaction. The percentage of patients who assessed their therapy as excellent was 64.9% for PD and 35.3% for HD. 56.9% of the surveyed patients on HD indicated they had been prescribed this drug, versus 38.6% of the patients on PD. The dose prescribed for patients on HD was 0.94 ± 1.47 vials (100 mg) per week, versus 0.67 ± 0.62 vials/ week for PD patients (no significant difference); the estimated value of each vial was US$17.10 (10). According to this, the annual cost of this drug per patient was US$475 for HD and US$230 for PD. The third element considered was erythropoietin (EPO). In the HD group, 39.2% of those surveyed declared they had been prescribed this drug; average dose was 2.24 ± 1.57 vials/ week/patient [2000-U vials, each at an estimated value of US$16 (10)]. In the PD group, 40.4% of those surveyed declared they had been prescribed this drug; average dose was 1.35 ± 0.85 vials/week/patient (p = 0.001, for the indicated dose of EPO). The annual cost of this item was US$731 for HD and US$454 for PD. Finally, the fourth element considered was hospitalization. Patients on HD were hospitalized 0.55 times per year, with an average stay of 17.07 days; whereas PD patients were hospitalized 0.75 times, with an average stay of 7.11 days. As 0 to 100 scale 100.00 90.00 80.00 70.00 60.00 50.00 40.00 30.00 20.00 10.00 0.00 Physical Physical Emotional Corporal Mental Vitality Social General function role role pain health role health PD 86.14 59.21 69.59 72.46 70.11 59.04 68.20 50.26 HD 84.12 57.60 67.97 68.86 68.82 58.28 72.55 47.79 Figure 1 The eight dimensions of the SF-36 Health Survey. Part of the survey was applied to patients on peritoneal dialysis (PD; black bars) and patients on hemodialysis (HD; gray bars). 361

PACHECO et al. MAY 2007 VOL. 27, NO. 3 PDI patients did not clearly recall (or know) the cost their hospitalization incurred, days of hospitalization were indexed to the cost/bed/day in the common ward of a general clinical hospital in 2005. According to this, the annual cost of hospitalization amounts to US$2,769 for HD and US$1,328 for PD. According to this analysis, the total direct costs add up to US$14,884 for HD and US$16,666 for PD. INDIRECT COSTS The most outstanding indirect cost is that referred to as loss of productivity and unemployment. This calculation was made considering the patients health insurance systems (public or private) and their geographical origins (borough). According to this, they were assigned an average income based on publicly available CASEN report data (11). Patients on HD work 31.0 ± 13.3 hours/week and patients on PD work 38.5 ± 12.2 hours/week (p = 0.0001). The number of missed working hours and the percentage of income conservation were calculated, that is, the percentage of current income compared to the income patients had before they entered the dialysis procedure. Patients on HD had 68.68% income conservation, versus 78.15% on PD (no significant difference). Thus, what a patient in each therapy would stop earning annually (the so-called loss of productivity and unemployment ) is estimated to be US$5,508 for HD and US$4,061 for PD. The survey also considered information on transportation, estimated at US$411/patient/year for HD and US$15/patient/year for PD. The indirect costs amount to US$5,919 for HD and US$4,076 for PD. ANNUAL GLOBAL COSTS AND COST/UTILITY INDEX The global costs were obtained from the sum of direct costs and indirect costs associated with each treatment. In this way, we have an annual global cost of US$20,803 for HD and US$20,742 for PD. If we index these values to the HRQL indicator, we obtain a cost/utility index of 3.16 for HD and 3.10 for PD (Table 1). DISCUSSION Although different in size, the two groups surveyed were symmetrical in relation to their characteristics, and this validates the study. The only apparent difference between them was time on therapy (greater in HD). In performing the survey (which included the SF-36 questionnaire as a full part), no significant differences were found with respect to the dimensions with which the 362 quality of life of each group was evaluated. The final result obtained from this instrument showed that the HRQL indicator of each technique was almost equal; however, in the matter of user satisfaction, the difference in favor of PD was significant. In performing this direct costs analysis, which, for the purpose of this study, considered a summation of the fee for the procedure, the use of intravenous iron, the use of EPO, and the costs of hospitalization, the greatest differences between the two groups are the amount of EPO prescribed and the number of days of hospitalization, which were more favorable for PD. For indirect costs, the greatest difference was the number of hours worked per week (significantly greater in PD), which showed the differences in morbidity and social environment that these patients may have. In fact, in our study, the number of university graduates on PD was larger than the number on HD, and was borderline significant. The transportation cost per patient, although marginal, was added to the indirect costs because it explains the loss of useful time in accessing the treatment. The annual global costs were practically the same, close to US$21,000 for both HD and PD. As a result of the above, the cost/utility indicator is very similar for both therapies: 3.16 for HD and 3.10 for PD. CONCLUSION TABLE 1 Main Results of Indices in the Cost Analysis In this health economics study in a group of Chilean ESRD patients, HD and PD showed very similar annual global costs and cost/utility indicators. Therefore, economic reasons that only take into account or consider all or part of the direct costs should not be a limiting factor for the admission of ESRD patients to HD or PD therapy in our country. ACKNOWLEDGMENTS Hemodialysis Peritoneal dialysis HRQL indicator 65.75 66.88 Annual global cost US$20,803 US$20,742 Cost/utility index 3.16 3.10 HRQL = Health-Related Quality of Life on the SF-36 Health Survey. This study is part of the Study of the Economic Evaluation in Health: Cost-Effectiveness, Cost Utility of Peritoneal Dialysis and Hemodialysis in Chile and Its Implications for the AUGE

PDI MAY 2007 VOL. 27, NO. 3 COST/UTILITY STUDY OF PD AND HD IN CHILE Policy, performed by the Health Administration Institute of the University de Chile School of Economy and Business, with Baxter-Chile financing. We particularly thank the patients who kindly participated in the surveys, as well the facilities provided at the dialysis centers of University of Chile Clinical Hospital, Catholic University of Chile Clinical Hospital, Dialysis System-Santos Dumont at Santiago, Chile, Clinidial at La Serena, and the Dialysis Unit of Valdivia at Valdivia, and very specially the head nephrologists and peritoneal dialysis and hemodialysis nurses of these centers. REFERENCES 1. Palmer AJ. Editorial review: health economics what the nephrologist should know. Nephrol Dial Transplant 2005; 20:1038 41. 2. Hooi LS, Lim TO, Goh A, Wong HS, Tan CC, Ahmad G, et al. Economic evaluation of centre haemodialysis and continuous ambulatory peritoneal dialysis in Ministry of Health hospitals, Malaysia. Nephrology (Carlton) 2005; 10:25 32. 3. Allard B, Cogny-Van Weydevelt F, Bacquaert-Dufour K, Bénévent D, Lavaid S, Beaud J, et al. A cost-effectiveness analysis of continuous ambulatory peritoneal dialysis vs. self-care in-centre haemodialysis in France. Dialysis & Transplantation 1999; 28(2):70 4. 4. Arredondo A, Rangel A, de Icaza E. [Cost-effectiveness of interventions for terminal chronic renal insufficiency.] In Spanish. Rev Saude Publica 1998; 32:556 65. 5. Winkelmayer WC, Weinstein MC, Mittleman MA, Glynn RJ, Pliskin JS. Health economic evaluations: the special case of end-stage renal disease treatment. Med Decis Making 2002; 22:417 30. 6. Kirby L, Vale L. The Relative Cost Effectiveness of CAPD Versus Haemodialysis. Aberdeen, UK: Health Economics Research Unit, University of Aberdeen; 1996; Briefing Paper for the NHS in Scotland, No. 10. 7. Poblete H. 25 th Annual Report of Chronic Hemodialysis in Chile (as of August 31, 2005). Santiago, Chile: Chilean Society of Nephrology; 2005. 8. Fierro A. Peritoneal Dialysis Report 2005. Santiago, Chile: Chilean Society of Nephrology; 2005. 9. Annual Schedule of Fees of the National Health Fund (FONASA). Santiago, Chile: Ministry of Public Health, Chilean Government; 2005. 10. Consumers Drugstore Price List. Santiago, Chile: Asociación de Dializados y Transplantados de Chile; 2005. 11. National Socio-Economic Characteristics Survey (CASEN). Santiago, Chile: Ministry of Planning, Chilean Government; 2005. 363