Cost Utility Analysis of HD and CAPD in Ministry of Health Dialysis Centres, Malaysia

BACKGROUND AND SIGNIFICANCE

1. Background

End-stage renal disease (ESRD) represents a major problem for public health, and is a severe disease affecting hundreds of millions of people in the world and increasing rapidly. It brings about complex implications to social and economic structures of every nation. Renal replacement therapy (RRT) is a standard choice of treatment in patients suffering from ESRD which include hemodialysis (HD), peritoneal dialysis (PD) and transplant. The National Center for Chronic Disease Prevention and Health Promotion estimated that in 2014, more than 20 million people in the US may have different levels of Chronic Kidney Disease (CKD). According to Fresenius Medical Care, information gathered from 150 countries worldwide showed number of patients being treated globally for ESRD was estimated to be 3,200,000 at the end of 2013 and, with approximately 6% growth rate, continues to increase at a significantly higher rate than the world population.

Among these patients, 2,522,000 were undergoing dialysis treatment HD or PD and around 678,000 people were living with kidney transplants. The annual growth rate of HD, PD and transplant were estimated around 6-7%, 8% and 4-5% respectively. Reported incidence of ESRD across the globe showed an important trend ranging from under 100 to over 2,000 patients per million populations (pmp) in 2013. Taiwan, Japan and the USA continue to have some of the highest rates with 3170 pmp, 2,620 pmp and 2,080 pmp respectively. It averages about 1,090 p.m.p. in the 28 countries that make up the European Union (EU). From a global view, most dialysis patients can be allocated to three major geographical regions: the USA, the EU and Japan. Around 44% of all dialysis patients are treated in these 30 countries. Estimates on the Global Burden of Disease (GBD) for years 2000-2012, conducted by the World Health Organization (WHO), indicated that kidney diseases were responsible for 864,226 deaths (1.5% of the total number of global deaths) and 29,685,826 disability adjusted life years (DALYs) lost, corresponding to 1.1% of all global DALYs lost in 2012. In Malaysia, according to the 22nd Report of the Malaysian Dialysis and Transplant Registry (MDTR) 2014, 6107 new HD cases were reported representing an acceptance rate of 203 pmp while new peritoneal dialysis cases totaled 948, representing an acceptance rate of 31 pmp in 2014. Overall, the total number of HD and PD patients increased to 31,497 and 3270 respectively, giving a prevalence rate of 1046 pmp and 109pmp respectively from 2004 to 2014. A total of 3521 patients of HD and 494 of PD were reported died in 2014. In addition, the number of dialysis centres for the whole of Malaysia increased from 205 in 2000 to 758 in 2014 giving a rate of 13 pmp in 2004 and 25 pmp in 2014. The increase in dialysis centres was mainly contributed by the private dialysis centres which had tripled from 6 pmp in 2005 to 14 pmp in 2014.

The 22nd MDTR report also stated the number of new transplant patients was 172 in 2005 and this decreased to 81 (3 pmp) in 2014. The reduction in the number of new transplants was mainly due to reduction in commercial transplantation performed overseas. The number of functioning renal transplants increased from 1716 in 2005 to 1907 in 2011 but declined to 1844 in 2014. The Global Burden of Disease Study (GBD) 2010 conducted by Institute of Health Metrics and Evaluation (IHME) revealed that among the total number of years of life lost (YLLs) due to premature death in Malaysia, kidney disease ranked 14th from 25 causes of death with 62 YLL per thousand population which accounted for 1.1% of the total premature death. Ischemic heart disease, cerebrovascular disease, and lower respiratory infections were the highest ranking causes in 2010. Besides, the figures also show the Malaysia ranked 13th relative to the same comparator countries for the leading causes of DALYs for kidney disease in 2010 with 1 as the best performance and 15 as the worst.

ESRD imposes significant economic consequences in terms of direct loss of gross domestic product as a result for the management of patients with ESRD. The expenditure for the management of patients with ESRD in developed countries accounted for 2-3% of total healthcare expenditure, while ESRD patients represent only 0.02-0.03% of the total population. The United States Renal Data System reported for 2013 $30.9 billion medicare expenditure for ESRD (+1.6% versus previous year) (USRDS 2015). In the National Health Service (NHS) in England, annual spending on kidney care was estimated at £445 million in 2002 (∼£566 million in 2009-2010 prices) and programme budgeting analysis by the Department of Health estimated the total NHS expenditure on kidney care, including CKD, at £1.64 billion in 2009-2010. In Australia, kidney disease contributes substantially to health care expenditure in Australia and is increasing much faster than expenditure on total health care. In 2004-2005 it accounted for 1.7% of total expenditure ($898.7 million), an increase of 33% since 2000-2001 ($573.6 million).

In Malaysia, according to the report by Malaysia National Health Accounts (MNHA), Ministry of Health, total health expenditure was 4.49% per of GDP (RM 42,256 mil) with 6.22% of General Government Health Expenditure (GGHE) as percent General Government Expenditure (GGE) in 2012. Although limited data available of ESRD or CKD expenditure in Malaysia, the estimated cost of dialysis in 2005 reported to be RM379.1 mil. NHMA (2013) report stated that, in 2005, total health expenditure was 3.58% per of GDP (RM 19,447 mil).

II. Statement of Problem

The observed World Kidney Day annually beginning in March 2006 sends a clear message to the public, government health officials, physicians, allied health professionals, patients, and families that 'CKD is common, harmful, and treatable'. Renal replacement therapy is available in three different modalities which include haemodialysis, peritoneal dialysis and kidney transplantation from either a living donor or deceased donors.

Kidney transplantation offers a nearly normal life in both, quality of life and survival, and is considered the optimum treatment for eligible patients. Despite renal transplants from live donors, organ shortage remains a worldwide problem producing increasing waiting lists for transplantation and an inevitable necessity for dialysis treatments. According to Electronic Malaysian Organ Sharing System (eMOSS), there are around 19,500 people on the waiting list for kidney transplants in Malaysia as of 31st May 2016. New transplant rate in Malaysia is very low at 3 pmp in 2014. Hence, most of the patients require dialysis therapy. There are close to 35,000 are on dialysis (PD and HD). As compared to renal transplantation, dialysis is less effective in terms of 'survival' and 'quality of life'. In spite of a notable growth in dialysis provision rates by more than 10-fold between 1993 and 2014, Malaysia is still incapable to provide universal access to RRT.

The increased of dialysis centres are driven by the private dialysis centres which had almost tripled from 6 pmp in 2005 to 14 pmp in 2014. Nongovernmental organization (NGO) centres had only increased from 4 pmp in 2005 to 5 pmp in 2014. In contrast, the rate of growth was stagnant in the public sector, 5 pmp in 2005 and 2014. Most of the increases in the private dialysis centres occurred in the more economically developed west coast states of Malaysian Peninsula. Public sectors still provides most of the dialysis centres in the economically disadvantage states. Moreover, prevalence of ESRD is expected to rise significantly in the foreseeable future. Thus, the sustainability of dialysis therapy (HD and PD) in the future is uncertain.

III. Justification

ESRD is a chronic and serious illness with significant health consequences and high-cost treatment options. Despite rising economic pressure on the collective health care system, cost effectiveness studies of managing ESRD treatment are scarce in this country. An economic evaluation of HD and CAPD in MOH hospitals, Malaysia was conducted in 2005 which only included cost per life saved. This study was expanded in 2007 by incorporating utility measures to obtain the cost utility for both HD and PD. The measurement of improvement in quality of life was only associated with haemoglobin.

In addition, it is worth to note that both studies were performed based on 2001 costs data. Obviously, the cost of treatments in medical settings has changed tremendously according to current economic situation. For example, the cost of erythropoietin (EPO) was not adapted in 2007 study from the earlier study as there had been large reductions in the price of EPO, an Erythropoiesis Stimulating Agent (ESA) following the in introduction of EPO-β to the MOH in early 2004. As a result, the utilization rate of ESAs increased to 91% of patients on HD and 80% of patients on PD in 2013 as compared to 74% of patients on HD and 63% of patients on PD in 2004. In regards to the utility score, they also found that the quality of life of HD and CAPD patients were high by international standards (UK value set). Hence, a Malaysian EQ-5D questionnaire value sets was produced that can be used in cost utility studies.

Given the low organ donation rate and continual global growth in ESRD population, it is necessary and important to carry out effectiveness evaluation of HD and PD in terms of cost utility analysis so that the expenditure of its management can be justified without compromising outcomes. In economic evaluation studies, the choice of perspective (societal or provider) has an important effect on the cost-outcome components assessed and used in the evaluation. This study will be conducted from Malaysia Ministry of Health perspective. The information obtained may facilitate financial planning and health policy decisions by Ministry of Health regarding health care allocations for a more sustainable dialysis treatment program in the future.

IV. Objective (s)

General Objective

This study aimed to compare the cost-utility between HD and CAPD from a Malaysian Ministry of Health perspective by estimating cost per quality adjusted life year (QALY).

Specific Objectives

1. To determine the socio-demographic characteristics of patients on HD and CAPD.
2. To evaluate the mean cost of HD and CAPD treatments.
3. To determine life expectancies of HD and CAPD patients.
4. To assess predicted and measured quality of life (QoL) values in HD and CAPD patients using EQ-5D-3L questionnaire.
5. To evaluate the cost utility as cost per QALY gained in HD and CAPD patients.
6. To evaluate the cost-effectiveness of varying levels of PD use versus current practice using a hypothetical cohort Markov model.

METHODOLOGY

1. Study design

This study is commencing in two phases. Phase 1 is a prospective multicentre study to determine the cost utility of HD and CAPD from MOH perspective using activity based costing (ABC) and step-down approach. Costs of interventions (HD and CAPD) are obtained from the participating dialysis centres and dialysis patients' medical records. Health related quality of life (HRQoL) or utility score is evaluated using EQ-5D-3L questionnaire. Data from National Renal Registry (NRR) will used to estimate of participating dialysis patients. The data collection period is from 1st October 2016 to 30th September 2017. Phase 2 involves the development of a hypothetical cohort Markov model using TreeAge software with inputs from the earlier phase and other secondary data including transition probabilities and incident dialysis patients to evaluate the cost utility by varying levels of CAPD use versus current practice.

Phase 1 Cost utility analysis of sampled data

Stage 1: Selection of participating centres Stage 2: Patient recruitment Stage 3: Cost analysis Stage 4: Measurement of quality of life Stage 5: Measurement of patient survival (based on secondary data from NRR) Stage 6: Cost utility analysis Stage 7: Sensitivity analysis

Phase 2 Cohort Simulation Markov model Stage 1: Development of Markov Model using TreeAge Pro Software Stage 2: Inputs of cost and health utilities Stage 3: Transition probabilities Stage 4: Scenario analysis Stage 5: Sensitivity analysis

The sampling frame of a participating centre is all MOH state hospitals with HD and PD units that submit their patients' data to the Malaysian National Dialysis and Transplant Registry (MDTR) via National Renal registry (NRR). The MDTR was set up in 1992 to collect data for patients on RRT. The registry functions to describe the natural history of ESRD, describe the characteristics of patients with ESRD, its management, patients' survival and treatment outcomes (NRR). Following are the inclusion and exclusion criteria for selection:

1. Inclusion criteria

• Centres that commenced operations before 2011.
• Awarded compliance certificate and contribute >80% annual treatment return according to the MDTR.
• Possesses adequate clinical and administrative data. ii. Exclusion criteria
• Private and NGO dialysis centres.
• Centres with less than 50 CAPD and 50 HD patients.
• State hospitals from Sabah and Sarawak due to logistic reasons.

Following the inclusion and exclusion criteria, five state hospitals have been selected to participate in this study using purposive sampling including Hospital Kuala Lumpur and Hospital Tengku Ampuan Rahimah (Central), Hospital Tengku Ampuan Afzan (East), Hospital Sultanah Aminah (South) and Hospital Pulau Pinang (North). A purposive sampling was used after considering logistics problem, availability of a senior nephrologist to become principal site investigator at each sites, availability of medical assistants and nurses to assist on data collections and approval from the respective site's head of department and director. However, selected hospitals can be considered to represent MOH dialysis centres since they are distributed across various geographical regions except Sabah and Sarawak (Far East).

III. Patient recruitment

180 dialysis patients (90 HD and 90 CAPD) are included in the study. Patients are sampled using a multistage random sampling. Firstly, a latest master list of patient from the selected participating centres is obtained from NRR (registered from 2011 to 2015) according to the inclusion and exclusion criteria mentioned above. Secondly, the list is separated according to HD and CAPD for each centre and sorted in ascending order based on the patients registered identification number in the database. Next, a simple random sampling is applied to select participating patients. Selected patients who agreed to participate in this study are be asked to read the information sheet before signing the consent form. They can withdraw in any point of the study.

IV. Study tools

The following forms are designed and being used to elicit information on patient's sociodemographic characteristics, commodities, baseline lab test, quality of life/utility index, and costs (capital + recurrent) :

Form 1: Patient Particulars and resource utilization

This form is designed to collect patient's sociodemographic characteristics such as age, gender, ethnicity, highest education level and average annual income and commodities such as hypertension, cardiovascular disease and anemia. Data related to vascular access surgeries, medications, imaging, laboratory tests, hospitalization, referral to non-nephrology clinics and patient status (alive/dead/transplanted/withdrawn) is collected via this form.

Form 2: Centre Costing Form

This form is designed to collect data on capital cost of dialysis centre including building cost, equipment/instrument used for dialysis and . Recurrent cost of dialysis related consumables, staff salaries and other consumables such as office stationery and utilities including water, electricity and telephone usage will be collected via this form.

Form 3: EQ-5D-3L

The quality of life of HD and CAPD patients will be assessed using EQ-5D-3L instrument. . EQ-5D is an instrument which evaluates the generic quality of life developed in Europe and widely used. In general, the EQ-5D questionnaire, descriptive system measures health status in five dimensions of morbidity, self-care, usual activities, pain/discomfort and anxiety/depression. Each dimension has three level of severity, namely 1) no problem, 2) some problem and 3) extreme problem. The descriptive system contains a total of 243 theoretically possible combinations of EQ-5D questionnaire domains and problem levels, referred to as health states. It is also recommended that to comply with the Malaysian tariff to convert to utility scores using the "N3 Re-scaled VAS" scoring algorithm.

TreeAge Pro Software

TreeAge Pro Software will be used to construct the Markov model.