Estimated indirect costs of haemodialysis versus peritoneal dialysis from a patients’ perspective at an Academic Hospital in Pretoria, South Africa

Chronic Kidney Disease (CKD) is an increasing public health burden [1‐3], with associated high economic costs [4‐8]. In 2017, the prevalence of CKD was estimated to be 9.1% worldwide with the highest burden of CKD is in sub-Saharan Africa, Oceania, and Latin America [9]. In 2018, kidney failure was the eighth leading cause of death among South Africans between the ages of 15 and 44, while it was the tenth cause of death among those who were aged 65 years or older [10]. There is an urgent need within countries to understand the current impact of CKD in terms of morbidity, mortality, and costs as a prelude to developing effective strategies to better manage the current situation as well as instigate future preventative strategies. This is particularly important in South Africa, with high prevalence rates for CKD, enhanced by increasing rates of,coronary vascular disease (CVD), with hypertension being a major risk condition for CVD, and diabetes along with high rates of the human immunodeficiency virus (HIV) [11‐13].

CKD has different stages, with kidney failure being the last and most severe stage. A concern is increasing prevalence of kidney failure worldwide, especially in LMICs [14]. Peritoneal dialysis (PD), haemodialysis (HD), transplantation or conservative care are currently available treatment modalities for patients with kidney failure [14, 15]. Kidney transplant is the preferred method of kidney replacement therapy (KRT) [16]. However currently, there are less kidney tranplants carried out annually in the South African public sector compared to those in the private sector. The current annual rate of kidney transplantation for the public and private sector is low at 4.6 pmp (per million of population) [11]. Current obstacles to increased rates, certianly in the public sector, include a lack of resources and healthcare personnel as well as attitudes of the public and some health care workers towards organ donation [11].

In South Africa, 70.2% of patients on KRT are on HD, while 9.1% are on PD and 20.1% have a kidney transplant [17]. Haemodialysis is performed by a health professional three times a week to remove waste products [18]. However, patients can perform daily PD at home to remove waste products [18]. In view of this, PD has been shown to improve patient outcomes compared to haemodialysis (HD) in terms of both survival and health-related quality of life (HRQOL) [19‐21] as well as being the most cost-effective option [22‐25]. PD is also an accessible treatment option among sub-Saharan African countries because it can be managed in low-technology environments with unreliable electricity [26]. Tang et al. (2019) also suggest that PD would be suitable in countries with working-age dialysis patients where the hourly wage is of a higher value [27].

Having said this, some patients prefer HD over PD [25]. This could be a result of not being able to manage fully PD due to physical and cognitive barriers and therefore requiring some form of assistance when dialysing [25, 28, 29]. This may be why in South Africa, low HRQOL has been observed in patients undergoing PD in view of the psychological, physical, social and economic factors [30].

Assessing the costs of different treatment options is important in this population as the cost of treating patients with kidney failure has increased in several years due to an increasing rate of non-communicable diseases (NCDs), especially CVD and diabetes, increasing the burden on the national health budget [31]. We know that CKD is one of the complications of diabetes, which is a concern in sub-Saharan Africa with currently sub-optimal management of patients with diabetes [32‐34]. This is a concern as the economic consequences of CKD are becoming a growing issue to patients, their caregivers, and entities responsible for payment [35], which needs to be addressed going forward. Alongside this, increasing recognition of the need to critically evaluate doses of medicines prescribed especially in hospital in patients with renal impairment [36]. The economic burden of CKD can be calculated including both indirect and direct costs [35]. Direct costs reflect the monies spent directly on treating a disease [28], while indirect or productivity costs reflect the time the patient loses for health care reasons [37].

The burden of indirect costs is more evident in patients with renal failure [35] since there is an increase in unemployment among these patients [38]. The economic and social development of the community is also affected as patients with CKD G5 between the ages of 30–60 years old and therefore part of the working population [39].

According to Yousefii et al. (2014), all costs, including direct, indirect, and intangible costs, should be considered during planning and policymaking within health systems [40]. This is important as the exclusion of indirect costs in research in South Africa may result in underestimating overall expenditure on dialysis [31, 41], which is important as South Africa is introducing universal healthcare [42]. Currently, there is limited knowledge regarding the costs of kidney failure in the public healthcare system in South Africa.

Consequently, this study aims to address this evidence gap by estimating the indirect costs of Kidney Replacement Therapy (KRT) from a patient’s perspective. The findings can be used with studies on direct costs to provide future direction to the South African authorities deciding on their future investment decisions. This is especially important with the introduction of universal healthcare in South Africa with considerable competing demands for available resources and a growing burden of NCDs.

This is the first conducted in South Africa to assess the indirect costs associated with different forms of dialysis among patients with CKD. The results from this study show that patients with PD had lower productivity losses than those with HD, which is seen as beneficial along with lower direct costs, as documented in the study by Makhele et al. (2019) and generally [22, 31, 47]. Overall, the study highlighted that patients on HD incurred higher indirect costs and experienced the highest unemployment rate compared to those on PD. Secondly, while patients using PD incurred less costs than those undergoing HD, there were issues related to PD. These included increased hospitalisation days and visits due to complications. A study undertaken in Tygerberg Hospital in Cape Town also found that the patients on PD verbalised more symptoms related to kidney failure and the complications of their dialysis modality compared to patients on HD [48]. Our findings though that there were higher hospitalisation costs among PD compared with HD patients contradict those from Sweden [49]. We are not sure of the reasons for this; however, it could be due to issues such as training, proper sanitation at home and a regular supply of electricity to fully undertake PD [50]. Concerns with sanitation and a continual supply of electricity are especially in the informal settlements in urban cities, where most of our patients reside. In 2019, only 83.1% of households in Tshwane had access to improved sanitation, and 95.1% of households had access to piped or tap water [51]. Poor sanitation and lack of water could lead to infections. Patients on PD must have a toilet in the home for disposing of dialysate and cleanliness requirements generally when applying PD requires clean water [52]. A study conducted in Senegal which looked at non-infectious complications of PD reported that mechanical complications such as catheter migration and metabolic complications such as hypoalbuminemia could be corrected by having experienced staff members inserting catheters and giving good nutritional advice before serious complications arise [53].

Similar to this study, other studies have found that the indirect costs of HD are higher than those for PD, which is perhaps not surprising since PD can be performed at home daily without patients having to regularly travel to the hospital for dialysis [27, 54, 55]. Similarly, in Taiwan, patients on HD had a greater productivity loss per month compared to PD patients with greater outpatient care [27]. Patients on HD tend to be less productive because dialysis reduces the time spent on routine activities. They generally perform fewer activities than they would like because of their physical health., They may also not have the strength to fully deal with the pain during and after dialysis, negatively impacting on their productivity [56]. In addition, HD patients typically experience higher transport costs travelling to the hospital than PD patients, who typically only visit the hospital once a month and some even once every three months, especially during pandemics. The travel time to the dialysis facilities also impacts productivity, as most patients in this study had to travel two to three hours when visiting the hospital. As a result, substantially more patients on PD were employed in our study compared to those on HD, enhanced by a greater flexibility with their time for dialysis, with HD patients having to be at the hospital typically three times a week. Similarly, a study conducted in Spain found that 28% of the patients undergoing PD continued working, compared with only 13% of HD patients [54]. This is important with Muehrer et al. (2011) finding that patients on dialysis who are unemployed may well experience physical and psychological problems, including anxiety, depression, sexual dysfunction, and loss of self-esteem [38].

One of the key factors the South African government has with implementing universal healthcare is to improve access to tertiary hospital services by reducing the distance patients have to travel to access such services. However, this will take time, especially following the economic consequences of COVID-19 in South Africa [57, 58]. Increasing the number of nephrologists through increased training and distributing them evenly across the healthcare sectors, will help with earlier diagnosis and increase the option for PD [50, 59]. Their educational input should also help to reduce the fear among patients that they might not be able to manage PD at home due to physical and cognitive barriers as well as fears generally with potential home modifications [25, 28, 29]. In addition, assisting working age patients to stay employed will positively affect societal costs and their well-being [38]. Seeking to instigate good secondary prevention for patients who are in CKD stage 4 or 5 to avoid dialysis would also be beneficial [49]. This could be achieved by increasing the opportunities for early diagnosis for patients with risk factors, such as those with diabetes and hypertension, given concerns with underdiagnosis in South Africa and prescribing patients with early-stage CKD protective medicines against CVD and renal failure to avoid or slow down the progression of disease [60]. This is because NCDs are one of the key health priorities in South Africa [42, 61]. Such activities though need to be combined with measures to enhance adherence to these medicines due to ongoing concerns [62‐64].

Whilst PD patients in our study had fewer caregivers, other studies have shown the need for caregivers, especially in ageing PD patients [28, 65]. Older patients will also have comorbidities such as diabetes mellitus with retinopathy and poor visual acuity, which could make it harder for them to perform PD without a caregiver [66]. Whilst our study showed that a lower number of patients had diabetes as a comorbidity or a cause for kidney failure, this could be due to the fact that there is more focus on preventing kidney failure in diabetic patients [31]. In addition, our findings were based on the replies of patients in the questionnaire rather than any formal diagnosis and may not be totally accurate This must be considered when reviewing potential policies to manage patients with CKD, especially in South Africa, where there is currently a limited number or no home visits by HCPs to assist patients with their PD, exacerbated by threats to healthcare workers [50]. It would also be beneficial to work with patients to reduce the time taken for PD if this is an issue that can be addressed with some patients spending up to four times longer to perform their PD than others. In addition, exploring ways to improve access to a continuous supply of electricity.

We are aware of a number of limitations with this study. Firstly, we only performed the study in one centre with a limited number of patients. Secondly recall bias could have affected the results of our study. Thirdly there were some disadvantages with using the Human Capital Method as11.5% of PD patients were employed; therefore, assuming the same wage for all participants this approach would underestimate the total cost in the group. In addition, the Human Capital Method also does not account for the possibility of patients being replaced at work, which is a possibility in South Africa given current unemployment rates. Another disadvantage of this method is that it measures the potential value of production loss due to sickness instead of actual loss which the friction cost method which would have accommodated this [44]. However, despite these limitations, we believe our findings are robust and do provide guidance for the future in decision making in this increasingly priority area.

Patients undergoing HD incurred a higher productivity loss than those undergoing PD, although the difference was insignificant. Whilst the results for this variable were uniform for HD patients, there were substantial differences with PD patients. Some patients took as low as 1.25 h every day to perform dialysis, while others took 5 to 6 h. There are also concerns about greater hospitalisation among PD patients, which needs further exploration to reduce associated morbidity and costs. This is especially as more HD patients compared to PD patients were also unemployed due to the lack of flexibility that comes with performing HD. Overall, early detection of CKD and appropriate treatment to avoid progression to kidney failure would be the optimal cost saving approach. Such approaches, including improved lifestyles, would also be beneficial generally for patients with CVD and diabetes to prevent other complications, and we will be following this up in future research projects.