The costs of treating RA patients have increased rapidly in recent years. The discovery of new drugs over the past few years has made the treatment of this disease more complex and expensive, and although the drugs are very effective, they are associated with significant costs [25, 26]. However, reduced hospital admission rates, better functional status, and lower incidence of disability and absenteeism offset a large portion of pharmaceutical costs [13]. Hence, due to the lack of a comprehensive study on the economic burden of RA in Iran, the present research was carried out to determine the economic burden of RA on the patients referred to the medical centres affiliated to Shiraz University of Medical Sciences in southern Iran in 2019.
The results of this study indicated that the economic burden of RA was $ 3,744,967,993 PPP (2,935,245,184-4,655,906,153). In their study in Italy, Mennini et al. (2017) estimated the economic burden of RA at 2 billion euros [27], which is consistent with the results of the present study. In addition, Birnbaum et al. (2013) in the United States concluded that the economic burden of RA was $ 39 billion [11], which is inconsistent with the results of the present study. One reason for the inconsistency could be the addition of the cost of premature deaths as well as the intangible costs, accounting for 80% of the total costs.
The economic burden of RA alone was about 4.029% of the total health expenditures in 2019. The total cost of the health system in Iran was $93 PPP billion in 2019, accounting for 8.66% of the GDP [28, 29].
In addition, according to the results of the present study, the mean costs of the disease per patient were $11,925.98, $11,887.49, $12,803.45, and $12,155.08 PPP in Remission, Low, and Moderate to Severe status and Total, respectively. In this regard, the results are consistent with those of the studies by HU et al. (2018) in China[30], Huscher et al. (2015) in Germany [13], Malhan et al. (2012) in Turkey [31], and Ruof et al. (2003) in Germany [32].
The results also showed that DMCs were the greatest total treatment costs, accounting for 87.46%, 79.90%, 80.80%, and 84.42% of the total costs of the disease in Remission, Low, Moderate to Severe, and Total, respectively, suggesting that DMCs were the most important cost components for the patients with RA. Furthermore, it was found out that the highest share of DMCs was that of purchasing the main drug (80.50%, 80.23%, 78.97%, and 79.90% of the total direct medical costs at Remission, Low, and Moderate to Severe states and Total, respectively), the reason for which could be the high price of the drugs in Iran. In this study, the DNMCs accounted for 11.10%, 18.40%, 17.65%, and 14.07% of the total costs of the disease in Remission, Low, and Moderate to Severe states, and Total, respectively. Hsieh et al. (2020) conducted a systematic research examining 72 studies carried out from 2000 to 2019 on the economic burden of RA, and stated that the cost of the drugs was a major component of direct costs, and the costs increased over time [26]. In a study in China examining the burden of arthritis from a societal perspective, Hu et al. (2018) concluded that the mean direct costs were $ 2,559.06 ± 1,917.21 per patient a year, and the costs of purchasing the drugs accounted for over 50% of the total costs ($ 1,898.15 ± 1,283.89). On the other hand, age and income were significantly associated with indirect and intangible costs[30]. In a study on 689 patients in Turkey, Hamuryudan et al. (2016) examined the direct and indirect costs of arthritis and showed that the mean annual direct and indirect costs were € 4,954 (median € 1805) and € 2,802 (median € 608) per year. Drug purchase costs accounted for over 50% of direct costs (average € 2,777; median € 791), second to which were the RA-related consulting costs (average, € 1,600; median, € 696)[33]. Michaud et al. (2003) in the US examined the direct costs of RA and showed that the mean total costs of annual direct medical care in 2001 was $ 9,519 per RA patient. Drug costs were $ 6,324 (66% of the total costs), while hospitalization costs were only $ 1,573 (17%). About 25% of the patients underwent biological treatment. The mean total direct costs for the patients receiving biological agents were $ 19,016 a year, while the costs for those not receiving biological therapy were $ 6,164. In addition, direct medical costs of the RA patients who were in the worst quarter in terms of functional status were $ 5,022 more than the expenses incurred in the best quarter, but the costs decreased in all groups after the age of 65[34]. In this respect, the results are consistent with the present study.
However, Naqvi et al. (2020) conducted a study on 358 patients in Pakistan and indicated that the total annual costs of arthritis were $ 891.83, of which the mean costs of visits and physiotherapy sessions were $ 72.05 and $ 419.07, respectively. On average, the costs of drugs and medical devices were estimated at $ 63.25 and $ 49.13, respectively. Diagnostic and laboratory costs were $ 12.28, and the costs of travel were $ 40.95. Finally, the direct costs of RA management were $ 235.1. The results showed that physiotherapy accounted for the highest percentage of costs, and drugs accounted for 7.1% of the costs, and this is inconsistent with the results of the present study [35]. Catay et al. (2012) conducted a study on 165 patients in Argentina and found that the mean DMC and DNMC costs were $ 1,862 and $ 222, respectively. On the other hand, the mean ICs were $ 1,008 and the mean total annual cost was $ 3,093. Hospitalization accounted for 73% of the total DMCs, while medications and outpatient procedures accounted for 16% and 8% of the total DMCs, respectively. The results differ from those of the present study, the reason for which could be the high costs of hospitalization in that country [36].
The present study also showed that ICs accounted for the lowest total treatment costs and that they were 1.44%, 1.69%, 1.55%, and 1.51% of the total costs in Remission, Low, and Moderate to Severe status and Total, respectively. The results of this study are consistent with those of the studies by Xu et al. (2014) in China [37] and Osiri et al. (2007) in Thailand [38]. However, in a systematic review, Hsieh et al. (2020) showed that indirect costs were primarily related to absenteeism and disability, accounting for 39% - 86% of the total costs. Also, the indirect costs reported were highly sensitive to the estimation approach [26]. This is different from our study, one of the reasons for which could be the difference in the approaches used to estimate the costs. According to Zhu et al. (2011) who conducted a study on RA patients in Hong Kong, the mean total costs of RA were $ 9,286 per patient, of which more than 60% was related to ICs due to the productivity loss [39]. This might be due to the higher wages in Hong Kong. In their study, Filipovic et al. (2011) showed that productivity reduction and related costs varied in different studies, but all studies showed that RA associated with significant direct costs. They also argued that economic analyses that eliminated indirect costs underestimated the full economic impact of RA. This is inconsistent with the present study in which the indirect costs had a low share. Nevertheless, as mentioned, it is generally better to include ICs in the analyses to provide a more accurate estimate of the disease. However, it was stated that the methods used to calculate productivity loss had a significant impact on the results of IC analysis and that they should be carefully selected when designing such studies [40]. This could be one of the reasons why the ICs were different in this study.
One limitation of the present study was the self-report of the patients or their companions about DNMCs and ICs because they were more likely to forget or approximate some of the costs. In addition, due to the lack of accurate evidence on the number of RA patients in Iran, the opinions of some of the best experts in this field were used in the present study. It is worth mentioning that intangible costs were not calculated in this study due to the inability to measure them accurately.