Treatment costs and cost drivers among osteoporotic fracture patients in Japan: a retrospective database analysis

Osteoporotic fractures often have serious health impacts. For example, in a previous research, those with osteoporotic fractures had a higher mortality rate and decreased individual quality of life [1, 2]. It is established that a risk of osteoporosis and osteoporotic fracture increases as age increases [3], and Japanese population is going through the rapidly increasing elderly population. In 2015, people aged 65 years and older consisted 26.6% of total population, and it was estimated that this proportion would increase to 37.7% by 2050 [4]. Furthermore, particularly in Japan, the financial burden of osteoporotic fracture is immense. Moriwaki and Kamae estimated that the financial burden due to osteoporotic hip and vertebral fracture was 27.5 billion JPY (250 million USD when 1 USD = 110 JPY) in 2009 [5].

Under this situation, the health economic analysis in osteoporosis and osteoporotic fracture is featured more than ever. In recent years, new drugs, such as teriparatide and denosumab, have been launched in Japan with higher drug prices than conventional bisphosphonates. In line with the increasing varieties of osteoporosis drugs, attention by researchers has been paid to cost-effectiveness of these drugs. Ding et al. (2008) analyzed the cost-effectiveness of risedronate sodium compared with no-treatment [6]. Moriwaki and his colleagues examined cost-effectiveness of alendronate versus no-treatment in 2013 as well as cost-effectiveness of zoledronic acid versus alendronate in 2017 [7, 8]. Mori et al. (2017) reported cost-effectiveness of denosumab versus alendronate in 2017 [9]. In terms of treatment cost parameters such as the yearly treatment costs for fractures, most studies used the cost input retrieved from the Japanese medical fee schedule with a set of treatment assumptions or estimates based on small-sized patient population. This is because there has been no study which estimated the treatment costs of osteoporosis and osteoporotic fractures with large-sized real-world data (RWD). Given that a considerable size of claim database has recently become commercially available, estimating osteoporotic fracture treatment costs using them is considered valuable.

Therefore, the primary objective of this study is to estimate the current yearly treatment costs due to osteoporotic fractures as well as treatment costs of osteoporosis patients in Japan using large-sized RWD. It will not only help estimate the financial burden of osteoporotic fracture but also help conduct various health economic analyses. The secondary objective of this study is to identify cost drivers among osteoporotic fracture patients. We defined cost drivers as the factors associated with increased or decreased treatment costs of osteoporotic fracture hereafter and explored them via the retrospective database analysis.

This study estimated the total medical costs and the fracture treatment costs in osteoporotic patients in Japan. We also explored the common comorbidities among osteoporotic fracture patients, and examined the potential cost drivers for the treatment costs.

Although a few studies have reported the treatment costs incurred for osteoporotic fracture in Japan, the number of patients was limited and the hospitals from which the patients were recruited was underrepresented. Our cost estimation was based on the latest claim data of 12,898 patients including both inpatients and outpatients, thus considered to be adequately representing Japanese population despite limitation of the database itself. Kondo et al. (2009) investigated the treatment costs of 41 hip fracture patients in three hospitals in Japan [13]. They reported that the costs were $19,034 on average. Although their result was slightly more expensive than what our study has found, it would be mainly due to the different payment system used by the hospital. The hospitals studied in this study were using only DPC payment system (similar to Diagnostic Related Group [DRG] in US) whereas the study by Kondo et al. included the hospitals with non-DPC payment system. DPC payment hospitals tend to manage their resource more effectively, and therefore it may lead to the lower treatment cost estimates. In contrast, Hagiwara et al. (2000) found that the mean treatment costs of hip fractures with 101 patients who were 60 years and older were 1,470,000 JPY ($13,364 when 1 USD = 110 JPY) [14]. This is less than what we have estimated, probably because their estimate was based on the inpatient stay in one hospital and did not include outpatient or rehabilitation care after discharge. Furthermore, it was reported that an average treatment costs for vertebral fracture was 780,000 JPY ($7091 when 1 USD = 110 JPY) with 272 hospitalized patients aged 60 years and older [15]. This can also be explained by the different study populations. Our study includes patients who were cared solely by outpatient visits; therefore, our research produced less expensive estimate than the research that only recruited hospitalized patients in terms of vertebral fracture patient costs.

The treatment costs borne by the patients with the second fracture were significantly higher compared with single fracture patient, which is consistent with previous studies in USA and UK [16‐18]. In this study, those with the second vertebral fracture bore more than twice of total medical costs and fracture treatment costs when comparing their point estimates with those with a single vertebral fracture. This does not only suggest the importance of preventing the second fracture within 1 year from the index fracture but also indicate the great benefit of preventing the second fracture. From this perspective, prescription of osteoporosis drugs after the osteoporotic fracture can contribute in reducing financial burden of osteoporotic fracture.

The results of cost estimation indicate that total medical costs would be preferable in economic evaluation because the fracture treatment costs were almost only one third of total medical costs. Other two third would consist from non-fracture treatment costs such as doctor consultation costs, comorbidity treatment costs, and inpatient specific costs. The amount of these costs are not negligible, and hence they should be considered when evaluating the economic value of fracture treatment. To our best knowledge, there has been no study in Japan comparing the total medical costs and the fracture treatment costs, and the results would support the official guideline for the economic evaluation of drugs/medical devices recommending the inclusion of related comorbidity costs for cost parameters [19].

Interestingly, it was found that more fracture treatment costs were incurred among female hip fracture patients whereas less total medical cost was incurred female vertebral fracture patients, suggesting that more non-fracture (comorbidity) treatment costs were incurred among male than female. Male could be more likely to worsen their comorbidities. In fact, one study reported higher relative hazards of all-cause mortality after hip fracture in male during 0–1 year age interval [20]. Male may be more vulnerable after fractures, which lead to increasing non-fracture treatment cost.

Associations between comorbidities and treatment costs were explored, and nine comorbidities were detected as cost drivers. The associations with three medical conditions (hypertension, iron deficiency anemia, and diabetes) were consistent with previous research. Firstly, in our study, it was found that hypertension was associated with increased fracture treatment costs in hip fracture subgroups. This positive association was persistent when investigated in the total population. The association between hypertension and bone remodeling has been suggested [3], and one previous study showed that the BMD level of female patients with hypertension was lower than the BMD level of female patients without hypertension [21]. Another study further suggested that hypertension was a predictor for fracture nonunion [22]. Considering these empirical evidence, our result could additionally indicate that patients with hypertension have worse bone turnover, and it takes additional care to heal, leading to higher fracture treatment costs. This association was consistent with the total medical costs among hip and vertebral fracture patients.

Secondly, the result showed a strong positive association between iron deficiency anemia and treatment costs (both in the fracture treatment costs and the total medical costs among hip fracture patients). It was reported that several potential interactions between iron deficiency and bone loss (i.e., the role of iron in collagen synthesis or in vitamin D metabolism) [23]. Thus, it might be claimed that the patient with iron deficiency anemia bore more treatment costs due to its downside effect on bone metabolism. However, it should be noted that anemia can be an intermediate variable between other comorbidities and fractures as well. Anemia can be observed among the patients with chronic medical conditions such as CKD (the detail association between CKD and treatment costs will be explained later) [24]. Therefore, the estimated impact of iron deficiency anemia in this study may reflect the impact of such comorbidities on treatment costs.

Lastly, it is interesting to find that diabetes had an association with the decreased fracture treatment costs among hip fracture patients. One possible explanation for this negative association is the influence of some antidiabetic drugs. Although it has not been concluded yet, drugs such as insulin and metformin have had positive effects such as improved bone bridging or osteogenic effect in some experimental studies [25]. Therefore, patients with these antidiabetic drugs can have better bone quality and need less care. However, this should be further examined carefully as mixed results exist in this research area. For example, one recent study showed that the insulin use was negatively associated with bone mineral density increase [26]. In addition, this could be explained by behavioral aspect. One previous study in Japan which compared the healthcare resource use between osteoporosis diabetic patients and non-diabetic patients found that BMD testing was less frequently used among the patients with diabetes [27]. Although the reason behind this is unclear, our study results may indicate similar phenomenon.

Unpredicted positive associations were also found with three medical conditions (sleep disorders, CKD, and other functional intestinal diseases). Sleep disorders were positively associated with both the fracture treatment costs among hip fracture patients and the total medical costs among hip and vertebral fracture patients. Sleep disorders are common claims among the elderly population. In a previous study, it had potential associations not only with increasing risks of falls but also with osteoporosis [28], suggesting that sleep disorders may have some impact on bone metabolism. For example, Amstrap et al. (2015) showed that melatonin, the hormone regulating the sleep, improved BMD among women with osteoporosis [29]. Given this, our study may indicate that sleep disorders cause malignant bone metabolism among osteoporotic fracture patients, and consequently patients bear more total medical costs to recover from the fracture. Moreover, patients with CKD may also bear more total medical costs, although it was not observed in our analysis of fracture treatment costs. This may be because some osteoporosis treatment drugs such as bisphosphonates and painkillers can induce side effects such as hypocalcemia. Therefore, monitoring kidney conditions or treatments for such side effects may increase the total medical costs among CKD patients.

Lastly, other functional intestinal diseases had a positive association only with the total medical costs. Fracture patients would increase the risk of other functional intestinal disorder because of the physical inactivity during the hospitalization period. Vertebral compression fracture itself can cause constipation as complication [30]. Thus, it could be considered that those with other functional intestinal disorder represented the patient cohort with severe fractures and longer inpatient stay, which made their total medical costs higher than the costs of other patient group. Therefore, there may be no causal association between other functional intestinal disorder and treatment costs.

Our study also showed unexpected negative associations between treatment costs and the other three medical conditions (dorsalgia, disorder of refraction and accommodation, and COPD). Although the causal association between these medical conditions and treatment costs needs to be carefully interpreted with further research, potential causal pathways are hypothesized for each comorbidity as follows. The first one is dorsalgia. It had a negative association with the fracture treatment costs in hip fracture patients and the total medical costs in vertebral fracture patients, and the association between dorsalgia and fracture treatment costs remained in total population. The guidelines for prevention and treatment and osteoporosis suggest that dorsalgia is one of the factors for diagnosing osteoporosis [3]. General patients with dorsalgia during pre-fracture period might receive osteoporosis treatments earlier and therefore may not experience severe fractures. The second one is disorder of refraction and accommodation. It was negatively associated with the fracture treatment costs in vertebral fracture patients and total patient population, even after adjusting the demographic factors. One possible reason is that this variable may have worked as intermediate variable and reflected severity of other comorbidities such as diabetes, which is significant risk factor for retinopathy. The third one is COPD. It was demonstrated that COPD was associated with decrease in the fracture treatment costs among hip fracture patients. It was preserved even after the population was extended to total population. Given that the COPD is a well-known risk factor of osteoporosis and low BMD [3], this result looks counterintuitive. One potential association is that COPD patients often develop vertebral fractures, and therefore they could have already had preventive treatments before hip fracture, which would result in improving fracture recovery. However, further research is necessary.

The following limitations were recognized in this study. Firstly, a selection bias due to the nature of the database was possibly considered. The claim data used in this study were collected from acute care hospitals, allowing us to collect severer patients in Japan. On the other hand, these acute care hospitals may spend less treatment costs due to DRG-like payment system in Japan (DPC). In fact, it was shown that less treatment costs were incurred in DPC hospitals [13], which could underestimate total treatment costs. Furthermore, it could create the issue related to loss to follow-up. As there is no information regarding the clinical outcome of the patients, it is impossible to identify whether the patient was recovered, transferred to another care institution, or died. The Japanese government has introduced the policy that promotes specialization of medical institution into four categories (advanced acute-care, acute-care, rehabilitation-care, and long-term care) and collaboration between these hospitals in 2015 [31]. Hence, it is likely that the total medical costs of patients who were transferred to rehabilitation-care or long-term care institutions were not fully captured. In addition, accuracy of diagnosis record could be questioned. One study revealed that the validation indicators such as sensitivity or positive predictive value of diagnosis were varied depending on the medical conditions [32]. In this sense, generalizability of this study might be questioned as it is uncertain that the codes used in our study fully captured the osteoporotic fracture population. Finally, our study could not adjust patient behavioral and SES factors which could confound the results. Behavioral factors such as smoking and SES factor would not only affect patients’ health but also influence their treatment choice. These limitations should be considered carefully when interpreting the study results.

In conclusion, our study reported the treatment costs incurred to osteoporotic fracture patients in Japan and examined the cost drivers by analyzing the associations with comorbidities. To our knowledge, this is the first pragmatic study estimating the treatment costs of osteoporotic fracture patients through the analysis of large-sized RWD, and it showed financial burden of osteoporotic fracture in Japan including patients with the second fracture. This study also detected the novel associations between some comorbidities and the increase or decrease in treatment costs. The findings from this study would help clinicians, researchers, and policy makers understand the potential financial burden of osteoporotic fracture and would also support various health economic analyses in the future.