This study estimated costs of healthcare incurred from primary care, outpatient, emergency, and inpatient settings, and from prescribed drugs for patients diagnosed with advanced soft tissue sarcoma. Twelve months before and post the diagnosis of aSTS were selected arbitrarily to make gross comparisons. The study findings are notable for a significant increase in costs in managing patients in the 12 months following their diagnosis of aSTS. The median overall survival of aSTS patients is estimated as 12 to 18 months with the current study reporting 15.4 months. The first twelve months seemed like an appropriate timeframe, to cover most treatment-related costs, however, it should be noted that more than two-thirds of aSTS diagnoses in our cohort had relapsed disease rather than de-novo advanced diagnoses (with an average time to relapse of approximately 24 months). It is plausible that a significant healthcare cost may have already been incurred well before the diagnosis of relapsed aSTS.
Cancer-related healthcare expenditure has been soaring in recent decades with increasing attention given to health economics in delivering appropriate care. Sarcoma is a rare malignancy with limited systemic therapy options in contrast to more common cancer types. Palliative-intent systemic therapy for advanced soft tissue sarcoma has shown little survival benefit to date. Reviewing the cost of managing this group is an important consideration with an emphasis on achieving an acceptable balance between costs and benefits of medical intervention. A few retrospective studies have been conducted to date estimating the cost of aSTS management, mostly as single-site retrospective chart reviews. One large study by Jonsson et al. was a multi-site, retrospective chart review, conducted in nine countries, which estimated country-specific costs across seven distinct categories of resources: including chemotherapy, other concomitant medication use, hospitalisation, hospice care, outpatient visits, radiotherapy and imaging(2). They calculated expected per-patient lifetime costs. A total of 213 patients were analysed with small patient numbers per site, mostly from France, limiting the interpretation of the data. The expected lifetime cost of care for patients with aSTS from the treatment onset was equivalent to AUD $117,450 per patient (95%CI, $92,241 to $152,385), in which most costs came from intravenous chemotherapy (31.7%). The patient population in this study was limited to those with favourable responses to chemotherapy, which they defined as achieving stable disease or better according to WHO or RECIST criteria within at least one line of chemotherapy. Other costs were attributed to inpatient care (24.8%), concomitant medication (11%), and oral chemotherapy (8.9%). The economic burden to the healthcare system was deemed considerable but with some notable differences across countries with regards to differences in intensive use of supportive medication, frequent inpatient care. In another similar study by Mytelka et al., a retrospective medical record review from the UK, Spain, Germany, and France, the healthcare costs were largely driven by drug acquisition and administration costs(3). As more novel drugs are approved in this patient cohort, the cost is likely to increase as noted by the change in treatment patterns in the study by Nadler from the US community practices(9).
Healthcare costs in our patient cohort were dominated by inpatient hospital care delivery, accounting for 80% of the total costs for the first year. This is a substantially higher proportion than what was reported in the above international studies, most likely due to differences in the patient population. Our study looked only at prescription medicines, considered relevant to cancer care, namely anti-neoplastic agents, anti-emetics, anti-infectives, and immunostimulants. Nonetheless, the PBS costs only accounted for 5%, considerably less than costs incurred from other healthcare services. More recently developed novel therapies for aSTS such as Pazopanib and Eribulin were only added onto the PBS reimbursement in 2013 and 2017 respectively(10, 11). Therefore, the cohort in this study would have missed the opportunity to access these drugs, which are now more readily accessible. Up to 18% of patients in this study participated in clinical trials and/or accessed off-label drugs on patient access program or via self-funding. These costs, either absorbed by hospitals or other research funders or incurred by patients, would not have contributed to the overall costs reported in this study.
Large datasets such as MBS, PBS, VAED, and VEMD have not been accessed for research purposes traditionally as they were inherently designed for their billing and record purposes for regulatory bodies. In recent years, there has been increasing use of these datasets to elucidate aspects of patient care and healthcare delivery, which are often not easy to glean from clinical trials or retrospective studies. Goldsbury et al. estimated state-wide health services cost of cancer care in Australia, using data from The Sax Institute’s 45 and Up Study, a longitudinal study of over 266,000 people in NSW aged 45 years and older for incident cancers diagnosed from 2006 to 2010 (12). This study accessed data from Medicare and pharmaceutical claims, inpatient hospital episodes and emergency department presentations, similar to our study to show the mean excess annual cost of cancer care per case as AUD $16,222 for the year before diagnosis and AUD $33,944 for the first year post-diagnosis. These costs were calculated for each case by taking their total costs and subtracting the average cost for their matched controls. As we did not utilise a control cohort, we cannot directly compare this with our study. There was a substantial variation by cancer type and time since diagnosis. The study found that the cost incurred from the initial treatment phase was higher for those with shorter survival time after diagnosis and lowest for those who survived more the five years. Consistent with the result from our study, the main driver of the excess costs was inpatient hospital costs, comprising 68% of costs for the first year after diagnosis, while 13% was for prescription medicines(12).
Our study highlights the complexities in the Australian healthcare system, which is difficult to assess as a whole. Some of the costs absorbed by the state government are not accounted for such as pathology and radiology services delivered when patients are admitted to the hospital. The costs of cancer care at the societal level were not included in our study. This would have necessitated including indirect costs incurred by patients, families, and carers in providing unpaid care and productivity losses as a result of premature mortality and morbidities adding to the socioeconomic burden. This additional aspect of cancer care is equally important when assessing the financial impacts of cancer care. Furthermore, timely access to data was hindered by the often laborious and prolonged process of obtaining regulatory approval for data access, which took approximately two years in total. This limited the expansion of the patient cohort to more contemporary datasets. The primary hindrance in conducting linkage studies in Australia remains the time delay associated with navigating the regulatory process for data access. This obstacle may differ from other countries that have more harmonised jurisdictions.
Limitations of this study are worth noting given the retrospective and descriptive nature. Multiple data sources rely on the local jurisdiction process for their accuracy. The extent to which the costs incurred in our patient cohort are generalizable to other Australian sarcoma referral centres is unknown, but with a reasonable sample size, the findings from this study give a good insight into the healthcare resource utilisation costs and will serve as a baseline for future studies. In future studies, a larger sample size could be employed to explore potential predictors of healthcare resource utilisation and costs.