Healthcare Resource Utilization and Associated Costs During the First 5 Years After Diagnosis and at the End of Life: A Nationwide Cohort Study of Patients with Multiple Myeloma in Finland

Despite the evolvement of care and introduction of multiple new therapies and treatment options during recent decades, multiple myeloma (MM) is currently considered incurable [1]. In our previous study in Finland [2], we presented a median overall survival (OS) of 4.5 years for patients with newly diagnosed MM in 2015–2019. Multiple myeloma disproportionately affects the elderly, with a median age of 71 years at diagnosis [2]. The annual crude incidence was on average 8.8 while the crude prevalence at the end of 2019 was 32.7 cases per 100,000 in the population aged 18 years or older. The age-standardized (WHO World Standard Population [2000–2025]) incidence was estimated at 3.3 per 100,000. Despite the relatively stable age-standardized incidence [2], the number of new patients with MM is increasing due to the aging of the population. Thus, information on the expected resource use and healthcare costs among patients with MM in different phases of the disease is essential when evaluating the burden of MM to healthcare systems in the future.

Overall, healthcare costs associated with the care of patients with cancer are substantial, with costs and cost drivers differing across various phases of the disease. Based on existing literature [3‐6], the costs of cancer follow a U-shaped curve, the greatest costs occurring during the initial and terminal phases of care. European studies on MM suggest that the key drivers of healthcare costs during the active treatment phase are medication and inpatient stays [7‐10]. However, the estimated annual costs due to MM vary across studies.

A Danish nationwide study compared the mean annual healthcare costs (outpatient services, inpatient admissions, prescription drugs and primary health sector costs) of MM patients diagnosed in 2002–2005 (n = 1531) and 2010–2013 (n = 1972) [8]. Patients were stratified by age (<65 years or ≥65 years) and followed up for a maximum of 3 years onwards from the diagnosis. The estimated total annual healthcare cost per patient decreased from €51,000 to €44,000 in younger patients and increased from €27,000 to €31,000 in older patients between the time periods. In both age groups and time periods, the estimated annual costs decreased after the first year of follow-up. A French study of around 6400 patients diagnosed and treated for MM in 2013–2018 estimated the mean annual costs of disease (including healthcare costs, sick leave, disability pensions and transportation) per patient at €72,000 for the first year after diagnosis [7]. Over the whole follow-up (median: 22 months), the respective mean annual costs were €58,000 per patient. A Portuguese study of 1941 patients with prevalent MM reported that the average annual direct healthcare costs per patient were €31,000 in 2018 [9]. In a Finnish study of patients diagnosed and treated for MM in one large hospital district in 2013–2019, the average specialized care costs per patient-year were estimated at €33,000 and €20,000 for those treated with and without stem cell transplantation (SCT), respectively [11].

Cost-effectiveness analyses for new therapies are often carried out using a lifetime approach where all costs and benefits associated with different treatment options are estimated throughout a patient’s lifecycle, that is until death. Thus, the interest in end-of-life resource use and associated costs in MM has increased as well; however, the evidence is scarce, particularly in Europe. A single-center study conducted in the Netherlands in 2017–2021 estimated that the cost of hospital care during the last month of life was around €8000 per MM patient [12]. Inpatient stays and intensive care were the key cost drivers. The average end-of-life costs of hospital care were highest among patients who died within 5 years from diagnosis. In general, the end-of-life costs of patients with MM were higher compared with those of patients with other malignancies.

In Finland, every permanent resident is entitled to publicly funded healthcare services. Public health services are divided into primary care and specialized healthcare. Primary care services are generally provided by social and healthcare centers, and they include counseling, outpatient services in general medicine, oral healthcare and rehabilitation. Specialized healthcare services include diagnosis and treatment by clinical specialists, and they are mainly provided by hospitals and their outpatient clinics. Access to specialized hospital care is organized through a tiered treatment system with specific criteria and usually requires a referral.

After the implementation of the Finnish social and healthcare reform on Jan 1, 2023, the responsibility for organizing healthcare, social welfare and rescue services was transferred from over 300 municipalities and joint municipal authorities to 21 wellbeing services counties and the City of Helsinki. The wellbeing services counties are self-governing and responsible for adequate funding and allocation of healthcare services [13]. For example, medicines administrated in hospitals are acquired and funded by the wellbeing services counties whereas outpatient medicines dispensed at community pharmacies are paid by the patient and reimbursed by the Social Insurance Institution (Kela) if the medicine is under the national reimbursement scheme.

The treatment pathway for patients with MM in primary and specialized healthcare in Finland can vary according to the patient´s condition and characteristics, the stage and severity of the disease and the treatments provided. As we have previously presented, patients diagnosed with MM in Finland have multimorbidity and often have comorbidities mainly treated in the primary care setting [2]. Therefore, a holistic view of the total burden and costs associated with the treatment, healthcare resource use (HCRU) patterns, end-of-life care and palliative care of patients with MM in Finland is needed. Specifically, after the recent restructuring of healthcare organization in Finland, the newly established wellbeing services counties work under a fixed budget. This means that both the decision makers and care providers need to consider the costs associated with treating versus not treating patients and weigh their decisions against potentially achievable health gains.

To help fill gaps in the current knowledge, the aim of this nationwide register-based study was to determine all-cause and MM-specific healthcare resource utilization, patterns of service use and associated costs in Finnish patients treated for MM during the first 5 years from their first diagnosis, before end of life and during palliative care.

In this nationwide cohort study with 1615 newly diagnosed patients receiving treatment for MM in Finland during 2015–2020, we demonstrated that the total HCRU and healthcare costs were the highest in the first year after diagnosis and at the end of life. During the 5-year follow-up, the share of reimbursed outpatient medication was on average around half of the total healthcare costs, being the lowest in the first year after diagnosis and the highest in the fifth year after diagnosis. Moreover, we observed that the treatment of MM in Finland is highly focused on specialized healthcare, but at the same time patients are heavily treated in the primary care setting as well. To our knowledge, this is the first comprehensive nationwide study based on Finnish national healthcare registries assessing all-cause and MM-specific HCRU and costs for patients with MM. Distribution of healthcare contacts between primary and specialized care, utilization patterns over time, or end-of-life costs among MM patients have not been described previously in Finland.

Our observation that healthcare costs of patients with MM were the highest during the first year after diagnosis and during the end of life supports the results of prior research on MM and other malignancies [4, 6, 17, 18]; however, direct comparison between studies is difficult due to differences in methodology. The higher costs during the first year after diagnosis may be partly explained by more active treatment, including SCT and medications requiring administration in hospital. A single-center study of patients with MM in Finland demonstrated that the mean number of visits to specialized healthcare units for infusion medicines was higher compared with that observed for oral treatments [19]. In our study cohort, 30% of patients received SCT within the first year after diagnosis and, based on a subset of the cohort, 72% were administered bortezomib in hospital within a median of 15 days from diagnosis [2]. Accordingly, the trend in costs was clearly generated through the higher MM-specific specialized care costs during the first year after diagnosis compared with the following years. Particularly, the number of inpatient stays in internal medicine wards, where hematological cancers are mainly treated in Finland, seemed to reduce after the first year from diagnosis (data not shown).

We demonstrated that healthcare contacts increased in all contact categories relatively during 6 months before death compared with 12 months before death. Healthcare contacts further increased during palliative care in the primary care setting, but not in specialized care, indicating that, near a patient’s end of life, the focus shifts to primary care setting as patients are no longer actively treated for MM. The focus shift to the primary care setting is most likely explained by patients’ advanced age and the increased need for home care. Although MM-specific resource use decreases towards death, previous studies have shown that the end-of-life costs of patients with MM or other malignancies are higher compared with non-cancer patients [18, 20].

Our estimate of the mean HCRU costs PPY over the 5-year follow-up period was comparable to a previous estimate for patients with MM from one large hospital district in Finland [11]. More specifically, Vikkula et al. reported mean HCRU costs PPY of €33,000 and €20,000 for SCT-treated and non-treated patients, respectively. As 34% of the patients in that study were treated with SCT, the weighted average costs PPY equal those observed in this study (€25,000 PPY). A recent French study estimated the mean annual per-patient healthcare costs at €58,000 [7]. When combining all considered cost elements (including healthcare costs and sick leave) during the 5-year follow-up in the current study, total costs were estimated at €48,000 PPY. The difference in the estimated costs may be partly explained by differences in the follow-up times (22 months vs 31 months) and included cost elements. Specially, we were not able to include disability pensions in our analyses. Thus, we have most likely underestimated the costs due to productivity losses, since 30% of the identified patients were of working age at the time of diagnosis (under 65 years of age).

A Finnish cross-sectional study showed that the average share of hospital and outpatient medication in 2014 was around 30% of total costs, and costs varied across different cancer sites [21]. In our study with newly diagnosed MM patients, the reimbursed outpatient medication costs solely comprised 29% of the total observed costs in the first year, and the share increased up to 59% during the following years. During our study period, most of the novel MM treatment options introduced to treatment practice and to earlier lines of treatment were reimbursed outpatient medications. During the last 12 and 6 months of life, the share of reimbursed outpatient medication costs was 41% and 34% of the total costs, respectively, but <10% during palliative care. In summary, when considering the total cost, key cost drivers seem to be specialized care services during the first year, reimbursed outpatient medication costs during the following years and inpatient stays during palliative care. In the analysis regarding the last 12 or 6 months of life, the costs were distributed more equally across cost elements. The effect of reimbursed outpatient medication costs on the total costs, however, should be interpreted with caution. Reimbursed MM-specific outpatient medication costs may be overestimations as the three main cost drivers of the medication costs, lenalidomide, pomalidomide and ixazomib, were reimbursed under confidential financial agreements in Finland. This means that their total costs to society were lower than reported in this study. For future considerations, it should also be noted that the prices of lenalidomide products, which accounted for most of the costs of reimbursed outpatient medicines during this study period, have reduced by up to 98% after the loss of exclusivity and becoming generic in 2022 in Finland [22].

Our study highlights the importance of looking at all aspects of real-world data within a healthcare system and in a disease that mainly affects older people with multimorbidity. Although the treatment of MM in Finland is highly focused on the specialized healthcare, a majority of all-cause healthcare contacts of the patients with MM were captured in primary care, and only a small fraction of those contacts was MM-specific. This reflects the fact that patients with MM often have comorbidities that do not require treatment in specialized care. When considering only all-cause healthcare contacts, it is highly likely that HCRU directly related to the disease would be overestimated. Conversely, by collecting data on MM-specific events only, the holistic burden faced by patients with MM, especially in the primary care setting, would be underestimated.

Providing healthcare has an opportunity cost [23]. In healthcare systems with a fixed budget, this cost represents the health benefits forgone from other healthcare services that could have been provided with the same level of investment elsewhere in the system. Well-established methods of economic evaluation are used in many countries to inform decisions about the funding of new medical interventions [24]. New treatment options for MM are often first introduced as a later-line treatment option, and thus the corresponding economic evaluation is performed in heavily pre-treated patient populations with relatively low life expectancy and limited treatment options. The information provided in our study can be utilized as the most up-to-date information related to MM and to characterize patients’ end-of-life care in health economic evaluations. After the new social and healthcare reform in Finland, wellbeing services counties need to think more thoroughly about how to produce all statutory services with a fixed budget and already strained resources without compromising better outcomes for the patients. Our study highlights the fact that the HCRU and costs associated with the treatment of patients, especially if old and having multimorbidity, cannot be examined solely from a narrow disease-specific or organizational point of view.

The main limitations of this study are related to how administrative data were used for measuring clinical events and HCRU. First, initiation of palliative care was identified using diagnosis codes recorded in Hilmo. Based on clinical knowledge, it seems likely that both the occurrence of palliative care and its duration were underestimated, which may have further led to underestimation of HCRU and costs associated with palliative care. Second, MM-specific HCRU was identified using ICD-10 code C90.0 as the primary diagnosis. It is possible that not all MM-related contacts had C90.0 as a primary diagnosis code, which may have led to underestimation of the share of MM-specific HCRU and respective costs. For example, infections and skeletal issues may be coded with other ICD-10 codes even if the events were related to MM or its medication. Third, as patient-level data on medicines administered in hospitals were available merely for a subset of MM patients and these data were not comprehensive nationwide nor over time [2], no analyses by treatment lines could be conducted. Although the unit costs for outpatient visits and inpatient stays in specialized care include in-hospital medication costs, no detailed, per-medication analysis for costs were possible. Fourth, when using administrative data, some inaccuracies and missing data are unavoidable. While the validity and accuracy of the Hilmo data has been shown to be very good [25], no validation studies are available concerning AvoHilmo, which was a relatively new register during the study years. Fifth, as data on the contact type was missing for about half of specialized care contacts in 2019 and for 85% in 2020, we used imputed data when estimating the mean numbers of each contact type PPY in specialized care. To evaluate the validity of our imputation, we conducted a complete case analysis. In this analysis, the contact type distribution remained the same as in the imputed analysis for each of the 5 years of follow-up (data now shown). Finally, due to lack of a control group without MM, the incremental costs of MM could not be determined in this study. This study did not address potential differences in HCRU and costs across patient subgroups either.

During active treatment (i.e. before palliative care), MM in Finland is primarily treated in the specialized care setting, with outpatient medication and visits to specialized care being the main cost drivers. The total HCRU is higher during the first year after diagnosis compared with the following years and increases towards the end of patients’ lives when the focus shifts to the primary care setting. The results of this study can be used to inform decision makers in local welfare service counties, hospitals and HTA bodies on the holistic real-life HCRU and cost burden associated with the treatment of patients with MM. Specifically, these results can help to estimate the need for care and expected costs over time for newly diagnosed patients receiving treatment for MM. In addition, the results can be used to inform health economic evaluations of MM.