Introduction

Cancer, a widespread and devastating disease, affects millions of people worldwide, leaving a profound consequence on individuals and society as a whole [1]. The repercussions extend beyond health, affecting the finances and productivity of individuals. The impact is particularly pronounced in low- and middle-income countries, where healthcare resources may be limited, and individuals may face challenges in accessing affordable and effective cancer care [2]. From an economic point of view, the burden of cancer is multifactorial, encompassing direct medical costs, productivity losses due to illness and premature death, and the costs of caregiving [3–5]. This financial burden is exacerbated for those with limited economic resources. Lower-income populations grapple with a distinct set of challenges in managing their healthcare costs, with OOP expenses emerging as a significant and often overwhelming aspect of the cancer journey [6]. The ASEAN Costs in Oncology (ACTION) study in 2012 for example reported almost half of the households with cancer patients experienced catastrophic health expenditure. This was particularly true for patients in advanced disease stages and those from socioeconomically disadvantaged groups [7].

Studies have consistently highlighted the disproportionate burden of OOP cancer expenses carried by lower-income individuals compared to their higher-income counterparts [6,8,9]. This situation worsens for those residing in rural and remote areas who have to spend more on transportation to access cancer treatment [10–13]. A study reported that 25% of the Malaysian population lives more than 100 km away from radiotherapy facilities [14]. This issue is particularly significant for rural patients, especially those in Sabah and Sarawak, who potentially be forced to travel long distances and struggle to afford transportation and lodging. Beyond the financial toll, there is a cascading effect on productivity. The intricate dynamics between cancer and productivity loss are well-documented, with cancer survivors facing unique challenges. Frequent medical appointments, treatments requiring time off work, and the overall physical and emotional toll of battling cancer contribute to heightened productivity loss and increased rates of bankruptcy [15]. Individuals burdened by OOP expenses and productivity loss may find it challenging to access essential healthcare services, potentially leading to unmet needs and worsening health outcomes. The identification of relevant costs impacting cancer patients is thus crucial for understanding and addressing the financial challenges they face. This knowledge can assist healthcare organizations in tailoring financial assistance programmes and support services to meet the unique needs of patients.

Malaysia forms an interesting case study of cancer OOP expenditures as its widely acknowledged universal healthcare is provided through a combination of the public-funded healthcare system and private facilities. Public healthcare services are particularly heavily subsidised for all Malaysian citizens, with only nominal fees applied to specific services. Despite the significant subsidies and minimal charges at government hospitals in contrast to private facilities, research indicates that the diagnosis of cancer and subsequent care potentially impose a substantial financial burden on the bottom 40% income group of households (B40) in Malaysia [16]. Despite the evident need for targeted support mechanisms, research on the financial burden of cancer remains in its infancy. This paucity of information poses a significant impediment to effective planning and execution of health interventions. Therefore, this study endeavors to fill this critical gap by meticulously estimating the costs incurred by cancer patients in the lower-income group. Through a comprehensive exploration of OOP expenses and productivity loss, this study also hopes to unravel the coping strategies adopted by these patients. The present study seeks to inform policymakers, healthcare providers, and researchers and pave the way for targeted interventions for mitigating the financial impact of cancer among lower-income individuals.

Methods

Study participants and sampling

The study included cancer patients from low-income households attending the study sites, diagnosed with any type of cancer, and actively undergoing treatment. Inclusion criteria comprised low-income households falling within the bottom 40% (B40) of the Malaysian population, earning below MYR 5,250 [17], and recipients of the Sumbangan Tunai Rahmah programme (STR). The STR is a government financial aid programme aimed at assisting the plights of vulnerable groups. Eligible participants were those currently on treatment, aged 40 years and above, and capable of responding to the interview. The criterion of being aged 40 years and above was determined based on the rising incidence rates observed in all cancers combined [18]. Additionally, in cases of less verbal patients, their carers were also interviewed as long as they are able to provide detailed information of the cancer-related expenses. On the other hand, exclusions encompassed patient’s incapable of responding and with no companion and those solely on follow-up.

All eligible patients attending the oncology centre during the designated period were identified and recruited by convenience sampling. This non-probability sampling method was chosen because of the practical limitation of reaching a specific population of low-income cancer patients. Although convenience sampling does not provide the same level of generalizability as probability sampling, it was considered adequate for the research nature of this study.

Data collection method

The study employed a face-to-face interview using a structured and validated questionnaire, as detailed in the published pilot and feasibility study [16]. These questionnaires (see S1 Appendix) encompassed general sociodemographic and clinical profiles, cancer OOP expenses, costs associated with productivity loss (encompassing both paid work and unpaid household activities such as caregiving and housework), and details regarding financial coping mechanisms. Data were gathered rei during the interview, with patients were asked to recall their past expenses related to cancer care. Only expenses incurred within public healthcare facilities were included, while costs associated with private healthcare facilities were excluded. The operational definition of various costs was summarised in the Table 1.

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Glossary of the operational definitions of various costs variables.

https://doi.org/10.1371/journal.pone.0311815.t001

OOP and productivity loss costs estimation

OOP expenses included the reported medical and non-medical costs, factoring in any reimbursements that were deducted. Medical costs encompass self-reported OOP expenses made for diagnosis and treatment. On the other hand, non-medical costs include expenses related to transportation, lodging, meal expenditure, childcare, nutritional supplements and alternative treatments incurred by the patients. For patients utilising personal transportation, the cost was computed by considering fuel consumption relative to the travel distance from their residence to the treatment site (MYR 0.50/km) [19]. In cases of alternative transportation modes, the calculation involved multiplying the reported transport fee by the number of trips. Both medical and non-medical cost components were collected over the three months preceding the interview and then annualized to determine the total OOP cost per year.

Productivity losses due to absenteeism, resulted from individuals missing work due to cancer, affecting both short-term and long-term work absences. The cost associated with productivity loss was estimated by utilising self-reported wages from patients who were currently employed or had any other source of income. The calculation involved multiplying the number of missed workdays by the daily wage, assuming each absent day equated to eight working hours. Additionally, health issues arising from cancer impede individuals from performing routine activities with genuine economic value, such as household work, childcare, or volunteer work [20]. The loss in ’production’ was either forfeited or taken on by others, who had to allocate scarce time that would have been otherwise spent on different activities. Some estimates of lost productivity incorporated unpaid productivity. To address this, the productivity loss for unemployed patients was determined using the National Minimum Wage Malaysia 2022, set at MYR1500.00. Furthermore, the productivity loss for elderly cancer patients aged 60 and above was calculated based on their estimated productive hours, which were reported as 6 hours [21]. A conceptual framework of total cancer costs, as depicted in Fig 1, underscores the financial burden faced by patients, encompassing medical, non-medical, and productivity loss.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Overview of total cancer costs: The sum of annual OOP expenditure and productivity loss.

https://doi.org/10.1371/journal.pone.0311815.g001

Results

Sociodemographic and clinical profiles

The sociodemographic details and clinical profiles of the 430 study participants are summarized in Tables 2 and 3. The majority (38.6%) of patients undergoing treatment fell within the age group of 60 to 69 years. A predominant proportion were female (63.5%), married (79.3%) and possessed a secondary level of education (57.7%). A significant number of patients were not currently in the workforce, consisting of those retired (n = 48, 11.2%) and unemployed (n = 336, 78.1%) individuals. Among the unemployed, approximately 30% had to quit their jobs due to cancer. The mean household size was 4.14 (2.04) members, with a mean monthly household income of MYR 3,211.39 (MYR 2,576.46).

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Sociodemographic profiles of study participants.

https://doi.org/10.1371/journal.pone.0311815.t002

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Clinical characteristics of study participants.

https://doi.org/10.1371/journal.pone.0311815.t003

In terms of disease-related characteristics, approximately 31.4% of patients were diagnosed with breast cancer and 24.7% were gastrointestinal cancer. A majority presented with later-stage cancer, with 47.0% classified as Stage 4 and 32.3% as Stage 3. A significant portion of participants (78.1%) were undergoing chemotherapy treatment during the interview. Notably, about 39 study participants sought treatment at private facilities, although the costs associated with such facilities were excluded from the calculation. The majority of the participants (52.8%) had been living with a cancer diagnosis for less than a year. On average, the hospitalisation frequency was 1.68 (1.97) visits per year, with an average length of hospital stay of 10.29 (12.30) days, and an outpatient frequency of 30.11 (25.38) visits per year.

Out-of-pocket costs and productivity loss.

Table 4 presents a detailed breakdown of the cost components and the overall costs of cancer. The median (IQR) and mean (SD) of medical costs were MYR 777.30 (MYR 1,648.25) and MYR 1,624.72 (MYR 2,593.63), while the non-medical costs were MYR 2,711.70 (MYR 3,398.60) and MYR 3,797.28 (MYR 3,788.04), respectively. The primary contributors to the average total OOP costs of MYR 5,422.00 (MYR 4,731.50) were transportation at MYR 1,707.94 (MYR 1,933.74), nutritional supplements at MYR 1,582.12 (MYR 2,862.41), outpatient treatment at MYR 761.39 (MYR 1,746.10), and medical supplies at MYR 601.09 (MYR 1,451.95).

Download:

• PNG
  larger image
• TIFF
  original image

Table 4. Out-of-pocket spending and productivity loss costs due to cancer.

https://doi.org/10.1371/journal.pone.0311815.t004

On the other hand, the median and mean productivity loss costs due to absenteeism were MYR 6,923.08 (MYR 21,923.08) and MYR 14,981.54 (MYR 16,653.84), respectively. Additionally, productivity loss due to limitations in productive activities incurred a median and mean costs of MYR 1,846.08 (MYR 6,489.00) and MYR 4,593.75 (MYR 5,938.17), respectively. In totality, these costs contribute to an average total costs per patient per year of approximately MYR 11,102.85 (MYR 10,040.19).

Highlighting the significance of these costs, productivity loss emerges as the major driver, constituting 51.2% of cancer patients’ total costs. Non-medical and medical costs contribute around 34.2% and 14.6% to the total costs, respectively.

While overall OOP costs for cancer patients are higher in Peninsular Malaysia than in Sabah and Sarawak, this difference is not statistically significant (S2 Appendix). However, a closer examination reveals that transportation costs, a significant component of total OOP expenses, differ markedly between these regions, with a significant p-value of 0.002. Table 5 provides a breakdown of the annual transportation costs aggregated by Peninsular Malaysia and Sabah and Sarawak. A significant difference (p<0.05) in the average round-trip travel distances was observed, with those in Peninsular Malaysia and Sabah and Sarawak covering 91.8 km and 261.5 km, respectively. Residents in the Peninsular spent an average of MYR 1,465.94 (MYR 1,429.20), while those in Sabah and Sarawak incurred a higher average expense of MYR 2,254.28 (MYR 2,502.05). Analysing the cost breakdown, participants from both Peninsular and Sabah and Sarawak allocated the highest expenditure to petrol, followed by public transport. Toll fees were covered exclusively by Peninsular residents, and parking fees were the lowest among the listed expenses.

Download:

• PNG
  larger image
• TIFF
  original image

Table 5. Annual transportation costs (MYR) breakdown.

https://doi.org/10.1371/journal.pone.0311815.t005

Factors associated with OOP expenditure

The final model for independent predictors for OOP costs due to cancer is presented in Table 6. Results of the multivariate regression analysis indicated that factors such as age, ethnicity, employment, monthly household income, and annual outpatient visits may influence OOP costs. Specifically, patients in the 50–59 age are experiencing less burden of paying OOP expenses (exp β = 0.74, 95% CI: 0.56–0.98; p = 0.03). Chinese ethnicity (exp β = 1.44, 95% CI: 1.15–1.79; p<0.001) showed higher OOP expenses during the cancer treatment compared to Malay ethnicity. Regarding employment status, retiree patients had lower OOP costs compared to those employed (exp β = 0.54; 95% CI: 0.37–0.79; p = 0.002). Additionally, patients in households with income MYR 2000–5000 and >5000 were associated with higher burdens of OOP costs (exp β = 1.31; 95% CI: 1.07–1.60; p = 0.01) and (exp β = 1.40; 95% CI: 1.10–1.78; p = 0.01), respectively. Furthermore, patients with higher annual outpatient visits (exp β = 1.00; 95% CI: 1.00–1.01; p = 0.02) were also linked to increased OOP expenses. The model demonstrated an adequate fit to the data, as indicated by the deviance of 0.91, Pearson Chi-Square statistic of 1.09, suggesting a reasonable explanation of the variance in OOP costs among the studied population. Additionally, the likelihood ratio test showed a significant improvement over the null model (p < 0.001), supporting the inclusion of the selected predictors.

Download:

• PNG
  larger image
• TIFF
  original image

Table 6. Factors affecting OOP costs on cancer patients using multivariate regression analysis.

https://doi.org/10.1371/journal.pone.0311815.t006

Financial coping mechanisms

The study participants adopted various coping strategies to address their cancer OOP expenses, as illustrated in Fig 2. A predominant majority (90.7%) relied on their own, besides household salary and savings, indicating a common approach to managing the financial burdens associated with cancer. About 34% of participants sought financial assistance from relatives and friends to alleviate the imposed financial strain. Furthermore, support from non-governmental organizations and social welfare centres played a role in covering expenses for 27.9% of participants. Additional funding was acquired through borrowing without interest (15.1%), the sale of assets (10.7%), utilisation of health insurance (3.0%), and the least common strategy of borrowing with interest (0.5%).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Financial coping strategies used by cancer patients to finance their cancer OOP expenses, % (n).

Frequencies and percentage would not be added up because multiple responses were possible. ^a It includes salary and/ or savings (i.e., Employee’s Provident Fund, EPF) of personal and household; ^b Financial support received from relatives and friends which is non-refundable; ^c Any means of selling household assets (i.e., jewellery, property, land and other household items); ^d Borrowings or taking loans from individuals (non-household members or friends); ^e Borrowings or taking loans from financial institutions; ^f Other financial sources received from non-governmental organizations (i.e., MAKNA Cancer Support) and social welfare centres (i.e., Zakat).

https://doi.org/10.1371/journal.pone.0311815.g002

Discussion

The study demonstrated that cancer places considerable financial strain on patients, even in a highly subsidized healthcare system such as Malaysia. Despite patients paying only a minimal fee at public healthcare facilities, their financial burden remained high. This was predominantly attributed to productivity loss and non-medical expenses. A shift in major cost contributors from medical to non-medical as often observed in other countries occurred because the subsidy extended only to medical treatment and hospital charges. This leaves the relative proportion of other additional expenses covered by the affected patients and households to be larger. Furthermore, the data also showed that even with subsidies, specific outpatient charges such as for radiotherapies and out-of-formulary medication purchases, in addition to medical aids can still add on to form a significant OOP medical expenditure. This was in line with several studies where such expenses can compound and in certain circumstances, become lifelong financial burdens for the patients [16,30–32].

Interestingly, travel and nutritional supplement costs surpass all other OOP cost categories. Transportation costs, frequently acknowledged as burdensome, appeared to hold significant importance, especially among individuals from lower socioeconomic statuses [16,30,32,33]. The combination of frequent visits and extended travel distances to cancer referral hospitals in this study contributed to the overall escalation of OOP costs. The high unit cost of transportation further magnifies the financial strain, particularly for cancer patients residing in distant locations. This is particularly true for the vast regions of Sabah and Sarawak, each with only one public cancer referral centre. The geographical limitations necessitate longer travel distances and higher expenses, including additional costs such as vehicle rentals, airplane tickets and boat fares, to reach cancer centres in urban areas. This limited access to cancer centres in these regions exacerbates the overall escalation of OOP costs. Additionally, expenditures were heightened by the intake of nutritional supplements, such as fortified milk and health supplements [16,30,34]. While the data was not able to distinguish whether such practices were prescribed or self-initiated, it is worth noting that such behaviours were also common among the lower-income groups.

The multivariate regression analysis identified as age, ethnicity, employment status, monthly household income, and the annual number of outpatient visits as significant predictors of OOP expenses among cancer patients. Contrary to some expectations, the analysis revealed that patients aged 50–59 experience a reduced burden of OOP expenses. This finding contrasts with a recent population-based study by Schneider et al. [35] among breast cancer patients in Germany, which highlighted increased OOP expenses among older patients. The difference in findings could be due to variations in healthcare utilization across age groups in different healthcare systems. Interestingly, while the mean costs calculated in this study indicate that patients aged 60 and above have higher spending, the GLM analysis suggests that the burden of OOP expenses is less for those aged 50–59. This discrepancy may be explained by differences in healthcare needs and financial capacities, where older individuals might face higher absolute costs due to more frequent healthcare utilization, but this does not necessarily translate to a higher relative burden after adjusting for other factors [36].

On the other hand, Chinese ethnicity was shown to also be a predictor for higher OOP expenditure. They specifically had a larger medical care expenditure compared to other ethnicities in public healthcare. Further analysis to explore possible reasons for such a trend showed that ethnicity was not associated with cancer types, severity, employment categories, and expenditures on nutritional supplements. The possible reason for this might align with a prior local study by Wan Puteh et al. [37] who postulated that the higher OOP burden among the minority groups may be associated with employment in the public sector. This is because in Malaysia charges are waived for current and retired civil servants. The current national statistics demonstrated that while ethnic minorities account for 42% of the population, only 22% are public service employees [38]. Thus, the additional fees waivered may have led to differences in the medical cost in the public sector.

Contrary to prior research which showed cancer patients residing in lower-income households were prone to higher OOP expenses [37,39], our data among the lower-income B40 households demonstrated that the poorest group with an income less than RM 2000 had a reduced likelihood of bearing a higher OOP burden. This discrepancy might be attributed to their scarce financial resources, which potentially limits their OOP expenditures, especially for non-medical expenses. While other studies have suggested that a longer time since diagnosis as well as higher hospitalisation frequency and longer length of hospital stay tend to influence OOP expenses [35,40], this study surprisingly reveals the opposite effect. Individuals with a higher number of outpatient visits, as supported by previous studies surprisingly bear a greater OOP burden [37,41].

The oversight of productivity loss estimation is evident in numerous cancer cost studies, despite its significant contribution to the overall costs of cancer care [6,42]. In reality, the costs attributed to productivity loss due to cancer could potentially constitute up to half of the overall expenditures [43]. This was echoed by the findings the average total productivity loss due to absenteeism and limitations in performing productive activities costs surpassed the OOP costs. The incapacity of cancer patients to work or the necessity for extended leaves for example not only affects their financial situations but also strains the overall income of the household. Moreover, the emotional toll and heightened caregiving responsibilities may further diminish the productivity of other household members, thereby creating a compounding effect on the economic well-being of the entire family [44,45]. It is worth highlighting that the losses are expected to be even higher if productivity loss from presenteeism and possible premature death were included in the long run.

Cancer patients may grapple with financial challenges even within a publicly funded healthcare system [46]. While patients can tap into their financial reserves, such as savings and health insurance, the persistent nature of cancer can swiftly deplete these resources. This study reveals a notable proportion of cancer patients experiencing unemployment, leaving them with little recourse other than relying on support from family members. However, in households lacking substantial initial savings, a cancer diagnosis can prove particularly devastating, resulting in financial sacrifices and household impoverishment that affect all members [47,48]. This study underscores that families with limited financial resources, impacted by cancer, often resort to borrowing and selling assets to meet associated costs. This finding aligns with previous studies that highlighted selling assets and borrowings as major financial coping strategies [49–52]. Moreover, a study by Azzani et al. [4] reported that households of colorectal cancer patients resort to selling household items or borrowing money from relatives and friends to navigate their new financial circumstances. Despite the presence of subsidies in public healthcare facilities, patients still find themselves reaching into their own pockets to cover non-subsidized treatments, medications, and hidden non-medical costs.

It was also observed that patients and their family members allocated nearly 34% of their total monthly income to finance cancer OOP expenses. This percentage greatly contrasts with the Malaysian household expenditure on healthcare, which usually ranges from 2.6% to 3.5% of the total monthly income [53]. Essentially, the costs associated with cancer impose a noticeable economic strain on those affected. A study in Canada reported that among cancer patients who perceive OOP as significant spent, on average, 34% of their total monthly income [54]. Similarly, in Southeastern United States, breast cancer survivors allocated as much as 31% of their monthly income to these expenses [39]. Given that Malaysia’s healthcare expenditure is currently below the WHO-recommended threshold (< 5%–6% of the annual GDP), allocating additional funds for healthcare, subsidising the costs of medical aid, and providing financial assistance programme could potentially alleviate the financial burden and improve healthcare access among lower-income cancer patients [55,56].

This study is subject to several limitations. Firstly, the estimation of expenses is confined to public referral cancer centres under the MOH. Thus, possible expenses incurred at private hospitals were not included. Eventhough evidence showed that lower-income patients are less likely to pursue care in private facilities [57], our sample showed that patients may still seek selected services such as surgery and investigations in private facilities. Thus, it may lead to an underestimation of the total financial impact on patients. Future research should consider including both public and private sector costs to provide a more comprehensive assessment. Secondly, the reported OOP spending and productivity loss are also limited by the recall period. This approach could result in cost exaggeration or underestimation, depending on the type of service and the length of the recall period, leading to potential inconsistencies in the data. However, a pilot study was conducted prior to the main study to evaluate these recall periods, and it was found that patients were generally able to recall and provide the required information accurately, supporting the feasibility of this approach despite the limitations. Furthermore, the study did not include the loss of productivity of caregivers due to the unavailability of companions for most interviews. This gap in knowledge, in addition to inaccurate recall of absenteeism, tends to grossly underestimate productivity losses. Lastly, the annualization of costs using one and three months’ expenses may potentially have influenced the accuracy of the total costs of cancer calculated in this study. This is because cancer expenditures are dependent on time, with expenses peaking in the first six months of diagnosis and year.

Albeit limitations raised, this study supplements the expanding body of literature addressing the financial burden of cancer from the patient’s perspective. Acknowledging the significant financial strain accompanying a cancer diagnosis is crucial for developing a patient-centred plan that addresses the specific needs of both patients and their families. The findings from this study play a pivotal role in advancing initiatives aimed at addressing the financial hardships experienced by low-income cancer patients [58], providing an estimate of the OOP expenses related to cancer care. Furthermore, the study highlights the ongoing role of household members as primary contributors to funding cancer-related care, even within publicly subsidised healthcare systems. Based on these findings, the PeKa B40 scheme, which offers financial assistance for cancer treatment and transportation appears to be timely and well-needed. This initiative commendably alleviates the financial burden on cancer patients, especially for expenses such as transportation and medical equipment [58]. However, it is worth exploring whether these initiatives reach the lower-income groups and if they alleviate their financial burdens. This study is thus poised to facilitate discussions around planning and decision-making on the redistribution of allocations based on needs and risks.

Conclusions

In summary, this research emphasizes the significant impact of cancer OOP expenditures on the financial well-being and productivity of lower-income households. These families grapple with the dual challenges of reduced income and increased expenses. Despite the presence of heavily subsidized public healthcare as the foundation of universal health coverage in Malaysia, it may remain inadequate to shield cancer patients and their families from catastrophic expenditures. Thus, there is still a need for policymakers to incorporate a social safety net into the healthcare financing system, with a focus on prioritizing the disadvantaged population. Such integration should ensure access to health services without subjecting individuals to excessive financial strain while balancing the long-term sustainability of public healthcare. The implementation of targeted measures such as PeKa B40 cancer incentives [59] may greatly benefit these population subgroups. Furthermore, the inclusion of health insurance coverage and financial support schemes also has the potential to alleviate the financial difficulties faced by affected patients. Ultimately, it is hoped that the findings will help guide policy-makers in developing holistic patient-centred cancer care while planning for future resource reallocations and investments.

Supporting information

Acknowledgments

We thank the Director-General of Health, Malaysia for making this publication possible. We also thank all collaborators from the Radiotherapy and Oncology Department, Kuala Lumpur General Hospital, Penang General Hospital, Cancer National Institute, Sultan Ismail Hospital, Sabah Women and Children Hospital and Sarawak General Hospital who enabled the data collection to happen. We would like to thank the reviewers for their thoughtful comments and suggestions.