Introduction

The costs of cancer treatments have been steadily increasing and even in environments of publicly funded healthcare some of that burden is falling on patients and their families. Additionally, patients are often not able to maintain their full employment, resulting in decisions to reduce or stop working as a consequence of the cancer, its treatment, or associated side effects which limit the ability to work. Caregivers may also experience limited capability to work, further adding to the financial burden on the family.

In recent years, the term financial toxicity has been used to describe the “distress and hardship arising from the financial burden of cancer treatment” [1].

The issue of financial toxicity has become increasingly relevant for many developed countries. This is due in part to a number of changes in healthcare system factors including the increased costs of newer cancer therapies [2], the higher rate of healthcare-related debt for patients and their families in certain countries [3], and the impact on ability to work for both patients [4] and their caregivers [5]. As a consequence, patients and their families are experiencing high rates of financial burden from cancer diagnosis and its treatment across both public and private funding settings.

There has been a significant body of literature surrounding this topic, including systematic reviews in the USA [6,7,8] and Australia [9]. However, none of these reviews focused on publicly funded healthcare systems in very high development index countries. It might seem reasonable to assume that financial burdens in these publicly funded environments are much less than those that are experienced in countries with a strong private sector component, as these public schemes are typically designed to reduce the patients’ financial burden associated with accessing healthcare. However, to date, this has not been investigated. In an effort to address this gap in knowledge, we examined the top 10 cancers globally (Globocan 2018: Global Cancer Observatory (http://gco.iarc.fr)) in very high development index countries to better understand these burdens; previous reviews have focused on all countries or predominately privately funded systems. The intent was to investigate the magnitude and extent of financial toxicity between these countries, and if some countries are managing patients’ financial toxicity better than others.

Methods

The review was conducted and reported following the PRISMA systematic review in healthcare guidance [10].

Categories of financial burden

There are a number of terms related to this financial toxicity that have appeared in the literature that should be defined in order to be clear about what we are trying to measure, and what we are not trying to measure. Generally speaking, there are several categories of financial toxicity/burden. We considered out-of-pocket costs (OOPC), income loss for patients and caregivers, travel and accommodation costs, and patient perceived stress/strain. For the purposes of this review, we are not focused on psychological adaption, insurance-based issues (except how it pertains to sub-populations of the other categories), or qualitative examinations. We sought to systematically identify studies which reported on the four categories of financial toxicity/burden, as stated above, from the perspective of patients and/or informal caregivers.

Generally, OOPC refers to patient expenditures related to their cancer treatment that are not reimbursed by government or insurers. This would typically include direct costs such as those for drugs, devices, homecare, complementary and alternative medicine, health professionals, hospital fees, and other related costs, and studies were eligible if they reported on any of these costs individually or OOPC overall. We note that one of the earliest references to this type of cost characterization was in relation to breast cancer patients’ burden [11].

Income loss can result from reductions in hours or days of work, leave of absence, or early retirement, and can occur for the patient or for caregivers in order to provide care and support for the patient. Studies were eligible if they reported income loss in monetary terms, in time, or as a percentage of the sample with lost work.

Travel and accommodation costs can be included in OOPC or can be broken out or reported separately and include fares by taxi, train, plane, fuel costs when travelling by car, parking costs, or government reimbursement calculations based on specific reimbursement per kilometer per mile (imputed). We treated this separately from OOPC because some of these costs are imputed values rather than directly incurred OOP expenses. Studies were eligible if they reported travel costs separately or within OOPC.

Patient perceived financial stress, strain, and distress have a variety of definitions in the literature. Financial stress has been described as “… the impact of the cancer diagnosis on the ability of the household to make ends meet” and is an “objective” measure related to the costs incurred [12]. Financial strain in the cancer context has been defined as “…the impact on the individual” (or the household) and “…how (they) felt about the financial situation they were in” and is considered a more “subjective” measure of financial impact [12]. Financial distress is closely linked to financial toxicity as it represents the psychological response that patients have to the financial toxicity [8]. Studies reporting the frequency of any of these concepts were eligible for inclusion in the review. Any of these categories of burden could be reported as weekly monthly, yearly, or multi-year.

Inclusion/exclusion criteria and outcomes

The overall review protocol was developed using the PROSPERO template (see Appendix B for details). In the screening process, studies were eligible if they included one or more than one of the 10 most common tumour types (namely lung, colorectal, breast, prostate, bladder, cervix uteri, stomach, esophagus, thyroid, and liver), and were conducted in a very high development index countries (Sources: World Health Organization: International Agency for Research on Cancer, cancer incidence list by tumour type for the world (http://gco.iarc.fr/today/explore) (Multibar option); (HDI country list 2018 http://hdr.undp.org/en/content/human-development-index-hdi)) with publicly funded universal healthcare. (Studies in the USA were therefore ineligible). We recognize that the term publicly funded healthcare can take a variety of forms including a National Health Service model, part of a social security system, or a mix of public and private systems; these different models have been examined in some detail by others [13]. We have not examined each of these types separately, but rather included countries where the dominant system is a public one regardless of model type.

We focused on full papers reporting quantitative studies which included those where the subjects were older than 18 years of age at the time of cancer diagnosis (no pediatric studies) because we wanted to focus on perspectives of working adult patients and/or their informal carers. Eligible studies could have a cross-sectional or prospective observational design. Those which assessed toxicity beyond 5 years post-diagnosis were excluded because the primary focus was on the impact of active treatment and its initial follow-up. Clinical trials were excluded to place the focus on the financial burden experienced by patients being managed in routine clinical care. Additionally, studies with less than 100 subjects were excluded because they are less likely to be generalizable. Only studies that reported on at least one of the following: out-of-pocket costs (OOPC); income loss for patients and carers; costs associated with travel to healthcare facilities; and patient perceived financial stress, strain, or distress were included. We excluded studies focused on qualitative examinations of financial hardship.

The main outcomes of interest were OOPC associated with co-payments or cash payments for medically related costs, and non-medically related support (e.g. family care), costs related to accommodations and travel, lost time from work for both patients and their carers, and financial stress/strain including decisions to forego care, loss of savings, and impact on future earning potential post-treatment. These could be weekly, monthly, yearly, or multi-year evaluations.

Searches

The search strategy was conceived and developed collaboratively between two authors (CJL and LB). Using a combination of keywords and database-specific controlled vocabulary, the search strategy incorporated the following concepts: cancer as defined by the top 10 reported adult cancers; financial toxicity including terminology related to financial burden; and patient’ stress or burden on the individual and/or the caregiver. Eligible studies were peer-reviewed, published in English. They were first screened by title and then by abstract, and finally by full article, with the last two steps undertaken by two researchers. Individual searches were carried out in the following databases: Ovid Medline, Ovid Embase, Ovid PsycINFO, CINAHL, Business Source Complete, and EconLit. All searches encompass the publication period of January 1, 2005, through March 7, 2019. For an extract of the OVID Medline search, please see Appendix A. Our decision to use this timeframe was because other reviews have covered the period prior to 2005, and it becomes increasingly difficult to compare monetary costs over longer periods of time.

Screening

One reviewer (split across a team of 5) independently screened the titles, followed by two teams of two independent reviewers (CJL and MF; LS and PH) screening the abstracts and then full texts of papers identified as being potentially eligible for inclusion. Disagreements at each stage of screening were resolved through mechanisms within the Rayyan software, which allows blinded review and highlights where discrepancies in decisions occur (inclusion or exclusion). This facilitated discussions [14] among reviewers; if necessary, a third reviewer (LB or CJL) was consulted in the event that any remaining conflicts were unresolved.

Data abstraction

One reviewer (CJL) extracted data from eligible articles into a standardized data abstraction form that best fit our intended examination of included studies [15]. The team resolved any conflicts through discussion and, if necessary, a second reviewer (LB) resolved any discrepancies that persisted. Data abstracted included study characteristics, cancer patient characteristics, measures of OOPC, travel/accommodation costs, lost income, and stress/strain and toxicity. Details on country, year of study, study design, and comparison group (where applicable) were also abstracted. We finally categorized the outcomes into one (or more) of the four categories based on “quantitative” evidence in the paper.

All study outcomes for OOPC and travel/accommodation costs were presented in local currency and were inflated to 2018 values using the OECD Inflation chart (CPI) (https://data.oecd.org/price/inflation-cpi.htm#indicator-chart) based on enrollment dates for study patients. As a second step, in order to convert expenditures into a common currency and time frame to facilitate comparisons, we used the OECD purchasing power parity calculator (https://data.oecd.org/conversion/purchasing-power-parities-ppp.htm) to adjust all currency values to $US/month for comparison purposes. Those studies that reported different outcomes (% change, odds ratios, etc.) were reported separately but not as a comparison between countries. Where studies reported factors associated with variation in financial burden, this information was also abstracted.

Quality appraisal

To assess the quality of eligible studies, we applied the QualStat checklist [16]. The authors’ defined quality “in terms of the internal validity of the studies, or the extent to which the design, conduct and analyses minimized errors and biases” [16, pg. 2]. The checklist included 14 items with a range of elements deigned central to internal study validity. Scoring ranged from “yes” = 2, “partial” = 1 to “no” = 0 (with non-applicable items denoted “n/a”). A summary score for each paper was calculated by summing each criterion score across the 14 items and dividing by the total possible score. We selected a minimum threshold for inclusion of 60% given the broad range of financial topics under review and the consequent range in methodological quality.

Results

We identified 7303 citations between Jan 1, 2005, and March 7, 2019 (Embase 3620, Medline 2676, CIHAHL 536, Business Source Premier 269, PsychInfo 119, EconLit 83). After removing duplicates, we identified 7117 unique articles to include in our title review. Following title review, we eliminated all but 751 articles. Abstract review then identified 41 qualifying articles (39 studies). Full paper review eliminated an additional 9 articles. Hence, our final evaluation includes 32 articles [17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48], representing 30 studies, as shown in the PRISMA flow diagram (Fig. 1).

Fig. 1
figure 1

PRISMA diagram

Full size image

These studies came from Canada (7), Australia (6), Ireland (6), UK (2), Denmark (2), Germany (2), Finland (1), South Korea (1), France (1), Malaysia (1), and the Netherlands (1). Eighteen studies included breast cancer, 16 studies included colorectal cancer, 13 studies included prostate cancer, 7 studies included lung cancer, 4 studies included gastro-intestinal cancers, 2 studies included cervical cancer, and 1 included thyroid; the other cancer types were presented in two papers that included all cancers but did not break them out by tumour.

The most frequent categories of burden reported were as follows: Out of pocket costs (OOPC) (Table 1) 18 studies; income loss (Table 2) 18 studies; travel costs (Table 3) 14 studies; and toxicity/stress and strain (Table 4) 6 studies. Twenty-five studies reported financial burden from the perspective of patients, one from the perspective of caregivers, and four from both perspectives. In Tables 1 and 3 below, we have presented the “key findings” in their native currencies as reported in the paper but converted to $US 2018 when possible and appropriate for OOPC and travel costs to facilitate comparisons. In Tables 2 and 4, where possible, we reported the percentage of lost wages and percentage of those experiencing stress, strain, or distress as appropriate, again to facilitate comparisons.

Table 1 Studies reporting OOPC (papers) n = 18 (n = 20)
Full size table
Table 2 Lost income studies (papers), n = 18 (n = 20)
Full size table
Table 3 Travel and accommodation studies (papers) n = 16 (n = 14)
Full size table
Table 4 Financial stress, strain studies, n = 6
Full size table

Out of pocket costs

Most OOPC studies reported actual costs and the duration of observation varied from weekly to 2 years (n = 18). Some studies included travel (n = 5) in their OOPC, while others provided these as two categories and reported both separately. Some studies examined a single cancer type, while others examined multiple cancers; not all of these latter studies reported OOPC for each tumour type.

In addressing those OOPC studies that could be converted to 2018$US/month, and where travel costs were included (n = 5), the highest was in Ireland ($503; [32]) and the lowest, including travel costs, was in Australia ($271; [23]), where both studies included a variety of cancers (Table 1). In studies where travel costs were not included, the highest costs were $418/month in breast cancer in Canada [17] and lowest was $17/month in prostate cancer in Canada [27]. One study [29] also reported OOPC by stage of disease with values of US$183/month (stage 1), $219/month (stage 2), $247/month (stage 3), and $243/month (stage 4).

The remaining two studies reported data differently. Paul et al. [31] included the percentage of patients (in a variety of cancers) who found particular cost types that influenced decisions (11% for treatment costs) and Azzani [29] presented OOPC currency results but by stage of disease in colorectal cancer ($183 stage I, $219 stage II, $247 stage III $243 stage IV), although this was still converted to 2018$US.

Income loss

Income loss was most commonly presented as percentage change in the number of patients with reduced income due to reduced hours and/or days, time away, left work force, or retired (n = 11 studies). A few studies (n = 7) reported actual days lost or dollars of income lost.

In terms of those studies reporting percentage change, the highest income loss was for Australians predominantly with breast and colorectal cancers, with 67.3% of those employed experiencing a change in employment status [31]. The lowest was in the UK (England, Wales, North Ireland, and Scotland) where only 18% of prostate cancer patients became unemployed or retired [44].

Of the studies which reported absolute losses, most (n = 4) focused on lost income of patients, and 2 studies included carers. An Australian study of 272 breast cancer patients reported a median loss of income AUS$5078 in the first 6 months and an additional AUS$1553 between 13 and 18 months [22], a Finnish study of 508 colorectal cancer patients and carers reported income loss ranging from €405 to €5078 depending on stage of treatment [41], a UK study reported breast/colorectal/prostate cancer caregiver’s mean income loss of £70 [28], and a Malaysian study reported income loss for colorectal cancer patients in local currency between RM296 and RM1151 [29]. One Canadian study reported lost days of work as 12.6 days for patients and 7.0 days for caregivers per month across breast, colorectal, lung, and prostate cancers [18,19,20].

Travel and accommodation costs

Some studies only included fares and/or cost of fuel (n = 3) while at the other extreme, they included imputed costs related to wear and tear on personal vehicles using government reimbursement rates for travel per kilometer per mile (n = 3).

Across studies, the average travel (and accommodation) costs were US$139. The highest travel costs were from a Canadian study reporting US$393 per month for breast cancer; this included imputed costs for private car wear and tear [24]. The lowest was in the UK in a mix of breast, colorectal, and prostate cancers with a cost of US$8/month but does not include personal vehicle wear and tear costs [28].

Toxicity, stress, distress, and strain findings

We found that for financial strain (subjective financial burden), the lowest prevalence was 7% at 12 months after diagnosis for Australian colorectal cancer patients [42] and the highest was 39% in Ireland for colorectal cancer patients [30]; in that study, the levels of financial stress (objective burden) were even higher. One Denmark study reported on financial toxicity (based on the financial question in the EORTC QLQ-C30), which was reported by 22% of working patients and 27% of non-working patients, but did not specifically address financial stress and strain [46]. Finally, in the Irish colorectal cancer study, 49% of patients reported depletion of savings; with low savings, borrowing money, and loans from family/friends all increasing the risk of both financial stress and strain [47]. We note that the time since diagnosis varies across these studies as outlined in Table 4.

Quality appraisal

The quality review revealed studies to be of mid to high quality with a relatively high average score across the included literature of 87.1%. The scores ranged between 63.6% (36.4% prior to inclusion criteria threshold) and 100%. In terms of time period, the 2006–2010 sample scored lower (85.8%) than more recent papers published between 2011 and 2019 (87.5%). Older papers revealed a minimum score of 77.3% compared with 63.6% for newer papers suggesting that while the average methodological quality of papers was improving over time, the variability in that quality also grew. Details on the quality scores by study can be seen in Table 5.

Table 5 Quality assessment of sample literature on cancer-related financial toxicity (2006–2019)
Full size table

While the sampled papers generally scored high across the literature, with 10 of the 11 criteria scored, on average, between 80.6 and 100%, a distinct weakness arose in relation to controlling for confounding in the sample. This criterion scored 48.4% with, on average, 32.2% of the literature in the sample either failing to control for confounding, or not reporting the attempt, generally in relation to their statistical models. The reporting of results criterion scored 2nd lowest at 80.6% indicating that deficiencies also arose in relation to cost presentation where full results were only reported for some outcomes making it difficult to assess the overall financial burden of the cancer under study. In some cases, this was likely due to a lack of comprehensive data collection. The remainder of the criteria was above 83.9% indicating a high level of quality across the majority of the papers.

Discussion

We identified a range of financial impacts related to OOPC, income loss, travel costs, and financial stress/strain/toxicity across eleven countries and ten cancer types in publicly funded healthcare systems. Although there is some evidence of financial toxicity across all jurisdictions, and evidence of significant amounts across some jurisdictions despite the public healthcare focus of the review, it is less clear that we have comparable patient populations. However, it is clear that all of these countries have some evidence and varying frequency of financial toxicity with certain patient and disease characteristics increasing risk. Although there is some evidence that recent increases in costs of care [5], and changes in level of public coverage are exacerbating this issue [5], more details on these phenomena are still needed. Additionally, we might expect that differences observed are strongly influenced by the healthcare system structure (e.g. the co-pay or deductibles required, and the degree of privatization) and by social welfare provisions (e.g. extent of income replacement during sickness absence), which vary significantly across countries. In fact, we anticipated differences across countries specifically because of these types of factors. The relevant point here, from our perspective, is that these health policy decisions have varying impacts on patients’ and their families’ financial burdens, but that all systems result in some degree of financial toxicity.

There are many ways to examine these findings, and in actuality, one of the biggest challenges is to decide how best to portray these financial burdens across the 30 studies identified. In one sense, the most efficient way to examine this is to look to the patient and determine their level of stress and strain or distress caused by the financial challenges (toxicity). Our identified studies present a range with a low strain of 7% 12 months after diagnosis for Australian colorectal cancer patients [42] and a high of 39% in Ireland for colorectal cancer patients [30] with levels of financial stress even higher. Although this is a broad range, even the 7% rate is of concern and certainly when numbers greater than a third of cancer patients experience financial strain, it suggests that the healthcare system is not able to fully support patients financially. It could be argued that this is not the role of governments, and with budget constraints, this is a fair argument. However, there should be a threshold for considering alternate strategies when a significant portion of the population is experiencing financial stress and strain related to their cancer treatment and follow-up. Some researchers have gone beyond these concepts to consider the psychological impact, the effect on overall well-being, and the need for better assessments of patients’ overall ability to cope with these financial impacts to support vulnerable patients more effectively [49,50,51].

Any comparisons across countries are limited by differences across studies by patient cancer types, stage at diagnosis, and whether patients are in active treatment [18, 19] or follow-up care [27]. As we were unable to control for this heterogeneity, it is less clear whether one country fares better than another, but suffice it to say that the majority of studies (14/16), when examining patients in active treatment, have OOPC that exceed US$100/month with many having costs that exceed US$300 (5 of 16).

We observed variation in lost income by country. Differences could be a consequence of better income replacement programs funded through social welfare systems by government using, for example, partial compensation for income loss as in the Netherlands [40] or partial wage continuance as in Germany [45]. These social system factors likely account for much of the differences observed as disease treatments are not likely to vary significantly, so impact on ability to work should be similar. In any case, it is clear that income effects are common for patients and, as has recently been reported in another systematic review (de Boer et al., in press), these negative impacts on income may be evident years after the cancer diagnosis [48]. Much less appears to be documented around caregiver lost income with only a few of the studies examining this in a more focused way [28, 41].

Few studies examined travel costs separately, although several embedded travel costs into the OOPC totals. Larger geographies of some countries appear to show greater travel expenses (Canada and Australia). This is to be expected, especially considering the number or rural settings that exist in both countries where healthcare services for cancer may be unavailable locally [23, 24, 35]. As might be expected in each of these studies, those travelling greater distances to centres experienced higher travel costs. In this regard, comparing across countries is a challenge when examining travel costs when geographies vary both within a country and between countries, again making straightforward comparisons almost impossible.

We attempted to look at the top 10 cancers based on worldwide incidence; however, in fact, we found that specific studies within the top 10 that included analyses of the actual tumour type were limited for many with breast (18), colorectal (16), prostate (13), lung (7), GI (4) (which included stomach and esophagus), and cervix (2) identifying multiple studies. However, just one study each for bladder and thyroid cancers was identified, and a few additional studies listing other cancers (3) or all cancers (2) with no details on individual tumours, likely due to smaller sample sizes. Detail across tumour type was mostly limited to the top 8 cancers.

The three most commonly studied cancers were breast, colorectal, and prostate and, in many of these cases, a number of observations were made that mostly aligned between countries. These commonalities included an increased financial burden (in most categories) for individuals with low income [18, 31, 37]; under retirement age (60–70 years) [18, 31, 35]; with more severe disease [29, 41, 44]; with shorter time since diagnosis [22, 27, 39]; without supplemental health insurance [18, 34]; and living further from cancer treatment centres [24, 32]. Each of these findings highlights that the disease and the time since diagnosis has a significant influence on the severity of the financial toxicity patients experience.

We should also compare these results with literature in a country; the USA has often been studied, where publicly funded healthcare is not the primary method of delivery. A recent US review of financial toxicity [6] suggests high rates of productivity loss although the difference is not that clear when comparing with publicly funded countries. Travel costs in the USA appear to be within the range observed in our publicly funded studies [6]. Differences in OOPC are evident; these are higher in the USA with monthly values ranging from $250–900/month when measured directly [6]. US data on financial stress (28–73%) and strain (16–32%) suggest that it is higher on average than that seen in publicly funded countries [8], although the effect for those under 65, hence not age-eligible for Medicare, is greater which mirrors that seen in the publicly funded countries. The rate of medical debt and/or bankruptcy was between 5 and 62% in the US studies [6] a number that although not well described in our identified studies are likely lower than that seen in the USA.

A 2015 systematic review has suggested that as the costs of healthcare increase, the burden on patients also increases both in cancer and in other diseases like cardiovascular disease and rheumatoid arthritis [52]. These costs are most likely to increase in the future, highlighting the urgency to address the financial gaps. Additionally, a recent ASCO abstract from late 2019 suggests that in specific cases, such as the use of tyrosine kinase inhibitors (TKIs) therapy for non-small cell lung cancer, the impact on patients’ finances in the USA has also had a negative impact on patient outcomes [53].

This review identified a number of the gaps in the existing literature including a shortage of data relating to the 3–5-year period post-diagnosis; literature suggests that lost wages can persist up to 5 years in some patients [48] but little has been published on other aspects of the financial burden. A further limitation relates to the lack of research on the work impact for caregivers; only 5 studies captured any detail on this population. We also noted that there are a variety of ways to capture lost income or OOPC, suggesting that standardization of measures would be valuable to allow easier comparison across jurisdictions.

Lastly, how do we put these outcomes into the current context as cancer care continues to evolve? Will the increased price tags on cancer drugs result in higher OOPC for patients? Will the use of more oral agents result in a reduction in travel costs, but increase the rates of emergency department visits and admissions or other services? When new treatments are less toxic, will it allow patients to work longer hours or more frequently? Will the increase in care costs have an impact on insurance co-payments or strategies to minimize government and private payor burdens? What role might debt and bankruptcy have on perceived financial distress, and what influence might public systems play here compared with private systems? We are hopeful that this investigation therefore encourages others to address these unanswered questions and provide a more fulsome explanation of the current burden faced by patients in publicly funded healthcare systems.

Limitations, of this review

We recognize that “publicly funded healthcare systems” are not all the same, as differences between countries may be influenced by their structure, culture, and political differences, among other factors [13]. We do not attempt to tease out all these differences, but acknowledge their examination and the role they play could be examined in future research.

Although we attempted to standardize outputs, some studies were presented in a way that did not allow them to be compared. In these cases, we summarized the results but not for comparison purposes. Comparisons are a challenge due to a variety of factors, including year of study, time since diagnosis, mix of cancer types, severity of illness, average age, and included costs within categories. Despite these limitations, it is clear that partly due to differences in support systems for patients between countries, the financial burdens appear to differ. This difference is particularly stark when looking at income losses for patients and their informal carers.

Our attempt to standardize to $US2018/month for OOPC and travel costs is also less than ideal as CPI adjustments are for all goods and services within a country and in many cases, healthcare cost changes are slightly different from other goods. We used purchase price parity from OECD tables but again, this may not be the same for healthcare as the PPP values are for the net effect of all goods and services and may be slightly different for healthcare. However, we believe this gives us a better sense than leaving each result in its native currency and year making it much more difficult to compare.

Lastly, we used the “whole world” (WW) cancer top ten, rather than the “high development index” (HDI) list which differs with liver and esophageal (WW) replacing melanoma and uterine (HDI). We also note that the use of HDI itself has some limitations as “high development” does not necessarily mean a strong publicly funded healthcare system nor does it ensure a similar level of care when compared with other jurisdictions.

Limitations, in the existing research

It is clear from this review that an agreed upon standard for measuring each of these outcomes has not been determined; hence, part of our challenge is making valid comparisons. Although we did our best to compare findings across studies, we recognize that this comparison is less than optimal, and comparisons made here need to be interpreted cautiously. As an illustration, not all lost income was reported similarly, as those who experienced reduced hours of work or lost productivity were not always captured, and these types of losses may be higher in more severe disease and younger populations. Additionally, some of these studies summarized a variety of cancer types, different stages of disease, and different time duration since diagnosis and it is known that each of these factors has an impact on the treatments and services required and in the ability of patients to return to work. Since virtually none of the studies matches exactly for all of these variables, comparisons are difficult and unreliable and again should be interpreted with caution. Finally, financial stress and strain studies used different time frames from diagnosis which is likely to have an impact on the frequency of these states; it is also unclear whether these studies used consistent definitions for these concepts.

Conclusions

Although government funded public healthcare exists in many very high development index countries, financial toxicity is still common among cancer patients and caregivers. The evidence suggests that those with a shorter time since diagnosis [22, 27, 39], not currently working [46, 47], and with more severe cancers [29, 41, 44] have higher rates of financial toxicity, including stress and strain. The studies also reveal that the rate of financial toxicity varies between these countries and yet still translates into high rates of financial stress and strain in the countries studied in this review. We believe this suggests that the current financial protections in many countries with publicly funded healthcare are still inadequate and room for improvement still exists. In comparison with the USA, OOPC burden in countries with public healthcare is smaller, as are the rates of financial stress and strain, but no evidence for differences related to travel costs or lost income was identified. This review also highlights the need for additional research, including standardization of outcomes and questions around some of the gaps in coverage that occur in too many of these countries, especially as it relates to lost income that persists well beyond the cancer diagnosis and its treatment.