Supportive Care in Cancer https://doi.org/10.1007/s00520-022-07044-z REVIEW ARTICLE Financial toxicity of cancer care in low‑ and middle‑income countries: a systematic review and meta‑analysis Andrew Donkor1,2  · Vivian Della Atuwo‑Ampoh3 · Frederick Yakanu4 · Eric Torgbenu1,5  · Edward Kwabena Ameyaw6  · Doris Kitson‑Mills7 · Verna Vanderpuye4  · Kofi Adesi Kyei7  · Samuel Anim‑Sampong7 · Omar Khader8 · Jamal Khader9 Received: 26 January 2021 / Accepted: 5 April 2022 © The Author(s) 2022 Abstract Introduction The costs associated with cancer diagnosis, treatment and care present enormous financial toxicity. However, evidence of financial toxicity associated with cancer in low- and middle-income countries (LMICs) is scarce. Aim To determine the prevalence, determinants and how financial toxicity has been measured among cancer patients in LMICs. Methods Four electronic databases were searched to identify studies of any design that reported financial toxicity among cancer patients in LMICs. Random-effects meta-analysis was used to derive the pooled prevalence of financial toxicity. Sub- group analyses were performed according to costs and determinants of financial toxicity. Results A total of 31 studies were included in this systematic review and meta-analysis. The pooled prevalence of objec- tive financial toxicity was 56.96% (95% CI, 30.51, 106.32). In sub-group meta-analyses, the objective financial toxicity was higher among cancer patients with household size of more than four (1.17% [95% CI, 1.03, 1.32]; p = 0.02; I2 = 0%), multiple cycles of chemotherapy (1.94% [95% CI, 1.00, 3.75]; p = 0.05; I2 = 43%) and private health facilities (2.87% [95% CI, 1.89, 4.35]; p < 0.00001; I2 = 26%). Included studies hardly focused primarily on subjective measures of financial toxicity, such as material, behavioural and psychosocial. One study reported that 35.4% (n = 152 of 429) of cancer patients experienced high subjective financial toxicity. Conclusions This study indicates that cancer diagnosis, treatment and care impose high financial toxicity on cancer patients in LMICs. Further rigorous research on cancer-related financial toxicity is needed. Keywords Cancer · Treatment · Financial toxicity · Low- and middle-income countries Introduction of cancer care, make it difficult to achieve high-quality pre- vention, early detection, diagnosis, treatment, survivorship New cases and deaths from cancer continue to increase in and palliative care services [3]. low- and middle-income countries (LMICs). During the The cost of care is an important barrier to many cancer period 2012–2018, the annual new cancer cases increased patients seeking treatment and care. Several LMICs spend from 8 million to 9.9 million and cancer deaths increased about 4 to 7% of their gross domestic product (GDP) on from 5.3 million to 6.7 million in LMICs [1, 2]. Govern- health, with regional differences in patients’ ability and ments have a responsibility of providing appropriate, acces- willingness to pay for medical and non-medical care [4]. sible and affordable services to the increasing number of In most LMICs, there is little or lack of widespread health cancer patients. However, multiple influential factors, such insurance coverage. Even among patients with health as unpredictable political climate, inadequately trained can- insurance, many are inadequately protected against the cer care providers, poor coordination and the increasing cost costly demands of cancer care because of high costs of insurance, including higher co-payments and increased deductibles. Cancer patients often spend relatively high * Andrew Donkor Andrew.Donkor@uts.edu.au out-of-pocket for cancer care [5]. The financial support of informal carers is substantial; yet estimates of informal Extended author information available on the last page of the article Vol.:(012 3456789) Supportive Care in Cancer caregiving costs in cancer care have been neglected. Can- Eligibility criteria cer patients and informal caregivers who are often, but not always, family members are vulnerable to losing employ- The inclusion criteria were as follows: primary studies of ment and have a greater risk of personal bankruptcy [6, 7]. any design that reported financial toxicity experienced by There remains a lack of a uniform terminology in the cancer patients, studies conducted in any country classified literature to describe the medical and non-medical can- as LMIC by the World Bank Group in 2020 (i.e. LMICs are cer care costs that result in financial burden for cancer categorised into low-income countries [$1045 or less], low- patients and their informal caregivers. A broad definition middle income [$1046 to $4095] and upper-middle-income of financial toxicity was recently proposed as “the possible [$4096 to $12,695]), studies that focused on people with any outcome of perceived subjective financial distress result- type of cancer, studies published in peer-reviewed journals ing from objective financial burden” [8]. Objective finan- and studies published in the English language to capture the cial burden refers to direct costs and indirect care–related current complexity of financial toxicity. Editorials, opinion costs while subjective financial distress include material, pieces, comments, letters, reviews and studies focused on psychosocial stress, negative emotions and behavioural high-income settings were excluded. reactions to cancer care [7, 8]. Terms commonly used interchangeably with financial toxicity include financial or economic difficulty, financial hardship, financial risk and Information sources economic stress [9]. Efforts have been made to develop tools for measuring cancer patients’ risk of experiencing Four electronic databases were searched, namely Ovid financial toxicity, which include COmprehensive Score Embase, Ovid MEDLINE® and In-Process & Other Non- for Financial Toxicity (COST) [10], Personal Financial Indexed Citations, Cumulative Index to Nursing and Allied Wellness (PFW) Scale [11] and Cancer Survivors’ Unmet Health Literature (CINAHL) and Cochrane Library. A hand Needs (CaSUN) measure [12]. These tools were developed search of the reference lists of included studies was per- and/or validated with cancer patients from high-income formed to supplement the database search. countries (HICs) in mind. The lack of practical guidance and tools that are psychometrically acceptable across set- Search strategy tings in LMICs for identifying cancer patients at risk of developing financial toxicity hinders cancer care providers Databases were searched on September 7, 2020. The search from implementing policies. strategy included terms relating to the following concepts: A recent systematic review with included studies mostly cancer, cancer patients, delivery of health care, cost of ill- from HICs identified that cancer patients who were younger, ness, cancer survivors and LMICs. Medical subject head- non-white, unmarried, living with dependents and residing ings, keywords and free text terms were combined using in non-metropolitan service areas are more at risk of finan- “AND” or “OR” Boolean operators. The initial search strat- cial toxicity [13]. There has been proliferation of studies egy was developed in MEDLINE (Ovid) (Supplementary using quantitative design to investigate financial toxicity Table 1). among cancer patients in LMICs. Hence, it seems timely to conduct a systematic review and meta-analysis to objectively summarise the results to address significant gaps in terms of Study selection designing and implementing innovative strategies in LMICs. The study aimed to determine the prevalence, determinants Two authors independently screened titles and abstracts of and how financial toxicity has been measured among cancer citations retrieved by the search for relevance against the patients in LMICs, which will be helpful in future studies of inclusion criteria, and full texts of articles were obtained. financial toxicity. Ten per cent of the articles was independently screened by a third author. Disagreements were resolved through discussion. Methods Data extraction This systematic review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) An electronic data extraction form was developed, and full- guideline [14]. It was registered with the international Pro- text data extraction was performed by three authors. The spective Register of Systematic Reviews (PROSPERO) extracted data was reviewed and discussed in a team meet- (CRD42020207205) [15]. ing, and disagreements were resolved through consensus. 1 3 Supportive Care in Cancer Data extracted included general information, study eligibil- relational, community and societal levels. Individual-level ity, setting, cancer type, study design, data collection, par- factors relate to person characteristics such as age, gender ticipants, outcome measures and results. and health conditions. Relational-level factors are defined by direct person-to-person interaction such as family, peer Quality assessment and social support or withdrawal. Community-level factors pertain to workplaces, neighbourhoods, churches and non- Two reviewers assessed the quality of the included stud- governmental/charity organisations. Social-level factors ies. Qualitative studies were assessed by using the Joanna include policies, laws and social and cultural norms [22]. Briggs Institute Critical Appraisal Checklist for Qualitative Research [16]. Quantitative studies were assessed according to the appropriate Joanna Briggs Institute Critical Appraisal Checklist, such as cross-sectional studies [17] and cohort Results studies [18]. Disagreements were resolved by discussion. To enable comparison, each item in the appraisal checklist The electronic databases searches yielded 4798 articles, was rated using a 3-point scale, with “1 = yes, − 1 = no and with another two identified through hand search. A total 0 = not applicable”. The sum was divided by the number of of 324 articles were excluded due to duplication. The title items in the appraisal checklist and multiplied by 100%. The and abstract of remaining articles were screened, and 4398 risk of bias scores were categorised as ≥ 80% (low), 60 to articles were excluded because they did not meet the inclu- 80% (moderate) and < 60% (high). sion criteria. The full text of the remaining 78 articles was then reviewed for eligibility, of which 31 were found to be Data synthesis and analysis eligible for inclusion. The PRISMA flow diagram provides detail of the screening process (Fig. 1). We used quantitative data to determine the prevalence and determinants of financial toxicity. Meta-analysis was employed for studies that reported quantitative data. A ran- Characteristics of included studies dom-effects meta-analysis of odds ratio (OR) was used to calculate pooled data with 95% confidence intervals (CIs). Table 1 presents the characteristics of included studies. Heterogeneity among studies was estimated using the I2 The 31 studies (30 quantitative and one qualitative) were index, with values classified as “low heterogeneity” (less conducted in four different regions, including Asia (China, than or equal to 25%), “moderate heterogeneity” (26–50%) n = 10; Iran, n = 3; Thailand, n = 3; Turkey, n = 3; Vietnam, and “high heterogeneity” (greater than 50%) [19, 20]. n = 2; and Malaysia, n = 2), Africa (Kenya, n = 2; Ethiopia, Leave-one-out sensitivity analysis was performed to exam- n = 1; and Morocco, n = 1), Middle East (Jordan, n = 1), ine whether single studies had a disproportionally exces- South America (Brazil, n = 1) and Europe (Serbia, n = 1), sive influence. Sub-group meta-analyses were conducted with a multinational study exploring financial toxicity to determine the potential sources of heterogeneity. Forest across Malaysia, Thailand, Indonesia, Philippines, Viet- plots were generated. Probability values below 0.05 were nam, Laos, Cambodia and Myanmar [23]. The quantita- considered statistically significant. Data were analysed using tive data were based on 14 retrospective cohort studies, 11 Review Manager 5.3. cross-sectional studies, four prospective longitudinal and Qualitative data investigates how certain coping strat- one observational cohort study. One-third of the studies egies were adopted to address financial toxicity. A nar- (n = 10) were published in 2018 and one-fifth (n = 7) in rative synthesis was undertaken for studies that reported 2019. The total sample size was 120,883, which ranged qualitative data by comparing similarities and differences from 30 to 45,692 participants. Majority of the partic- across studies [21]. Studies were independently coded by ipants were females (n = 65,564). The mean age of the two authors by applying the socio-ecological framework to participants was 57.7 ± 7.8 years and ranged from 42 to determine the coping strategies adopted to reduce financial 72 years. The majority of the studies focused on specific toxicity. Emerging themes were explored and refined, and cancer types, such as lung [24–28], breast [29–31], colo- any discrepancies were resolved through discussion. The rectal [32, 33], liver [34], ovarian [35], prostate [36] and socio-ecological framework is suitable to provide a multi- stomach [37]. level perspective and structured approach to understand- NR not reported, CHE catastrophic health expenditure, ing coping strategies for reducing financial toxicity among HRQoL health-related quality of life, OOP out-of-pocket, cancer patients in LMICs. It is a four-tier framework for H&N head and neck, GI gastrointestinal, PM pleural meso- organising factors, which then inform corresponding thelioma, M male, F female. coping strategies [22]. The four levels are individual, 1 3 Supportive Care in Cancer Fig. 1 PRISMA flow diagram Records identified through database searching CINAHL (1064) Additional records identified MEDLINE (1108) through other sources EMBASE (2079) (n = 2) Cochrane Library (547) Duplicates removed (n =324) Records screened Records excluded (n = 4476) (n = 4398) Full-text articles excluded, with reasons Full-text articles assessed for (n = 47) Childhood cancers (n = 12) eligibility Rapid review (n = 10) (n = 78) Editorial (n = 9) Wrong setting (n = 7) Non-cancer articles (n = 7) Non-English articles (n = 2) Articles included (n = 31) Prevalence of objective financial toxicity estimates (93.38% [95% CI, 87.21, 99.99]). However, the heterogeneity in the ratio of prevalence was extremely high Three studies provided the prevalence estimates of objec- (I2 = 100%). Objective financial toxicity was categorised into tive financial toxicity [40, 44, 45] enabling a meta-analysis. direct medical costs, direct non-medical costs and indirect The pooled prevalence of objective financial toxicity was costs. 56.96% (95% CI, 30.51, 106.32) (see Fig. 2). The random- effects meta-analysis showed that the pooled prevalence of Direct medical costs objective financial toxicity among cancer patients varied from 17.73% (95% CI, 15.76, 19.94) to 93.38% (95% CI, Table 2 presents the results of the mean estimates of can- 87.21, 99.99) in any cancer type after separating the data cer care costs using random-effects meta-analysis and sub- on rural and urban in one study [44]. Rural dwellers had a group meta-analysis. Direct medical costs were categorised substantially higher prevalence of objective financial toxicity into seven cost items: consultation; diagnosis; treatment, 1 3 Included Eligibility Screening Identification Supportive Care in Cancer 1 3 Table 1 Characteristics of the included studies Authors Year Aim Study design/ Country Sample size Cancer type Cancer stage Gender Mean age (years) Tools for meas- Key results data collection uring financial toxicity a. Included quantitative studies Ting et al. [38] 2020 To determine Prospective, Malaysia 429 Prostate = 366, I = 48, II = 179, III = 85 M = 414 and NR Used the •Greater objective the prevalence cross-sec- bladder = 31 and and IV = 117 F = 15 catastrophic and subjective and associated tional/ques- kidney = 32 health financial toxicities factors of tionnaire expenditure were associated objective and tool and Per- with poor HRQoL subjective sonal Finan- •Patients attending financial tox- cial Wellness a private tertiary icity among Scale to meas- hospital were urologic can- ure financial more likely to cer patients toxicity face objective in Malaysia. financial toxicity Secondly, it (OR = 258.14, investigated 95% CI = 22.16– the associa- 3007.58) tion between •Female respondents financial were more likely toxicity and to face average to HRQoL high subjective financial toxicity (OR = 44.88, 95% CI = 4.58–440.12) Su et al. [39] 2020 To estimate the Cross-sectional/ China 964 Breast = 398, NR M = 366 and NR Material finan- •Almost half of sur- proportion of questionnaire lung = 140, F = 598 cial problem vivors experienced Chinese can- colorectal = 198 questions: material financial cer survivors and stom- How much hardship experiencing ach = 228 did you or •10% of cancer financial your family survivors reported hardship and borrow or experiencing then examine how much behavioural finan- the relation- debt did cial hardship ship between you incur material and because of behavioural your cancer, financial hard- its treatment ship or the lasting effects of that treatment? Behavioural financial hardship questions: Have you ever forgone cancer treat- ments because of worrying about the costs? And if so, which cancer treat- ment is it? Supportive Care in Cancer 1 3 Table 1 (continued) Authors Year Aim Study design/ Country Sample size Cancer type Cancer stage Gender Mean age (years) Tools for meas- Key results data collection uring financial toxicity Kasahun et al. 2020 To examine the Cross-sectional/ Ethiopia 352 Breast = 130, NR M = 94 and 48 ± 13.2 Structured •74.4% of patients [40] incidence of questionnaire cervical = 58, F = 258 questionnaire experienced CHE catastrophic colorectal = 46, based on with mean overall health NPC = 13 and WHO Study expenditure of expenditure others = 105 of Global $2366 per patient and identify Ageing and •Inpatient services associated Adult Health accounted for factors and Household 2-thirds of the total coping strate- income and expenditure with gies among expenditure a mean cost of patients were meas- $1584 attending can- ured based on cer treatment respondents’ services in self-reported Addis Ababa, daily or Ethiopia monthly income and expenditure Zhao et al. [41] 2019 To measure the Cross-sectional/ China 200 General NR M = 96 and 54.87 ± 12.45 Used the •84.5% of the comprehen- question- F = 104 comprehen- patients had medi- sive needs naires sive needs cal insurance of cancer assessment •Patients who were patients and tool (CNAT) younger, female, explore the in cancer with low family possible for patients monthly income, at factors associ- to measure their own expense, ated with their financial more than 3 years needs burden after diagnosis, and with highly educated caregiv- ers had higher score of CNAT (49.13 ± 10.13) •Lowest score of CNAT was the need for physical symptoms (35.12 ± 16.68) Tekin and 2019 To determine Retrospective Turkey 26,664 Breast NR M = 2432 and NR Hospital billing •Total medical cost Saygili [29] the annual cohort/elec- F = 24,232 system of outpatients was direct medical tronic records $73,534,475.5 costs of all •Total medical cost breast cancer of inpatients was patients in $23,159,274.9 Turkey with •Total cost of drugs top-down cost and medical approach equipment was $14,805,009.2 Supportive Care in Cancer 1 3 Table 1 (continued) Authors Year Aim Study design/ Country Sample size Cancer type Cancer stage Gender Mean age (years) Tools for meas- Key results data collection uring financial toxicity Rozman et al. 2019 To describe Retrospective Brazil 2985 Lung = 370; colon NR M = 1629 and 64.4 ± 13.46 Hospital billing •Average cost [42] the resource cohort/elec- and rectal = 368; F = 1356 system per patient was utilisation tronic records H&N = 353; $12,335, ranging and costs stomach = 243; from $8269 for among cancer breast = 221; patients with patients female genital pancreatic cancer by cancer tract = 176; to $19,395 for localisation oesopha- patients with brain and per month gus = 145; cancer of treatment prostate = 139; before death bladder = 94; malignant melanoma = 92; haemato- logical = 90; liver = 87; pancreas = 80; kidney = 71; malignant neoplasm, not otherwise = 65; other sites in the digestive tract = 62; male genital tract = 18; thyroid = 14; others = 146 Piroozi et al. 2019 To investigate Cross-sectional/ Iran 161 Oesophageal = 36, I = 26, II = 30, III = 37 M = 152 and NR World Health •Lack of sup- [43] the prevalence questionnaire stomach = 34, and IV = 68 F = 9 Survey plementary health as well as the colon and Questionnaire insurance and low effective fac- rectum = 67 and developed by socio-economic tors on facing others = 24 the WHO status were the CHE after the significant factors implementa- affecting exposure tion of health to CHE transforma- •The rate of house- tion plan holds facing CHE was 72.7% Moghadam et al. 2019 To investigate Retrospective Iran 499 Prostate cancer NR M = 449 72 ± 9.25 Self-developed •The mean score [36] the economic cohort/ques- question- for HRQoL was burden of tionnaire naire based 0.62 ± 0.16 for all prostate can- on opinions patients cer patients of urology •Chemotherapy and their and oncology patients suffered health-related specialists the worst scores in quality of life and experts the physical well- in the field of being subscale economics (0.47 ± 0.24) Supportive Care in Cancer 1 3 Table 1 (continued) Authors Year Aim Study design/ Country Sample size Cancer type Cancer stage Gender Mean age (years) Tools for meas- Key results data collection uring financial toxicity Leng et al. [44] 2019 To explore the Retrospective China 792 Lung = 187, NR M = 539 and Urban dwellers = 64.75 Unvalidated •94.3% of urban prevalence, cohort/ques- intestinal = 53, F = 253 Rural dwellers = 64.01 questionnaire households and determinants tionnaires gastric = 126, that collected 96.1% of rural and conse- liver = 135, informa- households quences of oesophagus = 86 tion such as spent 40% or CHE among and others = 205 demographic more of their urban and characteris- monthly income rural end-of- tics, health as out-of-pocket life cancer services cancer health care patients in utilisation and expenditure China end-of-life •The poorer the out-of-pocket household, payments the higher the prevalence of catastrophic health expenditure •Health insurance did not adequately compensate for CHE Bhoo-Pathy 2019 To examine the Prospective Malaysia 1294 GI = 345, res- I = 59, II = 153, M = 470 and Median = 53 Unvalidated •Overall incidence of et al. [45] incidence, longitudinal piratory = 62, III = 103, IV = 170 F = 824 questionnaire financial toxicity cost drivers cohort/cost breast = 424, and unknown = 531 that collected among the cancer and factors dairies and female repro- cost data on survivors at 1 year associated questionnaire ductive = 81, conventional was 51% (n = 665), with financial urogenital = 14, medical care, ranging from 33% toxicity after hemato- traditional and in MOH hospitals cancer in an logic = 278 and complemen- to 65% in the upper–mid- other = 90 tary medical public university dle-income care and hospital and 72% country with goods and ser- in the private universal vices related hospitals health cover- to cancer care •Low-income status, age (transporta- type of hospital tion, meals, and lack of health lodging, park- insurance were ing, childcare strong predictors and personal of financial toxicity items directly •Payments for incurred conventional by patients medical care made and not up 39% of the total reimbursed by OOP costs borne insurance by the affected households Supportive Care in Cancer 1 3 Table 1 (continued) Authors Year Aim Study design/ Country Sample size Cancer type Cancer stage Gender Mean age (years) Tools for meas- Key results data collection uring financial toxicity Zheng et al. [46] 2018 To evaluate the Retrospective China 1344 General NR M = 775 and NR Questionnaire •Incidence of CHE medical eco- cohort/medi- F = 569 surveys was 42.78%. Influ- nomic burden, cal records among cancer encing factors were including total and question- patients, length of stay, type current cura- naire which of health insurance tive expendi- collected and location of ture and CHE information of household on cancer demographic •Among the in Liaoning charac- households, those Province, teristics, with oesophageal China household cancer patients income and were most likely expenditure, to experience CHE medical of which the inci- expenses and dence rates were compensation 60.29%, 57.89% and 46.89% Sun et al. [37] 2018 To examine the Retrospective China 14,692 Stomach I = 2357, II = 2590, M = 10,092 and 58.1 ± 12.6 Hospital infor- •Average medical costs of the cohort/medi- III = 3452, IV = 4838 F = 4205 mation system expenses of the first course cal records and unknown = 1060 that collected first course treat- treatments medical ments were about in Chinese expenses $6851 patients with for cancer •Contributing factors stomach treatments included long stay cancer and including pay- in hospital and an the associated ments (both of increased number trends out-of-pocket of episodes of care payments and payments by insurance plans) of each patient for admissions and outpa- tients from the first admission date to the last discharge date Supportive Care in Cancer 1 3 Table 1 (continued) Authors Year Aim Study design/ Country Sample size Cancer type Cancer stage Gender Mean age (years) Tools for meas- Key results data collection uring financial toxicity Saengow et al. 2018 To determine Cross-sectional/ Thailand 437 Colorectal M = 183 and 58.4 ± NR Willingness to •Less than half of [32] the willing- questionnaire F = 254 pay for colo- participants were ness to pay rectal cancer willing to pay for for faecal screening colonoscopy immunochem- questionnaire •Presence of com- ical test and panion, female and colonoscopy family history of and examine cancer were influ- an effect of ential factors proposed co- payment on uptake rates Qiu et al. [34] 2018 To understand Retrospective China 12,342 Liver I = 905, II = 3089, M = 9638 and 54.91 ± 12.29 Individual case– •Pharmaceuticals the medical cohort/ques- III = 4683, IV = 2556 F = 2704 based medical accounted for expenditure tionnaire and unknown = 1109 care cost the biggest part for liver records, with of the medical cancer during data such as expenditure, and it 2002–2011 in demographic, rose from 48.01 to urban areas of diagnostic 52.96% China information and detailed expenditure information relating to different types of service including registration, ward bed, diagnosis, examination, treatment, surgery, labo- ratory, nursing and drugs Perin et al. [24] 2018 To assess the Retrospective Serbia 187 Lung IIIB = 69 and IV = 118 M = 137 and NR Extracted •The average hospi- hospital costs cohort/elec- F = 50 resources and tal cost per patient of diagnosing tronic records procedures was $3309.40 and treating from the •37% of the hospital patients with integrated cost was due to stage IIIB and hospital infor- medication IV non-small mation system cell lung to estimate the cancer cost for each patient Supportive Care in Cancer 1 3 Table 1 (continued) Authors Year Aim Study design/ Country Sample size Cancer type Cancer stage Gender Mean age (years) Tools for meas- Key results data collection uring financial toxicity Owenga and 2018 To assess the Cross-sectional/ Kenya 334 Cervical I = 52, II = 40, III = 63 F = 334 NR Self-designed •Financial challenges Nyambedha financial questionnaire and IV = 129 questionnaire were costs of medi- [47] challenges that collected cation 291 (87%), and sources information cost of travel 281 of financial such as socio- (84%) and cost of assistance for demographics diagnostic tests cervical can- and health 250 (75%) cer patients history, •13% of patients financial received assistance challenges of from charity cervical can- organisation, 27% cer patients, received assistance patient care from friends, 9% and informa- received assistance tion needs from colleagues, and spiritual 10% received needs assistance from relatives and 10% received assistance from church Nguyen, et al. 2018 To estimate Retrospective Vietnam 52 patients and Cervical NR F = 62 NR Reviewed medi- •The unit costs for [48] the medical cohort/medi- 10 experts cal records of precancer services costs for the cal records 52 patients fluctuated from treatment of and expert with cervical $18.26 to $33.31 cervical can- discussion cancers to •The main cost cer patients document the driver of radical medical pro- hysterectomy and cedures, types radiotherapy was and quantity the staff payments of resources (59%) needed for the service Supportive Care in Cancer 1 3 Table 1 (continued) Authors Year Aim Study design/ Country Sample size Cancer type Cancer stage Gender Mean age (years) Tools for meas- Key results data collection uring financial toxicity Liao et al. [30] 2018 To assess the Cross-sectional/ China 2746 Breast I = 546, II = 1236, F = 2746 49.6 ± 10.0 Questionnaire •Overall average economic bur- questionnaire III = 603, IV = 285 comprising of expenditure was den of breast and unknown = 76 demographic $8450 (medi- cancer (BC) characteris- cal expenditure: diagnosis and tics, clinical $7527; non-med- treatment in information ical expenditure: China through and relative $922) a multicentre expenditure •Average loss of cross-sec- information time was $1529 tional study (dates of and to obtain diagnosis and a theoretical treatment, evidence for all medical policy-making expenditure [self-pay and healthcare costs], non-medical expenditure [trans- portation, accommoda- tion, meals, nutrition and employee escort fees]) Supportive Care in Cancer 1 3 Table 1 (continued) Authors Year Aim Study design/ Country Sample size Cancer type Cancer stage Gender Mean age (years) Tools for meas- Key results data collection uring financial toxicity Chen et al. [25] 2018 To examine Cross-sectional/ China 227 Lung I = 12, II = 28, III = 57 M = 159 and 59.48 ± 9.42 Financial •Financial difficulty the effect medical and IV = 130 F = 68 information was perceived of financial records and was collected in 83.7% of the burden, using question- using 4 ques- participants objective and naires tions: How •Mean direct subjective much did medical costs indicators, on you pay for was $2518.83 the HRQoL the medical with a median of in lung cancer expense $1515.01 and a patients last month? range from $60.60 (direct medi- to $18,180.17 cal costs) •27.3% reported that How much did the healthcare cost- you spend on to-income ratio the disease- was less than 40% related expenses other than medical expenses, such as buy- ing health supplements, last month? (direct nonmedical costs) What proportion of your annual household income do you spend on healthcare annually? (healthcare cost-to- income ratio) Have your disease and treatment caused you and your fam- ily financial difficulty? (perceived financial dif- ficulty) Supportive Care in Cancer 1 3 Table 1 (continued) Authors Year Aim Study design/ Country Sample size Cancer type Cancer stage Gender Mean age (years) Tools for meas- Key results data collection uring financial toxicity Atieno et al. [49] 2018 To evaluate the Cross-sectional/ Kenya 412 General NR M = 152 and NR The self- •Patients on chemo- economic questionnaire F = 261 designed therapy alone burden of questionnaire cost an average of treating can- collected $1364.3, surgery cer patients information cost $1265.6, radi- such as patient otherapy $1175.1 demograph- and combination of ics, medicines all 3, $3291.8 per prescribed patient and their costs, cost of radiologic tests, costs of laboratory tests, any surgery and associated costs and quantity and costs of any medical devices used Supportive Care in Cancer 1 3 Table 1 (continued) Authors Year Aim Study design/ Country Sample size Cancer type Cancer stage Gender Mean age (years) Tools for meas- Key results data collection uring financial toxicity Zhuyan et al. 2017 To assess the Prospective, China 123 Ovarian NR F = 123 53.4 ± 9.3 Reviewed medi- •Patients with low [35] association of longitudinal cal records financial status financial sta- cohort/medi- to document had a significantly tus and QoL cal records, informa- higher risk of dete- among Chi- question- tion such as riorating HRQoL nese women naires demographic in physical func- actively and clinical tioning (p = 0.001), undergoing data (age, role functioning chemotherapy marital status, (p = 0.0140), emo- for recurrent education, tional functioning ovarian cancer occupation, (p = 0.021), pain type of insur- (p = 0.010) and ance, financial financial difficul- status and ties (p = 0.003) number of recurrences and intervals between recurrences) Financial status was based on self-reported annual family income minus expenses Quality of life was evaluated using the simplified Chinese ver- sion (3.0) of the European Organization for Research and Treatment of Cancer (EORCT) 30-Item Core Quality of Life Question- naire (QLQ- C30) and the simplified Chinese version of the QLQ-OV28 question- naire which is specific to ovarian cancer Supportive Care in Cancer 1 3 Table 1 (continued) Authors Year Aim Study design/ Country Sample size Cancer type Cancer stage Gender Mean age (years) Tools for meas- Key results data collection uring financial toxicity Wenhui, Sheng- 2017 To analyse Retrospective China Shanghai = 600 General NR Shanghai Shanghai = 63.5 ± 12.4 Hospital records •The average OOP lan [50] the health cohort/medi- Beijing = 600 M = 289 and Beijing = 63.2 ± 13.5 of par- as the proportion services cal records Fuzhou = 600 F = 311 Fuzhou = 67.1 ± 10.0 ticipants were of household’s utilisation Chong- Beijing Chongqing = 67.4 ± 13.7 extracted. capacity to pay was and financial qing = 608 M = 325 and Average total 87.3% (Chong- burden of F = 275 expense per qing), 66.0% insured cancer Fuzhou visit, average (Fuzhou), 33.7% patients and M = 311 and out-of-pocket (Beijing) and identify the F = 289 payments and 19.6% (Shanghai) gaps of finan- Chongqing average reim- cial protection M = 346 and bursement provided by F = 262 rate were insurance in analysed urban China Thongprasert 2015 To evaluate the Cross-sectional/ Thailand 150 Lung I–II = 5 and III– M = 78 and 60.9 ± 10.40 The question- •Patients’ willing- et al. [26] patient and questionnaire IV = 145 F = 72 naire collected ness to pay was public will- information associated with ingness to pay such as socio- quality of life, for a quality- demographic financial difficul- adjusted life data, respond- ties, health insur- year for lung ents’ health ance, diarrhoea cancer treat- status/utility and wealth ments using and willing- Thailand as an ness to pay example Supportive Care in Cancer 1 3 Table 1 (continued) Authors Year Aim Study design/ Country Sample size Cancer type Cancer stage Gender Mean age (years) Tools for meas- Key results data collection uring financial toxicity The Action 2015 To determine, in Prospective, 8-country 4584 Breast = 1667; I = 415, II = 1181, M = 1300 and 51.3 ± 12.4 The question- •31% of participants Study Group this region, in longitudinal (Malaysia, mouth and III = 944 and F = 3284 naire collected incurred financial et al. [23] patients with cohort/ques- Thailand, pharynx = 388; IV = 483 information catastrophe. a first-time tionnaire Indonesia, stomach = 227; such as par- Women had greater diagnosis of Philippines, colon and rec- ticipants’ age, odds of financial cancer and in Vietnam, tum = 622; tra- sex, marital catastrophe than whom surgery Laos, Cam- chea, bronchus status, country men (OR, 1.35; was specified bodia and and lung = 116; of residence, 95% CI, 1.05–1.74) in their initial Myanmar) cervix = 341; highest level treatment uterus = 135; of education plans, the ovary = 179; and attained, incidence others = 779 employ- of financial ment status catastrophe and recent owing to experience out-of-pocket of economic payments for hardship treatment, whether in treatment dis- the previous continuation 12 months (as defined they were by whether unable to such patients make any proceed to necessary hospitalisation household by 3 months), payments and mortality, or needed as well as the assistance to factors associ- do so, annual ated with such household outcomes income and health insur- ance status Supportive Care in Cancer 1 3 Table 1 (continued) Authors Year Aim Study design/ Country Sample size Cancer type Cancer stage Gender Mean age (years) Tools for meas- Key results data collection uring financial toxicity Lkhoyaali et al. 2015 To assess Prospective Morocco 150 Lung, breast and NR M = 61 and 44.7 Participants’ •Economic resources [51] the social, cohort/ques- lymphoma F = 89 demograph- were exceeded in psychological, tionnaire ics, disease 78.7%. 56% used behavioural characteristics banking credits and and economic and social, sold properties. impact on economic and Work lay-off was patient’s fam- psychologi- recorded in 54% ily caregivers cal features were collected with a questionnaire. Psychological impact was assessed using Diagnostic and Statisti- cal Manual of Mental Disorders Ak et al. [27] 2015 To evaluate the Retrospective Turkey 275 PM I–II = 50, III–IV = 221 M = 146 and 63.2 ± 11.2 Medical records •Factors affecting relationship cohort/medi- and unknown = 4 F = 129 of par- the cost were his- between cost cal records ticipants were tology, treatment according to reviewed. type, received treatment type Direct medi- second- and third- and prognosis cal costs were line chemotherapy in malignant estimated as and number of PI the sum of hospitalisations hospital bills attributed to the disease. The phases of care were divided into 3 periods as diagnosis, treatment and terminal phase in chemotherapy and multimo- dality groups Supportive Care in Cancer 1 3 Table 1 (continued) Authors Year Aim Study design/ Country Sample size Cancer type Cancer stage Gender Mean age (years) Tools for meas- Key results data collection uring financial toxicity Nguyen et al. 2013 To estimate the Retrospective Vietnam 129 Breast I = 9, II = 73, III = 35 F = 129 51 ± 9.5 Medical records •Total direct medical [31] direct medical cohort/medi- and V = 12 were reviewed cost for a 5-year cost of a cal records to obtain treatment course 5-year treat- personal was estimated ment course information at $975 per for women (e.g. name, patient (range: with primary age, home $11.7–$3955) breast cancer address), date in Central of admission, Vietnam diagnosis and stage, treatment regimes, item- ized invoices and health insurance participation. Unit costs for treatments received over the study period were acquired from the hospital’s finance department Nazer et al. [52] 2013 To describe Retrospective Jordan 116 General NR M = 65 and 51.7 ± 14.8 Electronic medi- •Mean number the drug cohort/elec- F = 51 cal records of medications utilisation tronic records were reviewed prescribed per pattern and to determine patient were 11.7 drug cost in the total (SD ± 4.7) the treatment number of of cancer medications patients with prescribed, severe sepsis the type of and septic medications, and the cost of each medi- cation Chindaprasirt 2012 To identify Retrospective Thailand 45,692 Colorectal NR M = 24,068 and NR Medical records •Colorectal cancer et al. [33] admission cohort/medi- F = 21,624 were reviewed contributed to 98.5 rates and cal records to collect data per 100,000 adult healthcare such as age, persons admission cost of colo- gender, level rates rectal cancer of hospital, •Average hospital regions of charge per hospital, admission were admission rate $1360.06 and hospital costs Supportive Care in Cancer 1 3 Table 1 (continued) Authors Year Aim Study design/ Country Sample size Cancer type Cancer stage Gender Mean age (years) Tools for meas- Key results data collection uring financial toxicity Edis and Karli- 2007 To evaluate the Observational Turkey 103 Lung NR M = 98 and 64 ± 9.3 Hospital billing •Average survival kaya [28] individual and cohort/medi- F = 5 system were was 6.8 months societal costs cal records reviewed •Average cost of lung cancer to obtained per patient was derived from direct medical $5.480 ± 4.088 our patient costs and •Direct medical cost representative additional was $5.471 ± 4.091 medical costs associated with the diagnosis and treatment of lung cancer b.Included qualitative studies Moradian et al. 2012 To explore, Qualitative, Iran 30 Breast = 7, GI = 5, I = 8, II = 15 and III- M = 11 and 42 years and ranged Semi-structured •Major themes: [53] through quali- descrip- bladder = 2; tes- IV = 7 F = 19 between 19 and interviews financial issues tative semi- tive/semi- tis = 2, lung = 2, 59 years with interview (cost of treatment structured structured sarcoma = 2 and guide focused and interference interviews, interviews other = 10 on the main with their ability to Iranian cancer theme: work), psychoso- patients’ ‘patients need cial issues (social needs to express and significant feelings about others support, the disease, distress and fear impact of from future) and cancer on care satisfaction their daily (accessing infor- life and their mation and nursing experiences care) of existing services’ Supportive Care in Cancer Fig. 2 Random-effects meta- analysis of studies that reported the prevalence of financial toxicity including surgery, radiotherapy, chemotherapy, hormone per breast cancer patients, $1372.50 per any cancer type, therapy, combined modalities and palliative/supportive $10,540.00 per lung cancer patient, to $14,181.30 per pros- care; inpatient care; outpatient care; and follow-up care. In tate cancer patient. Two studies presented data on mean total, 11 studies presented data on mean direct medical costs costs of combined surgery, chemotherapy and radiotherapy [25, 27–29, 31, 36, 37, 40, 42, 49, 52]. Overall mean direct [27, 49], with total costs of $9888.14 (95% CI, $ − 4480.83, medical costs were $2740.18, which ranged from $1953.62 $24,257.12) and a substantial heterogeneity (I2 = 100%). One to $3526.74 per cancer patient. Components of the over- study reported that combined surgery and radiotherapy for all mean direct medical costs included $2366.00 (95% CI, any cancer type resulted in even higher associated direct 1920.76, 2811.24) in any cancer type, $1902.95 (95% CI, medical costs ($1749.35 [95% CI, $1257.90, $2240.80]; $ − 655.85, $4461.74) in lung cancer, $4961.80 (95% CI, p < 0.00001) [49]. $4892.80, $5030.80) in stomach cancer, $91.60 (95% CI, Mean costs of palliative care were measured in four stud- $72.87, 110.33) in breast cancer and $6141.30 (95% CI, ies from Kenya [49], Vietnam [31], Brazil [42] and Turkey $5717.88, $6564.72) in prostate cancer, with GDP per capita [27] with GDP per capita ranging from $1817 to $9042. The ranging from $858 in Ethiopia to $10,262 in China. random-effects meta-analysis estimated direct medical costs Three studies reported data on diagnostic costs [31, 36, attributed to palliative care as $3741.28 (95% CI, $2241.19, 49]. Expressed as random-effects estimates, mean diagnosis $5241.37). Also, two studies conducted in Ethiopia [40] costs were higher for any cancer type ($138.90 [95% CI, and Turkey [29] reported data on costs of outpatient care, $126.59, $151.21]; p < 0.00001), as well as breast cancer which was significantly associated with higher financial in women ($16.02 [95% CI, $15.12, $16.92]; p < 0.00001) burden ($673.03 [95% CI, $488.40, $857.66]; p < 0.00001; and prostate cancer in men ($205.80 [95% CI, $168.32, I2 = 85%). One study from Vietnam [31] with GDP per cap- $243.28]; p < 0.00001). Consultation costs significantly ita of $2715 reported costs of follow-up care in breast cancer favoured higher medical costs (p < 0.00001) [49]. The ratio patients as $356.24 ranging between $311.36 and $401.12 of consultation costs to GDP per capita ranged from 1.77 to per patient. 2.16 in Kenya. Costs of surgery were measured in three studies from Kenya [49], Vietnam [31] and Iran [36] with GDP per capita Direct non‑medical costs ranging from $1817 to $5506. The pooled mean costs of surgery were $1678.80 (95% CI, $62.39, $3295.20; p = 0.04; The total direct non-medical costs as reported by one study I2 = 100%), which varied greatly from breast cancer ($82.35 from Turkey were $334.00 (95% CI, $333.74, $334.26) per [95% CI, $76.86, $87.84]; p < 0.00001) to prostate cancer lung cancer patient [28]. Direct non-medical costs were ($3709.50 [95% CI, $3396.01, $4022.99]; p < 0.00001). On observed to be significant (p < 0.00001). Components of the the other hand, data regarding overall mean costs of radio- direct non-medical costs included disease-related transfer, therapy were available in two studies [31, 36]. A non-sig- accommodation and informal and transportation costs. It nificant increase in total mean costs of radiotherapy favour- was observed that mean transportation costs ($162.00 [95% ing low costs burden was observed ($4131.50 [95% CI, CI, $125.307, $198.693]; p < 0.00001) were responsible for $ − 3923.69, $12,186.69]; p = 0.31; I2 = 100%), with higher 48% of the total direct non-medical costs incurred by lung heterogeneity. The ratio of radiotherapy costs to GDP per cancer patients [28]. Also, informal costs were associated capita ranged from 0.59 in Vietnam to 154.78 in Iran. with significantly higher direct non-medical costs among The sub-group meta-analysis of the total costs of chem- prostate cancer patients, with mean costs of $2454.70 rang- otherapy favouring high financial toxicity was observed ing between $2171.84 and $2737.56 (p < 0.0001) [36]. The ($6555.98 [95% CI, $ − 97.19, $13,014.76]; p = 0.05; ratio of informal costs to GDP per capita ranged from 39.44 I2 = 100%), showing increased mean costs from $476.48 to 49.72 in Iran. 1 3 Supportive Care in Cancer Table 2 Mean estimates of cancer care costs using random-effects meta-analysis and sub-group meta-analysis Cost variable Cancer type No. of Mean (95% CI) I2 (%) P value GDP per capita* Ratio of cost Reference articles range as GDP per capita** Overall medical costs Any cancer type 1 $2366.00 ($1920.76, < 0.00001 $858 223.86–327.65 [37] $2811.24) Lung cancer 1 $3199.25 ($3120.88, < 0.00001 $9042 34.52–36.25 [23] $3277.62) 1 $2518.83 ($1837.99, < 0.00001 $10,262 17.91–31.18 [21] $3199.67) 1 $5.48 ($4.68, $6.28) < 0.00001 $9042 0.05–0.07 [24] Sub-total 3 $1902.95 100 0.14 ($ − 655.85, 4461.74) Stomach cancer 1 $4961.80 ($4892.80, < 0.00001 $10,262 47.67–49.02 [33] $5030.80) Breast cancer 1 $91.60 ($72.87, < 0.00001 $9042 0.81–1.22 [25] $110.33) Prostate cancer 1 $6141.30 ($5717.88, < 0.00001 $5506 103.85–119.23 [32] $6564.72) Total 7 $2740.18 ($1953.62, 100 < 0.00001 – – $3526.74) Consultation Any cancer type 1 $35.70 (32.23, < 0.00001 $1817 1.77–2.16 [40] 39.17) Diagnosis Any cancer type 1 $138.90 ($126.59, < 0.00001 $1817 6.97–8.32 [40] $151.21) Breast cancer 1 $16.02 ($15.12, < 0.00001 $2715 0.56–0.62 [27] $16.92) Prostate cancer 1 $205.80 ($168.32, < 0.00001 $5506 3.06–4.42 [32] $243.28) Total 3 $119.02 ($13.71, 100 0.03 – – $224.33) Surgery Any cancer type 1 $1268.30 ($1098.57, < 0.00001 $1817 60.46–79.14 [40] $1438.03) Breast cancer 1 $82.35 ($76.86, < 0.00001 $2715 2.83–3.24 [27] $87.84) Prostate cancer 1 $3709.50 ($3396.01, < 0.00001 $5506 61.68–73.07 [32] $4022.99) Total 3 $1678.80 ($62.39, 100 0.04 – – $3295.20) Radiotherapy Breast cancer 1 $22.87 ($16.03, < 0.00001 $2715 0.59–1.09 [27] $29.71) Prostate cancer 1 $8242.60 ($7963.29, < 0.00001 $5506 144.63–154.77 [32] $8521.91) Total 2 $4131.50 100 0.31 ($ − 3923.69, $12,186.69) 1 3 Supportive Care in Cancer Table 2 (continued) Cost variable Cancer type No. of Mean (95% CI) I2 (%) P value GDP per capita* Ratio of cost Reference articles range as GDP per capita** Chemotherapy Any cancer type 1 $1372.50 ($1050.30, < 0.00001 $1817 57.80–93.27 [40] $1694.70) Lung cancer 1 $10,540.00 < 0.00001 $9042 73.16–159.97 [23] ($6615.09, $14,464.91) Breast cancer 1 $476.48 ($346.57, < 0.00001 $2715 12.77–22.33 [27] $606.39) Prostate cancer 1 $14,181.30 < 0.00001 $5506 250.70–264.42 [32] ($13,803.62, $14,558.98) Total 4 $6555.98 ($97.19, 100 0.05 – – $13,014.76) Hormone therapy Breast cancer 1 $4.25 ($1.63, $6.87) 0.001 $2715 0.06–0.25 [27] Prostate cancer 1 $2940.40 ($2786.29, < 0.00001 $5506 50.60–56.20 [32] $3094.51) Total 2 $1471.27 100 0.32 – – ($ − 1406.10, $4348.65) Surgery + radio- Any cancer type 1 $1749.35 ($1257.90, < 0.00001 $1817 69.23–123.32 [40] therapy $2240.80) Surgery + chemother- Any cancer type 1 $2547.75 ($1532.40, < 0.00001 $1817 84.34–196.10 [40] apy + radiotherapy $3563.10) Lung cancer 1 $17,210.25 < 0.00001 $9042 189.96–190.71 [23] ($17,176.28, $17,244.22) Total 2 $9888.14 100 0.18 – – ($ − 4480.83, $24,257.12) Palliative/supportive Any cancer type 1 $976.65 ($749.60, < 0.00001 $1817 41.25–66.25 [40] care $1203.70) 1 $12,327.00 < 0.00001 $8717 135.40–147.42 [49] ($11,803.00, $12,851.00) Sub-total 2 $6649.27 100 0.24 – – ($ − 4473.87, $17,772.40) Lung cancer 1 $1897.00 ($1849.16, < 0.00001 $9042 20.45–21.51 [23] $1944.84) Breast cancer 1 $4.50 ($2.02, $6.98) 0.0004 $2715 0.07–0.26 [27] Total 4 $3741.28 ($2241.19, 100 < 0.00001 – – $5241.37) Inpatient care Any cancer type 1 275.10 (241.87, < 0.00001 $1817 13.31–16.97 [40] 308.33) 1 1584.00 (1193.76, < 0.00001 $858 139.13–230.10 [37] 1974.24) Sub-total 2 $914.52 ($ − 367.84, 98 0.16 $2196.88) Breast cancer 1 $26.38 ($24.28, < 0.00001 $2715 0.89–1.05 [27] $28.48) Total 3 $436.51 ($190.65, 99 0.0005 – – $682.36) 1 3 Supportive Care in Cancer Table 2 (continued) Cost variable Cancer type No. of Mean (95% CI) I2 (%) P value GDP per capita* Ratio of cost Reference articles range as GDP per capita** Outpatient care Any cancer type 1 $782.00 ($638.85, < 0.00001 $858 74.46–107.83 [37] $925.15) Breast cancer 1 $591.60 ($572.87, < 0.00001 $9042 6.34–6.75 [25] $610.33) Total 2 $673.03 ($488.40, 85 < 0.00001 $857.66) Follow-up care Breast cancer 1 $356.24 ($311.36, < 0.00001 $2715 11.47–14.77 [27] $401.12) Indirect costs Lung cancer 1 $17.34 ($11.87, < 0.00001 $9042 0.13–0.25 [24] $22.80) Prostate cancer 1 $4873.93 ($3604.88, < 0.00001 $5506 65.47–111.57 [32] $6142.98) Total 2 $2402.47 98 0.32 – – ($ − 2356.15, $7161.09) Direct non-medical Lung cancer 1 $334.00 ($333.74, < 0.00001 $9042 3.69–3.70 [24] $334.26) Informal care Prostate cancer 1 $2454.70 ($2171.84, < 0.0001 $5506 39.44–49.72 [32] $2737.56) CI confidence interval, GDP gross domestic product. *The 2019 data. **The ratio of cost of care to GDP per capita was computed by dividing the mean cost range by the GDP per capita and multiplying by 100. Indirect costs Three studies highlighted coping behaviours at the indi- vidual level, which included using personal savings, selling Two studies conducted in Iran and Turkey with GDP per cap- assets, skipping bill payments, borrowing or incurring bank ita ranging from $5506 to $9042 reported quantitative data debt and delaying/forgoing treatment [40, 47, 51]. Two stud- on indirect non-medical costs [28, 36]. The overall pooled ies identified coping behaviours at the relational level, such mean indirect costs were $2402.47 (95% CI, $ − 2356.15, as receiving financial support from family and friends and $7161.09), with $17.34 (95% CI, $11.87, $22.80) per lung emotional support from partners, friends and family mem- cancer patient and $4873.93 (95% CI, $3604.88, $6142.98) bers [40, 53]. The major coping behaviour at the commu- per prostate cancer patient. However, there was high hetero- nity level was seeking financial assistance from workplaces, geneity (I2 = 98%). neighbourhoods, churches and non-governmental/charity organisations to cover the financial toxicity imposed on can- Prevalence of subjective financial toxicity cer patients and their household [40, 54]. There were two main coping behaviours at the social level, which included Included studies hardly focused primarily on subjective creating supportive policies (e.g. a waiver to help cancer measures of financial toxicity, such as material, behavioural patients offset their medical bills) and promoting a pleasant and psychosocial. We were unable to provide pooled preva- social support environment, such as food, accommodation lence of subjective financial toxicity because only one study and transport for treatment programme [53, 54] (see Fig. 3). provided prevalence estimate. The study reported that 35.4% (n = 152 of 429), 11.9% (n = 51 of 429) and 52.7% (n = 226 Determinants of objective financial toxicity of 429) of cancer patients experienced high, average and low subjective financial toxicity, respectively [38]. Psychosocial It was challenging to perform a meta-analysis of the fac- issues identified in one qualitative study were anxiety and tors associated with subjective financial toxicity because social relationship disruption through conflict and criticism the instruments and domains differed across studies. Fig- [53]. ure 4 presents pooled estimates of the determinants of objective financial toxicity. The sub-group meta-analyses 1 3 Supportive Care in Cancer Fig. 3 Coping strategies for reducing financial toxicity showed that cancer patients with a household size of more 0.27, 2.03]; p = 0.55; I2 = 97%) or income level (1.74% than four were associated with a significant increase in [95% CI, 0.68, 4.47]; p = 0.25; I2 = 97%). objective financial toxicity (1.17% [95% CI, 1.03, 1.32]; p = 0.02; I2 = 0%). There was no significant heterogeneity Measuring financial toxicity among the three included studies [38, 43, 46]. The meta- analysis revealed that cancer patients who received more Over one-third of the studies used unvalidated question- than six cycles of chemotherapy were almost two times naires to measure the financial toxicity related to cancer more likely to experience high financial toxicity (1.94% care [23, 25, 26, 30, 35, 36, 39, 44, 45, 51, 54]. Answers to [95% CI, 1.00, 3.75]; p = 0.05; I2 = 43%). In three of the questions, such as “How much did you pay for the medical included studies [40, 45, 46], it was observed that can- expense last month?” and “How much did you spend on cer patients who attended private health facilities during the disease-related expenses other than medical expenses?”, the course of their disease were statistically associated were often used to measure the objective financial toxicity with high-level financial toxicity (2.87% [95% CI, 1.89, during cancer treatment and care [25]. Catastrophic health 4.35]; p < 0.00001; I2 = 26%). One study indicated that expenditure was defined as “when previous one year patient prolonged length of hospital stay was significantly related households’ out-of-pocket expenditure for cancer care to cancer patients encountering higher objective financial exceeded 10% of total annual household income” [40]. One toxicity (1.88% [95% CI, 1.68, 2.11]; p < 0.00001) [46]. study applied a pre-existing generic financial assessment Using data from six studies [23, 38, 43–46], the pooled instrument, namely the PFW scale, which consists of five estimate for health insurance as a determinant of objec- items on the psychosocial, two items on financial resources tive financial toxicity among cancer patients was not a and one item on coping strategies [38]. One study utilised significant factor (1.19% [95% CI, 1.00, 1.42]; p = 0.06; the Chinese version of the cancer-specific comprehensive I2 = 33%). However, according to the leave-one-out sensi- needs assessment tool (CNAT) [41]. One-fifth of the stud- tivity analysis, the random-effects meta-analysis showed ies obtained financial information through hospital billing that not having health insurance was a significant risk systems [27–29, 31, 33, 42, 52]. factor for exposure to objective financial toxicity (1.29% Three instruments were used in six studies to measure the [95% CI, 1.03, 1.61]; p < 0.03; I2 = 42%) when removing health-related quality of life (HRQoL) of cancer patients in one study from China [44] from the pooled analysis. The general and disease-specific aspects of life [23, 25, 26, 35, sub-group meta-analyses indicate no statistically signifi- 36, 38]. The most frequently used HRQoL instrument was cant association with cancer-related objective financial the Functional Assessment of Cancer Therapy (FACT). In toxicity by gender (0.97% [95% CI, 0.65, 1.45]; p = 0.89; particular, the FACT is a two-part instrument that assesses I2 = 70%), stage at diagnosis (1.16% [95% CI, 0.79, 1.70]; general HRQoL related to cancer and cancer therapy p = 0.46; I2 = 32%), level of education (0.73% [95% CI, 1 3 Supportive Care in Cancer Fig. 4 Forest plot showing determinants of objective finan- cial toxicity (FACT-G) and tumour-specific measures, such as prostate which consists of 30 core items with five functional scales (FACT-P). (cognitive, emotional, physical, role and social), three Another instrument that was often used in the assess- symptom scales (fatigue, pain and vomiting/nausea) and ment of HRQoL in cancer patients was the European a global health and quality-of-life scale [26, 35]. Quality of Life Five Dimension (EuroQol/EQ-5D), which measured well-being in five dimensions: usual activities, Quality assessment self-care, pain/discomfort, anxiety/depression and mobil- ity [23, 26]. The EuroQol/EQ-5D combines self-assess- Supplementary Fig. 1 presents the results of the quality ment with a valuation of quality of life in which full health assessment of the included quantitative studies. Sixteen is scored at “one” and death at “zero”. Two studies used studies achieved an overall low risk of bias. Fifteen of the European Organisation for Research and Treatment of the quantitative studies were mainly rated low on overall Cancer Quality of Life Questionnaire (EORTC QLQ-C30), quality. Thirteen of the quantitative studies were rated as 1 3 Supportive Care in Cancer Fig. 4 (continued) 1 3 Supportive Care in Cancer a moderate risk of bias often because there were no identi- need to develop a simple and cost-effective instrument that fication of potential confounders, evidence of strategies to is applicable to LMICs. deal with effects of confounding factors and/or description The limited data in this study does not show clear evi- of statistical adjustment in data. Overall, two of the quan- dence that health insurance is a determinant of financial titative studies were rated as a high risk of bias because of toxicity. Data from six studies did not reach statistical sig- outcome measurement and statistical analysis issues. Out- nificance [23, 38, 43–46]. However, the inclusion of data come measurement issues were due to the use of unvalidated from China may, in part, explain this [44]. A recent study instruments and lack of clear definition and documentation has demonstrated that government’s health insurance cover- of outcomes. The qualitative study demonstrated low risk age significantly increased utilisation of expensive targeted of bias. It showed sufficient quality in terms of underlying anti-cancer medicines and improved patient’s affordability research method, data collection and analysis [53]. [58]. Despite the insufficient data to examine the relationship between health insurance and financial toxicity, it is critical to implement strategies to make health insurance systems Discussion sustainable and facilitate access to affordable cancer treat- ment and care. Previous studies have also highlighted that This systematic review and meta-analysis describes the prev- rural dwellers are less likely to access cancer treatment and alence of cancer-related financial toxicity, its determinants care due to the lack of health insurance, travel distance and and how it has been measured in LMICs based on avail- financial burden [59, 60]. Innovative strategies, such as tele- able data published from 2007 to 2020. The prevalence of consultation and cancer patient–assisted travel schemes, can objective financial toxicity among cancer patients in LMICs be implemented to reduce rural–urban health inequities by varied significantly, ranging from 17.73 to 93.38%. There are decreasing out-of-pocket costs. several direct medical costs, direct non-medical costs and The review shows that household size of more than four, indirect costs that have an impact on cancer patients, their multiple cycles of chemotherapy and private health facilities families and friends. For instance, the mean direct medi- are significantly associated with objective financial toxicity. cal costs per cancer patients were $2740.18 and the costs It is well known that cancer drugs remain unaffordable in attributable to surgery, radiotherapy, chemotherapy, hor- most LMICs, with a large number of cancer patients delay- mone therapy and palliative care were $1678.80, $4131.50, ing or skipping treatment resulting in decreased quality of $6555.98, $1471.27 and $3741.28, respectively. Direct life. To prevent the potential financial and clinical harms, non-medical costs, which included disease-related transfer, it is critical to provide cost-effective cancer care by reduc- accommodation and informal and transportation costs, were ing overuse of anti-neoplastic medication [61]. Also, cancer hardly measured in the studies reviewed. Similarly, there is patient groups, health professionals and governments can limited knowledge when it comes to measuring subjective engage pharmaceutical companies to implement policies or financial toxicity and included studies scarcely focused on interventions to lower the cost of cancer drugs. The associa- it. This finding confirms previous observation that there is tion between large household size and objective financial a lack of accepted definition of subjective financial toxicity toxicity is consistent with the literature on financial toxic- [55]. ity in traumatic injury [62]. Large household size in most The review shows the frequent use of unvalidated or unre- LMICs can be explained by the high infant mortality that liable instruments for measuring financial toxicity among translates into insecurity in families about the survival of cancer patients in LMICs. Unvalidated instruments may their children [63]. Previous studies have indicated that a generate data that do not contribute to a better understand- large number of children result in the decline of parents’ par- ing of cancer patients’ financial difficulties because that data ticipation in the labour force [64]. It also reduces household cannot be interpreted effectively. Similar results have been savings which exposes larger families to income shortfalls. reported by a previous systematic review, which synthesised Long-term, community ownership, community-led partner- methods for measuring financial toxicity after cancer diag- ship and results-based interventions must be considered to nosis with most of the included studies conducted in HICs, ensure sustainable development, poverty and child mortality such as the USA and UK [8]. Few standardised instruments reduction in LMICs. have been developed and validated in an attempt to quan- The results from this systematic review and meta-analysis tify financial toxicity in cancer patients. Examples of such support previous systematic reviews [65–67] and individual instruments include Breast Cancer Finances Survey Inven- studies [7, 9] showing that adult patients with newly diag- tory [56], PFW Scale [11] and COST [10, 57]. These tools nosed cancer experience significantly objective financial tox- were developed in HICs and available mostly in these coun- icity and impaired HRQoL. It is important to note that the tries where cancer patients’ experience of financial toxicity deteriorating HRQoL occurred in several domains, including differs from their counterparts in LMICs. Thus, there is a physical well-being, social well-being, emotional well-being 1 3 Supportive Care in Cancer and functional well-being. As demonstrated by a study from data analysis was done by AD and consensus discussions and finalising HIC, financial toxicity directly impacts the complete well- with VDA-A, ET, FY, EKA, DK-M, VV, JY, KAK, SA-S, JK and OK. being of gastrointestinal cancer patients with higher earn- Table design was completed by AD, DK-M and FY. All authors read and approved the final manuscript. ers reporting less challenges with accessing community resources, pain, fatigue, anxiety and depression [68]. Funding Open Access funding enabled and organized by CAUL and The results of this study show that cancer patients in its Member Institutions LMICs often need to finance their medical and non-medical costs by using personal savings, selling assets, skipping bill Data availability All data generated or analysed during this study are payments, borrowing or incurring bank debt. Waiving medi- included in this published article. cal bills and implementing social policies that assist with Code availability N/A. necessities, such as food, accommodation and transport for treatment are critical coping strategies to reduce the finan- Declarations cial impact on cancer patients and their families. However, previous studies [69, 70] have reported that most African Ethics approval and consent to participate This article is based on a countries have limited or no social protection systems to secondary analysis of the existing literature and does not contain any provide safety nets for patients, thereby forcing unsustain- studies with human participants or animals performed by any of the able coping strategies that increase the risk of bankruptcy. authors. The PRISMA guideline for reporting systematic and meta-analysis was followed. Consent to participate is not applicable. Strengths and limitations Consent for publication N/A. Strengths of this study include the comprehensive search Competing interests The authors declare no competing interests. strategies, rigorous selection criteria and a thorough review process. This is the first systematic review and meta-analysis Open Access This article is licensed under a Creative Commons Attri- to identify the extent of cancer-related financial toxicity and bution 4.0 International License, which permits use, sharing, adapta-tion, distribution and reproduction in any medium or format, as long how it has been measured in LMICs. There are limitations as you give appropriate credit to the original author(s) and the source, in this study. First, substantial heterogeneity in the included provide a link to the Creative Commons licence, and indicate if changes studies was detected. Hence, we applied random-effects were made. The images or other third party material in this article are model, which allows for the true effect to vary between stud- included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in ies. We also used sub-group analysis to help with the inter- the article's Creative Commons licence and your intended use is not pretation of results. Second, it was challenging to explicitly permitted by statutory regulation or exceeds the permitted use, you will model cost variables and determinants in the meta-analysis need to obtain permission directly from the copyright holder. To view a due to several reasons, including incomplete reporting and copy of this licence, visithttp://c reat ivecom mons. org/l icen ses/b y/4. 0/. the limited number of included studies. References Conclusion 1. Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A (2018) Global cancer statistics 2018: GLOBOCAN estimates of This systematic review and meta-analysis indicate that incidence and mortality worldwide for 36 cancers in 185 coun- cancer diagnosis, treatment and care impose high financial tries. CA Cancer J Clin 68(6):394–424 toxicity on cancer patients in LMICs. More high-quality 2. Bray F, Jemal A, Grey N, Ferlay J, Forman D (2012) Global cancer research on cancer-related financial toxicity is needed, par- transitions according to the Human Development Index (2008–2030): a population-based study. Lancet Oncol 13(8):790–801 ticularly from Africa. Future research needs to create and 3. Milroy MJ (2018) Outlining the crisis in cancer care. In: Hope- validate an instrument that will be available to LMICs to wood P, Milroy MJ editors. Quality cancer care: survivorship measure financial toxicity in cancer patients. before, during and after treatment. Springer; Cham, pp 1-12 4. O’Grady ET, Johnson J (2014) Health policy issues in changing environment. In: Hamric AB, Hanson CM, Tracy MF, O’Grady Supplementary Information The online version contains supplemen- ET (eds) Advanced practice nursing: an integrative approach, 5th tary material available at https://d oi.o rg/1 0.1 007/s 00520-0 22-0 7044-z. edn. Elsevier, Missouri, pp 579–606 5. Desai A, Gyawali B (2020) Financial toxicity of cancer treatment: Author contribution Study protocol and design were developed by moving the discussion from acknowledgement of the problem to AD, VDA-A, FY, ET, DK-M and EKA. All authors contributed to the identifying solutions. EClinicalMedicine 20:100269–100272 development of the manuscript. The article search and management 6. Yousuf ZS (2015) Financial toxicity of cancer care: it’s time to were performed by AD. Article screening was completed by ET, EKA intervene. J Natl Cancer Inst 108(5):1–4 and AD. Data extraction was completed by DK-M, FY and AD. Quality 7. Carrera PM, Kantarjian HM, Blinder VS (2018) The financial assessment and study description were performed by AD and FY. The burden and distress of patients with cancer: understanding and 1 3 Supportive Care in Cancer stepping-up action on the financial toxicity of cancer treatment. associated with surgically operable cancer in South-East Asia: CA Cancer J Clin 68(2):153–165 results from the ACTION Study. Surgery. 2015;157(6):971-82 8. Witte J, Mehlis K, Surmann B, Lingnau R, Damm O, Greiner W 24. Perin J, Zaric B, Dodic JE, Potic Z, Potic M, Sekerus V et al et al (2019) Methods for measuring financial toxicity after cancer (2018) The cost of hospital treatment of advanced stage lung diagnosis and treatment: a systematic review and its implications. cancer patients in a developing South East European country. J Ann Oncol 30(7):1061–1070 Cancer 9(17):3038–3045 9. Lentz R, Benson Iii AB, Kircher S (2019) Financial toxicity in 25. Chen JE, Lou VW, Jian H, Zhou Z, Yan M, Zhu J et al (2018) cancer care: prevalence, causes, consequences, and reduction Objective and subjective financial burden and its associations with strategies. J Surg Oncol 120(1):85–92 health-related quality of life among lung cancer patients. Support 10. de Souza JA, Yap BJ, Hlubocky FJ, Wroblewski K, Ratain MJ, Care  Cancer 26(4):1265–1272 Cella D et al (2014) The development of a financial toxicity 26. Thongprasert S, Crawford B, Sakulbumrungsil R, Chaiyakunap- patient-reported outcome in cancer: the COST measure. Cancer ruk N, Petcharapiruch S, Leartsakulpanitch J et al (2015) Willing- 120(20):3245–3253 ness to pay for lung cancer treatment: patient versus general public 11. Prawitz A, Garman ET, Sorhaindo B, O’Neill B, Kim J, Drentea values. Int J Technol Assess Health Care 31(4):264–270 P (2006) InCharge financial distress/financial well-being scale: 2 7. Ak G, Metintas S, Kose T, Bogar F, Girginer N, Batirel HF development, administration, and score interpretation. J Financial et al (2015) The relationship between the cost of treatment and Couns Plan 17(1):34–50 prognosis in malignant mesothelioma in Turkey. J Thorac Oncol 12. Hodgkinson K, Butow P, Hunt GE, Pendlebury S, Hobbs KM, Lo 2:s631–s2 SK et al (2007) The development and evaluation of a measure to 28. Edis CE, Karlikaya C (2007) The cost of lung cancer in Turkey. assess cancer survivors’ unmet supportive care needs: the CaSUN Tuberkuloz ve Toraks 55(1):51–58 (Cancer Survivors’ Unmet Needs measure). Psycho-Oncol J Psy- 29. Tekin RN, Saygili M (2019) Determining breast cancer treat- chol Soc Behav Dimens Cancer 16(9):796–804 ment costs using the top down cost approach. Eur J Breast Health 1 3. Mols F, Tomalin B, Pearce A, Kaambwa B, Koczwara B 15(4):242–248 (2020) Financial toxicity and employment status in cancer sur- 3 0. Liao XZ, Shi JF, Liu JS, Huang HY, Guo LW, Zhu XY et al (2018) vivors. A systematic literature review. Support Care Cancer Medical and non-medical expenditure for breast cancer diagnosis 28(12):5693–708 and treatment in China: a multicenter cross-sectional study. Asia 1 4. Moher D, Liberati A, Tetzlaff J, Altman DG, Group PRISMA Pac J Clin Oncol 14(3):167–178 (2009) Preferred Reporting Items for Systematic Reviews 31. Nguyen HL, Wongsa L, Stewart JF, Nguyen Dinh T, Coyte PC and Meta-Analyses: the PRISMA statement. PLoS Medicine (2013) Cost of treatment for breast cancer in central Vietnam. 6(7):e1000097–e103 Glob Health Action 6:1–10 15. Donkor A, Atuwo-Ampoh VD, Yakanu F, Torgbenu E, Ameyaw 3 2. Saengow U, Birch S, Geater A, Chongsuwiwatvong V (2018) E, Kitson-Mills D (2020) Financial toxicity of cancer care in low Willingness to pay for colorectal cancer screening and effect and middle-income countries: a systematic review. University of of copayment in Southern Thailand. Asian Pac J Cancer Prev York. PROSPERO: CRD42020207205. https://w ww.c rd.y ork.a c. 19(6):1727–1734 uk/ prosp ero/ displ ay_ record. php? Recor dID= 207205. Accessed 3 3. Chindaprasirt J, Sookprasert A, Wirasorn K, Limpawattana P, 06/10/2020 Sutra S, Thavornpitak Y (2012) Cost of colorectal cancer care 16. Joanna Briggs Institute. (2017) The Joanna Briggs Institute criti- in hospitalized patients of Thailand. J Med Assoc Thai 95(Suppl cal appraisal tools for use in JBI systematic reviews checklist for 7):s196–s200 qualitative research. Available from:https://j oanna brigg s.o rg/s ites/ 34. Qiu W-Q, Shi J-F, Guo L-W, Mao AY, Huang H-Y, Hu G-Y et al defaul t/fi les/2 019-0 5/J BI_C ritic al_A pprai sal-C heckl ist_f or_Q uali (2018) Medical expenditure for liver cancer in urban China: a tative_R esea rch201 7_0. pdf. Accessed 01/10/2020 10-year multicenter retrospective survey (2002–2011). J Cancer 17. Joanna Briggs Institute. (2017) The Joanna Briggs Institute critical Res Ther 14(1):163–170 appraisal tools for use in JBI systematic reviews checklist for ana- 3 5. Zhuyan S, Tao Z, Ping Z, Qiang W, Dan L, Shihua W et al (2017) lytical cross-sectional studies. Available from:https:// jbi.g lobal/ Association of financial status and the quality of life in Chinese sites/ defau lt/ files/ 2019- 05/ JBI_ Criti cal_ Appra isal- Check list_ women with recurrent ovarian cancer. Health Qual Life Outcomes for_ Analy tical_ Cross_ Secti onal_ Studi es2017_ 0. pdf. Accessed 15:1–8 01/10/2020 3 6. Moghadam MJF, Ayati M, Rangchian M, Pourmand G, Haddad 1 8. Joanna Briggs Institute. The Joanna Briggs Institute critical P, Nikoofar A et al (2019) Economic burden of prostate cancer appraisal tools for use in JBI systematic reviews checklist for in Iran: measuring costs and quality of life. Middle East J Cancer cohort studies. 2017 [Available from:https://j bi.g lobal/s ites/d efau 10(2):139–155 lt/ files/ 2019- 05/ JBI_ Criti cal_ Appra isal- Check list_ for_ Cohort_ 3 7. Sun X-J, Shi J-F, Guo L-W, Huang H-Y, Yao N-L, Gong J-Y et al Studie s2017_ 0. pdf. Accessed 01/10/2020 (2018) Medical expenses of urban Chinese patients with stomach 1 9. McKenzie JE, Beller EM, Forbes AB (2016) Introduction to sys- cancer during 2002–2011: a hospital-based multicenter retrospec- tematic reviews and meta-analysis. Respirology 21(4):626–637 tive study. BMC Cancer 18(1):1–13 20. Higgins JPT, Thompson SG, Deeks JJ, Altman DG (2003) Meas- 3 8. Ting CY, Teh GC, Yu KL, Alias H, Tan HM, Wong LP (2020) uring inconsistency in meta-analyses. BMJ 327(7414):557–560 Financial toxicity and its associations with health-related quality 21. Lucas PJ, Baird J, Arai L, Law C, Roberts HM (2007) Worked of life among urologic cancer patients in an upper middle-income examples of alternative methods for the synthesis of qualitative country. Support Care Cancer 28(4):1703–1715 and quantitative research in systematic reviews. BMC Med Res 39. Su M, Lao J, Zhang N, Wang J, Anderson RT, Sun X et  al Methodol 7(1):4–9 (2020) Financial hardship in Chinese cancer survivors. Cancer 22. Cramer RJ, Kapusta ND (2017) A social-ecological framework 126(14):3312–3321 of theory, assessment, and prevention of suicide. Front Psychol 4 0. Kasahun GG, Gebretekle GB, Hailemichael Y, Woldemariam AA, 8(1756):1–10 Fenta TG (2020) Catastrophic healthcare expenditure and coping 2 3. The Action Study Group, Jan S, Kimman M, Peters SA, Woodward strategies among patients attending cancer treatment services in M. Financial catastrophe, treatment discontinuation and death Addis Ababa. Ethiopia BMC Public Health 20(1):1–10 1 3 Supportive Care in Cancer 41. Zhao X-S, Wang H-Y, Zhang L-L, Liu Y-H, Chen H-Y, Wang Y 5 7. de Souza JA, Yap BJ, Wroblewski K, Blinder V, Araújo FS, (2019) Prevalence and risk factors associated with the comprehen- Hlubocky FJ et al (2017) Measuring financial toxicity as a clini- sive needs of cancer patients in China. Health Qual Life Outcomes cally relevant patient-reported outcome: the validation of the 17(1):102–112 COmprehensive Score for financial Toxicity (COST). Cancer 42. Rozman LM, Campolina AG, Lopez RM, Chiba T, De Soarez 123(3):476–484 PC (2019) Palliative cancer care: costs in a Brazilian quaternary 5 8. Diao Y, Qian J, Liu Y, Zhou Y, Wang Y, Ma H et al (2019) How hospital. Support Palliat Care 15:1–8 government insurance coverage changed the utilization and afford- 4 3. Piroozi B, Zarei B, Ghaderi B, Safari H, Moradi G, Rezaei S et al ability of expensive targeted anti-cancer medicines in China: an (2019) Catastrophic health expenditure and its determinants in interrupted time-series study. J Glob Health 9(2):020702–020711 households with gastrointestinal cancer patients: evidence from 59. Pace LE, Mpunga T, Hategekimana V, Dusengimana J-MV, new health system reform in Iran. IJHRH 12(4):249–257 Habineza H, Bigirimana JB et al (2015) Delays in breast cancer 4 4. Leng A, Jing J, Nicholas S, Wang J (2019) Catastrophic health presentation and diagnosis at two rural cancer referral centers in expenditure of cancer patients at the end-of-life: a retrospective Rwanda. The Oncologist. 20(7):780–8 observational study in China. BMC Palliat Care 18(1):43–53 60. Ambroggi M, Biasini C, Del Giovane C, Fornari F, Cavanna L 45. Bhoo-Pathy N, Ng C-W, Bhoo-Pathy NT, Saad M, Taib NA, Lim (2015) Distance as a barrier to cancer diagnosis and treatment: GC-C et al (2019) Financial toxicity after cancer in a setting with review of the literature. Oncologist 20(12):1378–1385 universal health coverage: a call for urgent action. J Oncol Pract 6 1. Schleicher SM, Bach PB, Matsoukas K, Korenstein D (2018) 15(6):e537–e46 Medication overuse in oncology: current trends and future impli- 46. Zheng A, Duan W, Zhang L, Bao X, Mao X, Luo Z et al (2018) cations for patients and society. Lancet Oncol 19(4):e200–e208 How great is current curative expenditure and catastrophic health 62. Murphy PB, Severance S, Savage S, Obeng-Gyasi S, Timsina LR, expenditure among patients with cancer in China? A research Zarzaur BL (2019) Financial toxicity is associated with worse based on “System of Health Account 2011.” Cancer Med physical and emotional long-term outcomes after traumatic injury. 7(8):4036–4043 J Trauma Acute Care Surg 87(5):1189–1196 4 7. Owenga JA, Nyambedha EO (2018) Perception of cervical cancer 63. Raza J, Warne GL (2012) Disorders of Sexual Development. In: patients on their financial challenges in Western Kenya. BMC Elzouki AY, Harfi HA, Nazer HM, Stapleton FB, Oh W, Whit- Health Serv Res 18(1):N.PAG-N.PAG ley RJ, editors. Textbook of clinical pediatrics. Springer, Berlin, 48. Nguyen AD, Hoang MV, Nguyen CC (2018) Medical costs for 3649–3674. https://d oi. org/1 0.1 007/ 978-3-6 42- 02202-9_ 383 the treatment of cervical cancer at central hospitals in Viet- 64. Orbetta A (2006) The more the poorer: Why large family size nam. Health Care Women Int 39(4):442–449. https:// doi. org/ 10. cause poverty. Policy Notes 2006(6):1–6 1080/ 07399 332. 2017. 14029 12 6 5. Altice CK, Banegas MP, Tucker-Seeley RD, Yabroff KR (2016) 49. Atieno O, Opanga S, Kurdi A, Martin A, Godman B (2018) Financial hardships experienced by cancer survivors a systematic Assessing the direct medical cost of treating patients with cancer review. J Natl Cancer Inst 109(2):djw205–djw21 in Kenya: a pilot study-findings and implications for the future. 6 6. Chan RJ, Gordon LG, Tan CJ, Chan A, Bradford NK, Yates P Pharmacoepidemiol Drug Saf 27(Supplement 2):302–312 et al (2019) Relationships between financial toxicity and symptom 50. Wenhui M, Shenglan T, Ying Z, Zening X, Wen C (2017) Finan- burden in cancer survivors: a systematic review. J Pain Symptom cial burden of healthcare for cancer patients with social medical Manage 57(3):646–662 insurance: a multi-centered study in urban China. Int J Equity 6 7. Gordon LG, Merollini KMD, Lowe A, Chan RJ (2017) A sys- Health 16:1–12. https:// doi. org/ 10. 1186/ s12939- 017- 0675-y tematic review of financial toxicity among cancer survivors: we 51. Lkhoyaali S, El Haj MA, El Omrani F, Layachi M, Ismaili N, can’t pay the co-pay. Patient-Patient-Centered Outcomes Res Mrabti H et al (2015) The burden among family caregivers of 10(3):295–309 elderly cancer patients: prospective study in a Moroccan popula- 6 8. LaRocca CJ, Li A, Lafaro K, Clark K, Loscalzo M, Melstrom LG tion. BMC Res Notes 8:347–350 et al (2020) The impact of financial toxicity in gastrointestinal 52. Nazer L, Al-Shaer M, Hawari F (2013) Drug utilization pattern cancer patients. Surgery 168(1):167–172 and cost for the treatment of severe sepsis and septic shock in 69. Parmar D, Williams G, Dkhimi F, Ndiaye A, Asante FA, Arhinful critically ill cancer patients. Int J Clin Pharm 35(6):1245–1250 DK et al (2014) Enrolment of older people in social health protec- 5 3. Moradian S, Aledavood SA, Tabatabaee A (2012) Iranian cancer tion programs in West Africa – does social exclusion play a part? patients and their perspectives: a qualitative study. Eur J Cancer Soc Sci Med 119:36–44 Care 21(3):377–383 7 0. MacLean LM (2002) Constructing a social safety net in Africa: an 5 4. Owenga JA, Nyambedha EO (2018) Perception of cervical cancer institutionalist analysis of colonial rule and state social policies in patients on their financial challenges in Western Kenya. BMC Ghana and Côte d’Ivoire. Stud Comp Int Dev 37(3):64–90 Health Serv Res 18(1):261–268 5 5. Pauge S, Surmann B, Mehlis K, Zueger A, Richter L, Menold N Publisher's note Springer Nature remains neutral with regard to et al (2021) Patient-reported financial distress in cancer: a system- jurisdictional claims in published maps and institutional affiliations. atic review of risk factors in universal healthcare systems. Cancers (Basel) 13(19):5015–5033 56. Given BA, Given CW, Stommel M (1994) Family and out- of-pocket costs for women with breast cancer. Cancer Pract 2(3):187–193 1 3 Supportive Care in Cancer Authors and Affiliations Andrew Donkor1,2  · Vivian Della Atuwo‑Ampoh3 · Frederick Yakanu4 · Eric Torgbenu1,5  · Edward Kwabena Ameyaw6  · Doris Kitson‑Mills7 · Verna Vanderpuye4  · Kofi Adesi Kyei7  · Samuel Anim‑Sampong7 · Omar Khader8 · Jamal Khader9 Vivian Della Atuwo-Ampoh Faculty of Health, University of Technology Sydney, vdatuwo-ampoh@uhas.edu.gh New South Wales, Sydney, Australia Frederick Yakanu 2 Department of Medical Diagnostics, Faculty of Allied f.yakanu@gmail.com Health Sciences, Kwame Nkrumah University of Science Eric Torgbenu and Technology, Kumasi, Ghana eric.l.torgbenu@student.uts.edu.au 3 Department of Medical Imaging, School of Allied Health Edward Kwabena Ameyaw Sciences, University of Health and Allied Sciences, Ho, edward.k.ameyaw@student.uts.edu.au Ghana 4 Doris Kitson-Mills National Centre for Radiotherapy, Korle-Bu Teaching doriskitsonmills@gmail.com Hospital, Accra, Ghana 5 Verna Vanderpuye Department of Physiotherapy and Rehabilitation Sciences, vanaglat@yahoo.com University of Health and Allied Sciences, Ho, Ghana 6 Kofi Adesi Kyei The Australian Centre for Public and Population Health kakyei@ug.edu.gh Research (ACPPHR), Faculty of Health, University of Technology Sydney, Sydney, NSW, Australia Samuel Anim-Sampong 7 tychicus77@gmail.com Department of Radiography, University of Ghana, Accra, Ghana Omar Khader 8 omar.j.khader@gmail.com Faculty of Medicine, University of Jordan, Amman, Jordan 9 Jamal Khader Radiation Oncology Department, King Hussein Cancer jkhader@khcc.jo Centre, Amman, Jordan 1 Improving Palliative, Aged and Chronic Care Through Clinical Research and Translation (IMPACCT), 1 3