Assessing the financial impact of cancer drug list (CDL) implementation on patients receiving chemotherapy in an ambulatory cancer centre in Singapore

Abstract

Background

On 1^st September 2022, the Cancer Drug List (CDL) was implemented to ensure the long-term affordability of chemotherapy and insurance premiums. This project aimed to determine the financial impact of CDL on out-of-pocket expenses (OOPE), acceptability and financial toxicity (FT) after CDL implementation.

Methods

A cross-sectional study was conducted at National Cancer Center Singapore. We analyzed OOPE before and after implementation of CDL by reviewing billing transactions in Aug-Sept 2022. Acceptability and FT were determined using theoretical framework of acceptability (TFA) and COmprehensive Score for financial Toxicity (COST) tool respectively via survey.

Results

Of the 314 patients finalized bills examined, majority (68.8%) experienced no change in their OOPE, and 69.1% had no OOPE. Most patients (72.6%) were unaware of implementation of CDL. Among patients aware of CDL, majority (61.9%) were accepting, but unsure on how it benefits them. FT was reported as low (mean COST score = 22.4 ± 9.9), and patients ≥65 years old or have at least pre-university education were less likely to experience FT.

Conclusion

Keywords

Cancer financial impact out of pocket expenses financial toxicity theoretical framework of acceptability

Introduction

Cancer represents a significant source of disease burden globally,¹ with an estimated global spending of US dollar$218 billion in 2023.² The use of novel therapies such as immunotherapies also poses concerns over patients’ cost of treatment.^3,4 Cancer patients and long-term survivors are frequently confronted with higher medical bills^5,6 as compared to those without a history of cancer.

Financial burden associated with medical expenses can be defined as “financial toxicity” (FT).⁷ Factors associated with greater FT in cancer patients include the female gender, younger age and unemployment status.^8–10 Other factors associated with high FT include: not having private medical insurance^11,12 and lower education level.¹² High FT may eventually worsen a patient’s health-related quality of life,^10,13 medication adherence,¹⁴ as well as survival rates.¹⁵

Cancer patients in all health systems are said to experience some form of FT.¹⁶ This is true even for patients with access to universal healthcare coverage.^17,18 As costs increase, some of this burden is shifted to patients through higher deductibles, rising copayments and coinsurance,¹⁹ resulting in higher out-of-pocket expenses (OOPE), and possibly greater FT.

In Singapore, healthcare costs are shared between the government and its people.²⁰ Medishield Life (MSHL), a national healthcare insurance provided by the government, will be tapped upon first to cover a portion of medical bills for outpatient chemotherapy. Citizens or permanent residents (PR) then co-pay any remaining amount using their Medisave (MSV), the national healthcare saving scheme, or self-pay as OOPE. Private insurance, such as Integrated shield plans (IP), can also be purchased to enhance medical coverage and reduce OOPE.²¹

Since 1^st September 2022, the Cancer Drug List (CDL) was implemented to ensure the long-term affordability of chemotherapy and insurance premiums. Only listed medications can be reimbursed via MSV, MSHL or IP.²² Further subsidies are also provided for clinically proven and cost-effective medications, via the Standard Drug List (SDL)²³ or Medication Assistance Fund (MAF), for specific indications.²⁴

To date, there has been no studies on the impact of CDL due to recency of policy implementation. The uniqueness of Singapore’s financing framework also makes it difficult to extrapolate the findings from other countries’ policies to Singapore’s context. Our current study therefore aims to (i) assess the impact of CDL on the OOPE for patient’s chemotherapy; and (ii) measure patient’s acceptability and FT score post-implementation of CDL.

Methods

Study design

This is a cross-sectional study conducted at the National Cancer Centre Singapore (NCCS), Singapore’s largest ambulatory care site for cancer. Eligible patients for our study were Singaporeans or PR, aged 18 and above and currently receiving treatment in NCCS. The study cohort were different for the two study objectives.

This study qualifies for exemption from CIRB review by the Singhealth Centralized Institutional Review Board (CIRB2023/2012).

Data collection

Determining OOPE after implementation of CDL

The bills of patients who visited NCCS between 1^st August 2022 and 30^th September 2022, and had their bills finalized by 31^st December 2022, were obtained from the electronic database on 5^th January 2023. Chemotherapy drugs were filtered using anatomical therapeutic chemical codes for anti-neoplastic agents, particularly those starting with L01 and L02. Supplemented by the list on National Cancer Institute,²⁵ the drugs were grouped into conventional or targeted drugs. Patient prices of these drugs in August and September were also extracted from database.

All patients were included if they were on the same chemotherapy protocol 1 month before and after CDL’s implementation and were excluded if they did not receive any form of chemotherapy, had a change in chemotherapy protocol or subsidy status. All line items for oral and intravenous chemotherapy regimes were included in analysis if they were present in both months.

Data regarding the patient’s characteristics (e.g., age, citizenship, gender, ethnicity), clinical (e.g., primary tumour site, chemotherapy treatment, number of drugs) and financial information (e.g. subsidy status) were extracted from the database.

Determining acceptability and FT score post-implementation of CDL

Data regarding acceptability and FT score were collected via a survey from 30^th January 2023 to 09^th February 2023. Patients surveyed had to provide verbal consent and be able to understand English or Mandarin.

Survey consisted of three components; characteristics, acceptability and FT. Patients were recruited via convenience sampling. Data such as the patient’s demographics, clinical and financial information was collected in the first section of survey.

Acceptability was determined using five questions developed for five dimensions of the Sekhon’s theoretical framework of acceptability (TFA), a validated questionnaire to assess the acceptability of healthcare interventions.²⁶

FT was assessed using the COmprehensive Score for financial Toxicity (COST) tool, a measure to assess FT in cancer patients.⁹ It consists of 11 items, each with a 5-point Likert scale from 0 to 4.

Reliability and internal consistency of the COST measure along with TFA was assessed and evaluated using Cronbach’s α coefficient, with values >0.85 considered excellent.²⁷

Data analysis

There were a total of 1117 unique patients during the survey period, and a sample size of 287 was desirable to achieve a confidence level of 95% and 5% margin of error.

For both objectives, patients’ characteristics, change in OOPE, acceptability, level of FT were summarized using descriptive statistics. Continuous variables are summarized as mean (±SD) or median (IQR) while nominal data was summarized using frequencies and percentage.

The difference in selling price of chemotherapy agents was compared using Wilcoxon signed ranked test between the 2 months.

The difference in OOPE in both months was split into two different categories based on whether there was an increase or no increase in OOPE, whereby the “no increase” category consisted of both decrease and no change in OOPE.

Overall acceptability of CDL could be estimated using a question which was framed after a general acceptability construct in the TFA.²⁶ Patients who answered “agree” or “strongly agree” were classified as “accept” while the rest - “no opinion”, “disagree” and “strongly disagree” were deemed as “do not accept”.

Patients’ COST score was summed from all 11 items, ranging from 0 to 44, with a lower score indicating higher financial toxicity. Items 2, 3, 4, 5, 8, 9, and 10 were reversely scored. Using the median COST score, patients were stratified into two categories of FT: Low FT (22-44), and High FT (0-21).²⁸

Bivariate analysis using Chi square and/or Fisher exact tests was then conducted separately between patients’ characteristics with change in OOPE, acceptability and levels of FT. Items with p-value <0.1 were used to conduct a multivariate logistic regression, whereby p-value <0.05 was deemed to be statistically significant predictors of OOPE, acceptability and FT. Adjusted odds ratios (OR) were reported with 95% confidence intervals (CIs).

Lastly, common themes were extracted from patients’ comments using Braun & Clarke’s six phase framework²⁹ and summarized as frequencies and percentages.

Data analysis was carried out using IBM’s SPSS Statistics version 28.

Results

OOPE

Patients’ characteristics

Our study included 314 patients, and their demographics are summarized in Table 1. The median age of patients is 67.0 (IQR: 59.0-73.3) years old, with most of them being Singaporean (95.5%), belonging to the ethnic group of Chinese (81.5%), were female (51.9%) and were subsidized (81.8%). Majority of patients were using one chemotherapy agent (54.5%), some form of targeted therapy (68.5%), and only CDL listed drugs (98.4%).

Table 1.

Descriptive statistics and bivariate analysis of patient’s characteristics with change in OOPE (n = 314).

Characteristics	Total number of patients (%) (n = 314)	OOPE
		Number of patients (%)		p-value
		No increase (n = 249)	Increase (n = 65)	p-value
Age
Median (IQR)	67.0 (59.0–73.3)
• Below 65	132 (42.0%)	100 (40.2%)	32 (49.2%)	0.206
• 65 and above	182 (58.0%)	149 (59.8%)	33 (50.8%)
Gender
• Female	163 (51.9%)	128 (51.4%)	35 (53.8%)	0.781
• Male	151 (48.1%)	121 (48.6%)	30 (46.2%)
Citizenship
•Singaporean	300 (95.5%)	242 (97.2%)	58 (89.2%)	0.012*
•PR	14 (4.5%)	7 (2.8%)	7 (10.8%)
Ethnicity
• Chinese	256 (81.5%)	201 (80.7%)	55 (84.6%)	0.484
• Non-Chinese	58 (18.5%)	48 (19.3%)	10 (15.4%)
o Malay	25 (8.0%)
o Indian	19 (6.1%)
o Others	14 (4.5%)
Subsidy status
• Subsidized	257 (81.8%)	198 (79.5%)	59 (90.8%)	0.046*
• Non-subsidized	57 (18.2%)	51 (20.5%)	6 (9.2%)
Primary tumour site
• Gastrointestinal (GI)	76 (24.3%)	54 (21.7%)	22 (33.8%)	0.118
o Colorectal	36 (11.5%)
o Liver	15 (4.8%)
o Pancreas	11 (3.5%)
o Other GI	14 (4.5%)
• Lung	73 (23.3%)	64 (25.7%)	9 (13.8%)
• Breast	69 (22.0%)	52 (20.9%)	17 (26.2%)
• Others	81 (25.9%)	68 (27.3%)	13 (20.0%)
o Urinary tract	18 (5.8%)
o Female reproductive organs	14 (4.5%)
o Leukemia/Lymphoma	13 (4.2%)
o Head and neck	12 (3.8%)
o Prostate	11 (3.5%)
o Other primary	13 (4.2%)
• Multiple	14 (4.5%)	10 (4.0%)	4 (6.2%)
• Unknown	1 (0.3%)	1 (0.4%)	0 (0.0%)
No. of drugs
• 1 drug	171 (54.5%)	148 (59.4%)	23 (35.4%)	<0.001*
• More than 1 drug	143 (45.5%)	101 (40.6%)	42 (64.6%)
o 2 drugs	102 (32.5%)
o 3 drugs	35 (11.1%)
o 4 drugs	6 (1.9%)
Type of therapy
• Conventional	99 (31.5%)	76 (30.5%)	23 (35.4%)	0.457
• Targeted/Combination	215 (68.5%)	173 (69.5%)	42 (64.6%)
CDL classification
• All drugs on CDL	309 (98.4%)	247 (99.2%)	62 (95.4%)	0.062*
• Not all drugs on CDL	5 (1.6%)	2 (0.8%)	3 (4.6%)
Median OOPE (IQR)
• August	0.00 (0.00-0.00)	-	-	-
• September	0.00 (0.00-4.77)
Change in OOPE
• Median (IQR)	0.00 (0.00-0.00)
• No increase	249 (79.3%)	-	-	-
o No change	216 (68.8%)
o Decrease	33 (10.5%)
• Increase	65 (20.7%)
Presence of OOPE
• August
o No	252 (80.3%)
o Yes	62 (19.7%)	-	-	-
• September
o No	220 (70.1%)
o Yes	94 (29.9%)

IQR, interquartile range; OOPE, out-of-pocket expenses; CDL, cancer drug list; PR, permanent resident.

*Items with p-values <0.1 are included in the multivariate regression.

The median OOPE of all patients is $0.00 (IQR: $0.00-0.00) and $0.00 (IQR: $0.00-4.77) in August and September respectively. Most of them (68.6%) did not experience a change in OOPE. Wilcoxon signed ranked test revealed no significant difference between the median OOPE paid in August and September (p = 0.218).

Association between OOPE and patient characteristics

Potential predictors of change in OOPE include citizenship, subsidy status, number of drugs used and the CDL classification of drugs used (Table 1). After adjusting for confounders, those who were permanent residents (PRs) had 5.14 times (95% CI: 1.62-16.3) the odds of experiencing an increase in OOPE. Subsidized patients were also found to have 3.33 times (95% CI: 1.21-9.14) the odds of having greater OOPE. Lastly, patients who were on more than one drug and on drugs that are not listed on CDL are respectively 3.42 (95% CI: 1.84-6.33) and 17.1 (95% CI: 2.26-130.0) times more likely to experience an increase in OOPE as well (Table 2).

Table 2.

Multivariate analysis of patients who are more likely to experience increase in OOPE (n = 314).

Characteristics	Adjusted odds ratio (95% CI)	p-value
Citizenship
• Singaporean (reference)
• PR	5.14 (1.62–16.3)	0.006*
Subsidy status
• Subsidized	3.33 (1.21–9.14)	0.020*
• Non-subsidized (reference)
No of drugs
• 1 drug (reference)
• More than 1 drug	3.42 (1.84–6.33)	<0.001*
CDL classification
• All drugs on CDL (reference)
• Not all drugs on CDL	17.1 (2.26–130.0)	0.006*

PR, permanent resident; OOPE, out-of-pocket expenses; CI, confidence interval; CDL, cancer drug list.

*p-values <0.05 are considered statistically significant.

Acceptability and FT

Reliability

The TFA and COST tool demonstrated excellent reliability and internal consistency, having a Cronbach’s α of 0.91 and 0.85 respectively.

Patients’ characteristics

A total of 230 patients were surveyed in the ambulatory treatment unit in NCCS. Amongst them, more than half were below the age of 65 (55.7%), female (53.5%), Singaporean (91.7%), Chinese (82.6%) and had education up to secondary level (61.3%). Most were using at least one form of targeted treatment (56.1%) (Table 3).

Table 3.

Descriptive statistics and bivariate analysis of patient’s characteristics with FT measured by COST. (n = 230).

Characteristics	Total number of patients (%) (n = 230)	Number of patients		p-value
Characteristics	Total number of patients (%) (n = 230)	Low FT^a (n = 109)	High FT^a (n = 121)	p-value
Age
Median (IQR)	63.0 (55.8–69.0)	53 (48.6%)	75 (62.0%)	0.047*
• Below 65	128 (55.7%)	56 (51.4%)	46 (38.0%)
• 65 and above	102 (44.3%)
Gender
• Female	123 (53.5%)	54 (49.5%)	69 (57.0%)	0.290
• Male	107 (46.5%)	55 (50.5%)	52 (43.0%)
Citizenship
• Singaporean	211 (91.7%)	101 (92.7%)	110 (90.9%)	0.643
• PR	19 (8.3%)	8 (7.3%)	11 (9.1%)
Ethnicity
• Chinese	190 (82.6%)	91 (83.5%)	99 (81.8%)	0.862
• Others	40 (17.4%)	18 (16.5%)	22 (18.2%)
o Malay	16 (6.96%)
o Indian	18 (7.83%)
o others	6 (2.61%)
Marital status
• Single	61 (26.5%)	27 (24.8%)	34 (28.1%)	0.654
• Married	169 (73.5%)	82 (75.2%)	87 (71.9%)
Education level
• Up to secondary	141 (61.3%)	58 (53.2%)	83 (68.6%)	0.021*
• Pre-university and above	89 (38.7%)	51 (46.8%)	38 (31.4%)
Current employment status
• Unemployed	142 (61.7%)	72 (66.1%)	70 (57.9%)	0.223
• Employed	88 (38.3%)	37 (33.9%)	51 (42.1%)
Most common mode of transport
• Public	164 (71.3%)	73 (67.0%)	91 (75.8%)	0.145
• Private	65 (28.3%)	36 (33.0%)	29 (24.2%)
Subsidy status
• Subsidized	201 (87.4%)	93 (85.3%)	108 (89.2%)	0.429
• Non-subsidized	29 (12.6%)	16 (14.7%)	13 (10.7%)
Type of therapy
• Conventional	101 (43.9%)	55 (50.5%)	46 (38.0%)	0.064*
• Targeted/Combination	129 (56.1%)	54 (49.5%)	75 (62.0%)
Presence of other comorbidities
• None	115 (50.0%)	58 (53.2%)	57 (47.1%)	0.428
• 1 or more	115 (50.0%)	51 (46.8%)	64 (52.9%)
o 1	63 (27.4%)
o 2 or more	52 (22.6%)
Presence of OOPE
• No	189 (82.2%)	92 (84.4%)	97 (80.2%)	0.491
• Yes	41 (17.8%)	17 (15.6%)	24 (19.8%)
Third party reimbursement^b
• No	111 (48.3%)	49 (45.0%)	62 (51.2%)	0.358
• Yes	119 (51.7%)	60 (55.0%)	59 (48.8%)
Awareness of CDL implementation
• Not aware	167 (72.6%)	83 (76.1%)	84 (69.4%)	0.300
• Aware	63 (27.4%)	26 (23.9%)	37 (30.6%)
Mean COST score (SD)	22.4 (±9.9)	-	-	-

FT, financial toxicity; CDL, cancer drug list; PR, permanent resident; OOPE, out-of-pocket expenses, IQR, interquartile range; SD, standard deviation.

*Items with p-values <0.1 are included in the multivariate regression.

^aLow FT is defined as scores 22-44, while high FT is defined as score from 0-21.

^bThird party reimbursement includes integrated shield plans, company insurance, medical claims pro-ration system (MCPS).

Of the patients surveyed, only 63 (27.4%) patients were aware of CDL. In this group, most of them were below the age of 65 (74.6%), married (77.8%), unemployed (60.3%) and used some form of third-party reimbursement (69.8%). More than half (57.1%) of them had education level of pre-university and above and were using at least one form of targeted treatment (55.6%) (Table 4). From their responses, majority were accepting of CDL implementation (61.9%), but were unsure on how it benefited them (68.3%) (Appendix A, Supplementary Table A1).

Table 4.

Descriptive and bivariate analysis of patient’s characteristics with acceptability, measured using TFA question 2. (n = 63).

Characteristics	Total number of patients (%) (n = 63)	Number of patients		p-value
Characteristics	Total number of patients (%) (n = 63)	Do not accept^a (n = 24)	Accept^a (n = 39)	p-value
Age
Median (IQR)	59.0 (50.0–65.0)
• Below 65	47 (74.6%)	21 (87.5%)	26 (66.7%)	0.065*
• 65 and above	16 (25.4%)	3 (12.5%)	13 (33.3%)
Gender
• Female	38 (60.3%)	14 (58.3%)	24 (61.5%)	0.801
• Male	25 (39.7%)	10 (41.7%)	15 (38.5%)
Citizenship
• Singaporean	7 (90.5%)	21 (87.5%)	36 (92.3%)	0.528
• PR	6 (9.5%)	3 (12.5%)	3 (7.7%)
Ethnicity
• Chinese	59 (93.7%)	22 (91.7%)	37 (94.4%)	0.612
• Non-Chinese	4 (6.3%)	2 (8.3%)	2 (5.1%)
o Malay	1 (1.59%)
o Indian	2 (3.17%)
o Others	1 (1.59%)
Marital status
• Single	14 (22.2%)	8 (33.3%)	6 (15.4%)	0.096*
• Married	49 (77.8%)	16 (66.7%)	33 (84.6%)
Education level
• Up to secondary	27 (42.9%)	9 (37.5%)	18 (46.2%)	0.500
• Pre-university and above	36 (57.1%)	15 (62.5%)	21 (53.8%)
Current employment status
• Unemployed	38 (60.3%)	10 (41.7%)	28 (71.8%)	0.018*
• Employed	25 (39.7%)	14 (58.3%)	11 (28.2%)
Most common mode of transport
• Public	40 (63.5%)	16 (66.7%)	24 (61.5%)	0.681
• Private	23 (36.5%)	8 (33.3%)	15 (38.5%)
Subsidy status
• Subsidized	56 (88.9%)	20 (83.3%)	36 (92.3%)	0.271
• Non-subsidized	7 (11.1%)	4 (16.7%)	3 (7.7%)
Type of therapy
• Conventional	28 (44.4%)	9 (37.5%)	19 (48.7%)	0.384
• Targeted/Combination	35 (55.6%)	15 (62.5%)	20 (51.3%)
Presence of other comorbidities
• None	31 (49.2%)	14 (58.3%)	17 (43.6%)	0.256
• 1 or more	32 (50.8%)	10 (41.7%)	22 (56.4%)
o 1	19 (30.2%)
o 2 or more	13 (20.6%)
Presence of OOPE
• No	52 (82.5%)	21 (87.5%)	31 (79.5%)	0.416
• Yes	11 (17.5%)	3 (12.5%)	8 (20.5%)
Third party reimbursement^b
• No	19 (30.2%)	3 (12.5%)	16 (41.0%)	0.017*
• Yes	44 (69.8%)	21 (87.5%)	23 (59.0%)
Mean COST score	21.4 (±9.98)	-	-
FT categories
• Low FT^c	24 (38.1%)	4 (16.7%)	22 (56.4%)	0.003*
• High FT	39 (61.9%)	20 (83.3%)	17 (43.6%)

FT, financial toxicity; CDL, cancer drug list; OOPE, out-of-pocket expenses; PR, permanent residents.

*Items with p-values <0.1 are included in the multivariate regression.

^aDo not accept is defined as those who responded, ‘no opinion’, ‘disagree’ and ‘strongly disagree’ while accept is defined as those who responded ‘agree’ and ‘strongly agree’

^bThird party reimbursement includes integrated shield plans, company insurance, medical claims pro-ration system (MCPS).

^cLow FT is defined as scores 22-44, while high FT is defined as score from 0 to 21.

Association between acceptability and FT with patient characteristics

Patient’s age, marital status, current employment status, presence of third-party reimbursement and FT category were found to be potential predictors of acceptability (Table 4). After adjustment, patients experiencing low FT had 9.42 times (95% CI: 2.16-41.0) the odds of accepting of CDL (Table 5).

Table 5.

Multivariate analysis of factors associated with respondents’ acceptability towards CDL’s implementation. (n = 63).

Characteristics	Adjusted odds ratio (95% CI)	p-value
Age
• Below 65 (reference)
• 65 and above	1.82 (0.340–9.71)	0.486
Marital status
• Single (reference)
• Married	3.37 (0.717–15.9)	0.124
Employment
• Unemployed	3.24 (0.85–-12.3)	0.084
• Employed (reference)
Third party reimbursement^a
• No	3.44 (0.721–16.4)	0.121
• Yes (reference)
FT categories
• Low FT^b	9.42 (2.16–41.0)	0.003*
• High FT (reference)

CI, confidence interval; CDL, cancer drug list; FT, financial toxicity.

*p-values <0.05 are considered statistically significant.

^aThird party reimbursement includes integrated shield plans, company insurance, medical claims pro-ration system (MCPS).

^bLow FT is defined as scores 22-44, while high FT is defined as score from 0 to 21.

Patient’s age, type of therapy and education level were found to be potential predictors of FT (Table 3). After adjustment, patients younger than 65 years old had 2.03 (95% CI: 1.16-3.55) times the odds of experiencing high FT as compared to those aged 65 and above. Patients who had education up to secondary level were also 2.28 times (95% CI:1.29-4.04) more likely to experience high FT as compared to patients who had at least pre-university education (Table 6).

Table 6.

Multivariate analysis of patients who are more likely to experience increase in financial toxicity. (n = 230).

Characteristics	Adjusted odds ratio (95% CI)	p-value
Age
• Below 65	2.03 (1.16–3.55)	0.013*
• 65 and above (reference)
Education level
• Up to secondary	2.28 (1.29–4.04)	0.005*
• Pre-university and above (reference)
Type of therapy
• Conventional (reference)
• Targeted/Combination	1.60 (0.931–2.74)	0.089

CI, confidence interval.

*p-values <0.05 are considered statistically significant.

Among patients who were aware of CDL, all TFA constructs except TFA3 along with the use of targeted therapy was found to be potential predictors of FT. After adjustment, patients who do not accept CDL, and those on targeted therapy were found to have 4.92 (95% CI:1.00-24.1) and 5.74 (95% CI: 1.47-22.3) times the odds respectively to have high FT (Appendix A, Supplementary Table A3).

Qualitative responses from patients

A total of 39 qualitative responses from patients regarding CDL’s implementation were provided by patients (Appendix B). Around half of them expressed a need for greater coverage (53.8%), either through the inclusion of more drugs or subsidized indications (25.6%) or increased claim limits (20.5%). Patients also felt uncertain regarding the effectiveness of CDL (30.8%). Other themes from their comments included the need for greater awareness (17.9%) and concerns regarding delays in billing (10.3%).

Discussion

OOPE

We found that most patients (68.8%) experienced no change in OOPE after the implementation of CDL. This was despite the decrease in claim limits for MSHL and MSV for cancer medications, suggesting that the reduction in claim limits may be largely offset by the reductions in drug prices (Appendix C).

Patients who are PRs are more likely to experience increased OOPE as PRs have less subsidies across all monthly household incomes brackets for medications listed under SDL and MAF^23,24 when compared to citizens. Furthermore, PRs are subjected to higher pro-ration factor for MSHL claims for subsidized outpatient treatment,³⁰ thus reducing the amount claimable and increasing the likelihood of increased OOPE for PRs.

Patients on multiple chemotherapy agents are also more likely to experience an increase in OOPE as patients’ claim limits are capped according to the drug with the greatest MSHL and MSV claim limit, rather than a summation of the claim limits from all chemotherapy agents.²²

Similarly, patients who are using non-listed CDL agents are also likely to pay greater OOPE as their treatments are not subject to any form of subsidy (MAF, SDL), nor can they utilize MSHL or MSV to reimburse a portion of their bills.²²

Subsidized patients were found to be more likely to experience an increase in OOPE.

Based on our preliminary analysis of drugs in NCCS, 93.6% of them had a decrease in MSHL claim limits and 100% of them had a decrease in MSV limits. There was a median decrease of $2200 and $600 for MSHL and MSV respectively (Appendix D). Subsidized patients are likely to experience an increase in OOPE as they are more reliant on public healthcare financing, in contrast to non-subsidized patients who tend to pay for their medical treatment, directly or indirectly through private insurance.³¹ Additionally, non-subsidized patients insured under IP are still covered under the term of their previous policy until April 2023, where IP providers are expected to mirror the coverage of CDL.²²

Acceptability

Patients who were experiencing low FT were more likely to accept CDL’s implementation, suggesting that patient’s acceptability of the scheme was dependent on their FT. This was also reflected amongst patients aware of CDL, whereby low acceptability and usage of targeted therapies were predictors of high FT (Appendix A, Supplementary Table A3).

Targeted therapies are associated with high financial burden due to their exorbitant costs.^4,32 Even though our study did not find an association between the use of targeted therapy and patient’s acceptability of CDL, it could be a possible factor given that it is associated with higher FT amongst those aware of the scheme (Appendix A, Supplementary Table A3).

Awareness of CDL was generally low (27.4%) and it has been found that those aware of CDL were more likely to be younger, be of Chinese ethnicity, possess higher education and utilizing third party reimbursement for their cancer bills (Appendix E, Supplementary Table E2). This is likely because younger and more educated individuals are associated with higher health literacy.^33–35 Patients insured under IP are also kept up to date regarding the CDL and how their policies are affected by it.^36,37 Patients of Chinese descent are more likely to be aware of CDL. However, this may not be true as suggested by another study,³⁸ thus this may be a type 1 error due to small sample size.

Overall, factors influencing awareness were not shown to be predictors of FT nor acceptability. Future studies can be done with greater sample size when the general public is more aware of the scheme to determine predictors of acceptability.

Financial toxicity

Younger patients below the age of 65 were more likely to experience high FT. This is likely due to older patients aged 65 and above enjoying additional subsidies for outpatient care, including ambulatory care. Citizens born between 1950 and 1959 (currently aged 64 to 73 years) are eligible for the Merdeka generation package³⁹ while those born on or before 1949 (currently 74 years or older) are eligible for the Pioneer generation package.⁴⁰ The Pioneer generation package also provides an additional MSV top-up for medical use.⁴⁰ These can serve as an aid to alleviate FT.

Younger patients are also more likely to be on more intensive chemotherapy regimens, which includes the use of multiple chemotherapy agents, as compared to older patients^41,42 as they are generally thought to be able to better tolerate the side effects of such regimens and enjoy better survival outcomes, despite evidence of such benefits being marginal⁴¹ or inconsistent.^41,43 Analysis from our database also suggests that there is an association between age and number of chemotherapy agents used, whereby patients below 65 had 1.60 (95% CI: 1.02–2.51) times the odds of using more than one drug (Appendix F, Supplementary Table F2)

Our study also revealed that having lower education, particularly that of secondary or below, was also found to be associated with higher FT. This was consistent with other studies,^16,44,45 which may arise from the association between lower education and poorer health literacy. In particular, poorer health insurance literacy^46,47 and overall financial literacy⁴⁸ was found to be associated with financial burden amongst cancer patients. This was also supported by our findings whereby more educated patients were more aware of CDL.

Having third party reimbursement was not found to be a predictor of FT (p = 0.358), unlike other studies that found an association between private insurance and lower FT.^11,12 One possible reason for this may be due to the variety of plans available for patients, where some plans cover the bill fully while others have a monthly cap.²¹ IP claims in Singapore also have a minimum co-payment to prevent excessive inflation of healthcare costs .⁴⁹ Alongside how IP providers are expected to follow the CDL and provide coverage for listed medications only, this may subsequently create financial stress for these patients with IP, especially for patients who expect IP to cover all of their medical bills.

Limitations

The findings of our study are limited by the numbers of finalized bills in August and September at NCCS, and the sample size and the use of convenience sampling for the survey. The study would have benefitted from a larger sample size to strengthen the associations identified. However, from the Cronbach alpha analysis, both COST and TFA demonstrated high reliability and internal consistency.

Most patients are unaware of the implementation of CDL, presumably due to the recency of the implementation. Therefore, there is a lack of respondents for the TFA portion and acceptability of this intervention may not be well documented. Efforts to conduct financial counseling and raise awareness of CDL are ongoing at NCCS. Another collaborative study is also being planned to further study the impact of CDL implementation linking it beyond economic to humanistic and clinical outcomes.

Conclusion

Our study found that after CDL implementation, most patients experienced no change in OOPE. Patients who were PR, utilizing multiple chemotherapy drugs or using non-CDL listed medications were likely to experience greater OOPE. The acceptability towards this policy was not well-documented due to the lack of awareness among the public. Generally, FT was reported as low, particularly for patients aged 65 and above, or have at least pre-university education.

The incidence of cancer is expected to increase in Singapore’s ageing population,⁵⁰ resulting in increased spending for cancer treatment. Therefore, greater enquiry towards public and private healthcare reimbursement is necessary to keep cancer treatment affordable for all in the long run.

Despite best efforts from policymakers, it is challenging to achieve a one-size fits funding or subsidy framework that can cater to the needs of all patients. Some groups of patients would inevitably still experience high OOPE and FT due to their personal or clinical circumstances. With our preliminary findings, we hope that targeted interventions can be made to improve the effectiveness of the scheme. This may include the revision of claim limits for patients who are subjected to higher likelihood of increased OOPE, such as those on multiple drugs. More safety nets and subsidy schemes can also be put in place to render assistance to vulnerable groups of patients. Patient education regarding CDL can also be enhanced, particularly towards elderly and those with lower education. Patients can be better empowered to seek financial assistance or resources when necessary.

Supplemental Material

Supplemental Material - Assessing the financial impact of cancer drug list (CDL) implementation on patients receiving chemotherapy in an ambulatory cancer centre in Singapore

Supplemental Material for Assessing the financial impact of cancer drug list (CDL) implementation on patients receiving chemotherapy in an ambulatory cancer centre in Singapore by Darren Lee, Li Qing Lim, Jo Lene Leow and Lita Chew in Proceedings of Singapore Healthcare

Footnotes

Acknowledgements

We would like to thank staff at the Ambulatory Treatment Unit of National Cancer Centre Singapore for their assistance.

Author contributions

DL and LLQ researched literature. All authors were involved in study conception and protocol development. LJL and LC obtained ethical approval. DL and LLQ were involved in patient recruitment and data analysis, and wrote the first draft of the manuscript. All authors reviewed and edited the manuscript and approved the final version of the manuscript.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethical statement

ORCID iDs

Jo Lene Leow

Lita Chew

Data Availability Statement

The datasets generated and/or analysed during the current study are available from the corresponding author.

Supplemental Material

Supplemental material for this article is available online.

References

WHO . Cancer. https://www.who.int/news-room/fact-sheets/detail/cancer (accessed 12 March 2013 2023)

Mikulic

. Global oncology spending from 2011 to 2023. 2023, https://www.statista.com/statistics/696208/oncology-costs-worldwide/ (accessed 12 March 2023).

Abrams

Durbin

Huang

, et al. Financial toxicity in cancer care: origins, impact, and solutions. Transl Behav Med 2021; 11: 2043–2054. DOI: 10.1093/tbm/ibab091.

Chan

Chiang

Low

, et al. Affordability of cancer treatment for aging cancer patients in Singapore: an analysis of health, lifestyle, and financial burden. Support Care Cancer 2013; 21: 3509–3517. DOI: 10.1007/s00520-013-1930-4.

Ekwueme

Zhao

Rim

, et al. Annual out-of-pocket expenditures and financial hardship among cancer survivors aged 18-64 Years - United States, 2011-2016. MMWR Morb Mortal Wkly Rep 2019; 68: 494–499. DOI: 10.15585/mmwr.mm6822a2.

Park

Hwang

. Health care use and expenditures attributable to cancer: a population-based study. Res Social Adm Pharm 2021; 17: 1300–1305. DOI: 10.1016/j.sapharm.2020.09.017.

NCI . Financial toxicity. Bethesda, Maryland: NCI. https://www.cancer.gov/publications/dictionaries/cancer-terms/def/financial-toxicity (accessed 12 March 2023 2023).

Carrera

Kantarjian

Blinder

. The financial burden and distress of patients with cancer: understanding and stepping-up action on the financial toxicity of cancer treatment. CA Cancer J Clin 2018; 68: 153–165. DOI: 10.3322/caac.21443.

de Souza

Yap

Wroblewski

, et al. Measuring financial toxicity as a clinically relevant patient-reported outcome: the validation of the COmprehensive Score for financial Toxicity (COST). Cancer 2017; 123: 476–484. DOI: 10.1002/cncr.30369.

10.

Gordon

Merollini

KMD

Lowe

, et al. A systematic review of financial toxicity among cancer survivors: we can't pay the Co-pay. Patient 2017; 10: 295–309. DOI: 10.1007/s40271-016-0204-x.

11.

de Alcantara Nogueira

Koller

Marcondes

, et al. Validation of the comprehensive score for financial toxicity for Brazilian culture. Ecancermedicalscience 2020; 14: 1158. DOI: 10.3332/ecancer.2020.1158.

12.

Harrison

Meyer

. Financial toxicity is common among gynecologic cancer patients. Int J Gynecol Cancer 2021; 31: 807. DOI: 10.1136/ijgc-2021-002701.

13.

Zafar

McNeil

Thomas

, et al. Population-based assessment of cancer survivors' financial burden and quality of life: a prospective cohort study. J Oncol Pract 2015; 11: 145–150. DOI: 10.1200/jop.2014.001542.

14.

Smith

Lopez-Olivo

Advani

, et al. Financial burdens of cancer treatment: a systematic review of risk factors and outcomes. J Natl Compr Canc Netw 2019; 17: 1184–1192. DOI: 10.6004/jnccn.2019.7305.

15.

Perrone

Jommi

Di Maio

, et al. The association of financial difficulties with clinical outcomes in cancer patients: secondary analysis of 16 academic prospective clinical trials conducted in Italy. Ann Oncol 2016; 27: 2224–2229. DOI: 10.1093/annonc/mdw433.

16.

Longo

Fitch

Banfield

, et al. Financial toxicity associated with a cancer diagnosis in publicly funded healthcare countries: a systematic review. Support Care Cancer 2020; 28: 4645–4665. DOI: 10.1007/s00520-020-05620-9.

17.

Longo

Fitch

Loree

, et al. Patient and family financial burden associated with cancer treatment in Canada: a national study. Support Care Cancer 2021; 29: 3377–3386. DOI: 10.1007/s00520-020-05907-x.

18.

Bygrave

Whittaker

Paul

, et al. Australian experiences of out-of-pocket costs and financial burden following a cancer diagnosis: a systematic review. Int J Environ Res Public Health 2021; 18: 2422. DOI: 10.3390/ijerph18052422.

19.

Ubel

Abernethy

Zafar

. Full disclosure--out-of-pocket costs as side effects. N Engl J Med 2013; 369: 1484–1486. DOI: 10.1056/NEJMp1306826.

20.

HealthHub . Costs and financing. https://www.healthhub.sg/a-z/costs-and-financing/5/costs_and_financing_overall (accessed 12 March 2023 2023).

21.

MOH . About integrated shield plans. https://www.moh.gov.sg/cost-financing/healthcare-schemes-subsidies/medishield-life/about-integrated-shield-plans (accessed12 March 2023 2023)

22.

MOH . Cancer drug list. https://www.moh.gov.sg/home/our-healthcare-system/medishield-life/what-is-medishield-life/what-medishield-life-benefits/cancer-drug-list (accessed 12 March 2023 2023)

23.

MOH . Subsidies for drug on the standard drug list (SDL) at public healthcare institutions. https://www.moh.gov.sg/cost-financing/healthcare-schemes-subsidies/subsidies-for-sdl-drugs-at-public-healthcare-institutions (accessed 12 March 2023 2023)

24.

MOH . Subsidies for drugs on medication assistance fund (MAF) list at public healthcare instituitons. https://www.moh.gov.sg/healthcare-schemes-subsidies/subsidies-for-maf-drugs-public-healthcare-institutions (accessed 12 march 2023 2023)

25.

NCI . List of targeted therapy drugs approved for specific types of cancer. https://www.cancer.gov/about-cancer/treatment/types/targeted-therapies/approved-drug-list (accessed 17 March 2023 2023)

26.

Sekhon

Cartwright

Francis

. Development of a theory-informed questionnaire to assess the acceptability of healthcare interventions. BMC Health Serv Res 2022; 22: 279. DOI: 10.1186/s12913-022-07577-3.

27.

Tavakol

Dennick

. Making sense of Cronbach’s alpha. Int J Med Educ 2011; 2: 53–55. DOI: 10.5116/ijme.4dfb.8dfd.

28.

Huntington

Weiss

Vogl

, et al. Financial toxicity in insured patients with multiple myeloma: a cross-sectional pilot study. Lancet Haematol 2015; 2: e408–e416. DOI: 10.1016/S2352-3026(15)00151-9.

29.

Braun

Clarke

. Using thematic analysis in psychology. Qual Res Psychol 2006; 3: 77–101. DOI: 10.1191/1478088706qp063oa.

30.

CPF . What MediShield Life covers you for. https://www.cpf.gov.sg/member/healthcare-financing/medishield-life/what-medishield-life-covers-you-for (accessed 12 March 2023 2023)

31.

Ramesh

. Realigning public and private health care in southeast Asia. Pac Rev 2008; 21: 171–187. DOI: 10.1080/09512740801990238

32.

Liao

Pan

, et al. Growing financial burden from high-cost targeted oral anticancer medicines among medicare beneficiaries with cancer. JCO Oncol Pract 2022; 18: e1739–e1749. DOI: 10.1200/OP.22.00171.

33.

Lee

Tsai

, et al. Health literacy, health status, and healthcare utilization of Taiwanese adults: results from a national survey. BMC Publ Health 2010; 10: 614. DOI: 10.1186/1471-2458-10-614.

34.

Lee

Toh

, et al. Development and validation of a general health literacy test in Singapore. Health Promot Int 2012; 27: 45–51. DOI: 10.1093/heapro/dar020.

35.

van der Heide

Wang

Droomers

, et al. The relationship between health, education, and health literacy: results from the Dutch Adult Literacy and Life Skills Survey. J Health Commun 2013; 18(Suppl 1): 172–184. DOI: 10.1080/10810730.2013.825668.

36.

AIA . Updateson shield's Coverage for cancer drug treatment, 2022. https://www.aia.com.sg/en/health-wellness/healthcare-services/shield-cancer-coverage-updates (accessed 25 March 2023 2023).

37.

PRUSHIELD . Prudential and pruextra coverage of cancer drug treatments, 2023 https://www.prudential.com.sg/-/media/prudential/pdf/prushield-and-pruextra-cancer-drug-list-cdl.pdf (accessed 25 Mach 2023 2023).

38.

Suppiah

Malhotra

Tan

, et al. Prevalence of health literacy and its correlates from a national survey of older adults. Res Social Adm Pharm 2023; 19: 906–912. DOI: 10.1016/j.sapharm.2023.02.013.

39.

MOH . Merdeka generation package. https://www.moh.gov.sg/cost-financing/healthcare-schemes-subsidies/merdeka-generation-package (accessed 12 March 2023 2023)

40.

MOH . Pioneer generation package. https://www.moh.gov.sg/cost-financing/healthcare-schemes-subsidies/pioneer-generation-package (accessed 12 March 2023 2023)

41.

Kneuertz

Chang

, et al. Overtreatment of young adults with colon cancer: more intense treatments with unmatched survival gains. JAMA Surg 2015; 150: 402–409. DOI: 10.1001/jamasurg.2014.3572

42.

Manjelievskaia

Brown

McGlynn

, et al. Chemotherapy use and survival among young and middle-aged patients with colon cancer. JAMA Surg 2017; 152: 452–459. DOI: 10.1001/jamasurg.2016.5050.

43.

Zhang

Duan

, et al. Does intensified chemotherapy increase survival outcomes of osteosarcoma patients? A meta-analysis. J Bone Oncol 2018; 12: 54–60. DOI: 10.1016/j.jbo.2018.04.001.

44.

Boby

Rajappa

Mathew

. Financial toxicity in cancer care in India: a systematic review. Lancet Oncol 2021; 22: e541–e549. DOI: 10.1016/S1470-2045(21)00468-X.

45.

Cheng

, et al. A systematic review of financial toxicity among cancer patients in China. Asia Pac J Oncol Nurs 2022; 9: 100071. DOI: 10.1016/j.apjon.2022.04.010.

46.

Zhao

Han

Zheng

, et al. Is health insurance literacy associated with financial hardship among cancer survivors? Findings from a national sample in the United States. JNCI Cancer Spectr 2019; 3: pkz061. DOI: 10.1093/jncics/pkz061.

47.

Edward

Rayens

Zheng

, et al. The association of health insurance literacy and numeracy with financial toxicity and hardships among colorectal cancer survivors. Support Care Cancer 2021; 29: 5673–5680. DOI: 10.1007/s00520-021-06036-9.

48.

Khera

Zhang

Hilal

, et al. Association of health insurance literacy with financial hardship in patients with cancer. JAMA Netw Open 2022; 5: e2223141. DOI: 10.1001/jamanetworkopen.2022.23141.

49.

MOH . COS. https://www.moh.gov.sg/docs/librariesprovider5/pressroom/current-issues/cos-2018-media-factsheet---keeping-healthcare-costs-sustainable-for-all-singaporeans.pdf (accessed 12 March 2023 2023)

50.

Annie

LLLF

Eric

Simin

, et al. Singapore cancer registry annual report 2020. Singapore: National Registry Of Diseases Office: MOH, 2022.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.37 MB