Part XVI · Disparities, access, and health systems · Chapter 90

Access, affordability, and financial toxicity

Cost is a treatment variable. It changes what is started, what is finished, and who survives.

1 · Insurance coverage, benefit design, and outcomes

Insurance is usually recorded as a category. Insured, Medicaid, uninsured. That categorisation is too coarse for what it is asked to explain, because two insured patients can face entirely different out-of-pocket exposure depending on deductible, coinsurance percentage, out-of-pocket maximum, formulary tier and network breadth.

The category still predicts a great deal. Among 177 075 women aged 40 to 64 diagnosed with stage I to III breast cancer, 20% of those uninsured or covered by Medicaid presented at stage III, against 11% of those with insurance. Insurance status mediated 45% to 47% of the racial difference in odds of locally advanced presentation¹. Coverage is therefore not a background variable in a disparities analysis. It is roughly half of the stage gap.

Benefit design acts later in the pathway than coverage does. Coverage governs whether care begins. Design governs whether it is completed, because the patient meets the deductible in January, encounters the specialty tier when an oral agent is prescribed, and discovers the network boundary when a referral crosses it. Those three events have nothing to do with each other and are all recorded as insured.

The practical consequence for a clinician is narrow and worth stating. Knowing that a patient has insurance tells you they are likely to start treatment. It tells you very little about whether they can afford to finish it, which is the subject of Financial toxicity as a clinical outcome.

2 · Financial toxicity as a clinical outcome

Financial toxicity is the patient-level burden arising from the cost of cancer care. It became a clinical outcome rather than an economic one when it was shown to be measurable and to predict survival.

Measurement first, because the measure defines the construct. The COmprehensive Score for financial Toxicity was validated in 233 patients with stage IV solid tumours receiving chemotherapy. Factor analysis produced a single latent variable. Scores correlated with income, with psychosocial distress and with health-related quality of life as measured by two independent instruments². It is a patient-reported outcome with the psychometric properties of one, not a proxy for income.

Then the outcome. Linking a regional cancer registry to federal bankruptcy records identified 4728 bankruptcy filings among 231 596 people diagnosed with cancer. Propensity score matching left 3841 patients in each group with similar stage, treatment and demographics. The adjusted hazard ratio for death among those who filed for bankruptcy was 1.79, with limits of 1.64 to 1.96³.

Two cautions belong with that number. Bankruptcy is a severe and late marker of financial distress, so the hazard ratio describes an extreme of the distribution and not the average patient. And the mechanism is not established. Treatment abandonment, delayed presentation for complications and psychological stress are all plausible, and the study design cannot separate them.

What the finding does establish is that cost belongs in the same conversation as toxicity. A regimen a patient cannot afford to complete has an effectiveness that is not the one in the trial report.

3 · Access to pathology, molecular testing, and imaging

Every treatment decision in this book rests on an assay. An assay that was not performed produces the same clinical result as an assay that was not available.

Genomic assay use makes the point at scale. Among patients with early-stage, oestrogen receptor-positive, HER2-negative disease in a population-based cohort of 2998 women, 62% of those eligible had no 21-gene assay performed⁴. The assay decides whether adjuvant chemotherapy is recommended, as set out in Prognostic and predictive assays and somatic profiling. A majority of eligible patients in that cohort had that decision made without it.

Imaging access has a facility component that is separately measurable. Among 46 185 abnormal screening mammograms recommended for biopsy, adjustment for individual and neighbourhood factors barely changed the racial difference in time to biopsy. Adding the identity of the screening facility reduced it, from a relative risk of 1.27 to 1.20 for Black against White women at 90 days⁵. Some of the delay is a property of where the mammogram was done rather than of who had it.

Three access barriers operate on molecular testing and they are usually conflated. Whether the laboratory offers the test. Whether the payer covers it. Whether the ordering clinician knows the indication. An institution that solves one of the three has solved none of the others, and only the first is visible in a laboratory audit.

4 · Access to surgery, radiation, and systemic therapy

The three treatment modalities fail differently under resource constraint, and the difference is structural rather than incidental.

Surgery is a single episode. It requires theatre time, anaesthesia and a pathology service, and once delivered it does not recur. Radiation is a course. It requires a linear accelerator within daily travelling distance and attendance on consecutive weekdays for several weeks. Systemic therapy is an extended commitment. It requires a supply chain, an infusion unit or a pharmacy, monitoring, and in the endocrine case adherence for five to ten years.

A patient can therefore be offered all three, complete one, partly complete another and never start the third, without any of that appearing in a record of what was recommended.

The effect of the fractionation schedule on access deserves naming, because it is the one variable clinicians control. Shorter radiation courses reduce the number of trips required, and a course requiring fifteen attendances is accessible to patients for whom twenty-five is not. The clinical evidence for hypofractionation is developed in Fractionation, hypofractionation, and acceleration. Its access consequence is separate from its efficacy, and it is large.

Universal coverage narrows these gaps without closing them. In the Military Health System, median time to breast cancer surgery differed by one day between non-Hispanic Black and non-Hispanic White women. Black women who underwent breast-conserving surgery still had a hazard ratio of 1.45 for death, with limits of 1.06 to 2.01⁶. Removing the payment barrier removed most of the timing difference and did not remove the outcome difference.

5 · Drug pricing, biosimilars, reimbursement, and supply

Four separate mechanisms determine whether a drug reaches a patient, and they are usually discussed as one.

Price is what the manufacturer charges. Reimbursement is what a payer will pay and under what conditions. Out-of-pocket cost is what the patient pays, which is a function of benefit design rather than of price. Supply is whether the product physically exists in the pharmacy. A drug can be cheap, covered, and unobtainable.

Biosimilars act on the first mechanism only. Competitive entry after biologic patent expiry reduces list and net price. It does not by itself change coinsurance percentage, prior authorisation requirements or formulary placement, and a percentage-based patient contribution falls only in proportion to the price reduction achieved.

Older generic cytotoxics illustrate the supply failure most clearly. Low-margin sterile injectables have few manufacturers, and a single production interruption can remove an agent from national supply. The drugs affected are frequently backbone agents rather than novel ones.

The global version of this problem is quantified in Global breast cancer care. A survey of 948 oncologists across 82 countries asked about the twenty most frequently prioritised cancer medicines. Universal availability of each was reported by 9% to 54% of respondents in low-income and lower-middle-income countries, against 68% to 94% in high-income countries. Catastrophic expenditure means spending more than 40% of household consumption net of food. Its risk was reported by 13% to 68% of respondents in lower-middle-income countries against 0% to 9% in high-income countries⁷.

6 · Transportation, employment, childcare, and digital barriers

These are frequently listed as practical inconveniences. They are better understood as a per-visit cost that the treatment plan imposes and the treatment plan can reduce.

Each visit consumes travel time, travel money, lost earnings and childcare. A regimen requiring weekly attendance for twelve weeks imposes that cost twelve times. The total is not recorded anywhere in the medical record, and it is the quantity the patient is actually deciding about when adherence fails.

Employment interacts with treatment schedule in a way that is specific rather than general. Salaried employment with paid leave absorbs appointments. Hourly employment converts every appointment into lost income, and shift work makes fixed daily radiation appointments impossible rather than difficult. The employment variable that matters is not whether the patient works. It is whether their hours are their own.

Travel burden is measurable and its effect is conditional. In a population-based cohort of 2998 patients, long travel time to both diagnosis and surgery was associated with delayed treatment initiation only among Black women. A pattern of short travel to diagnosis followed by long travel to surgery was associated with delay across the whole cohort⁴. That second pattern is a referral-network artefact rather than a residence one.

Digital barriers now sit on the same list. Patient portals, telehealth visits and electronic appointment booking each assume a device, connectivity and literacy. A service that moves a function online has moved it out of reach for part of its population unless it maintains the previous route.

7 · Geographic access, rural care, and treatment deserts

Rural residence is a bundle rather than a variable. It contains distance to the nearest mammography unit, distance to the nearest surgical service, distance to the nearest linear accelerator, local oncology workforce density, and the financial viability of the local hospital. These have different remedies and are frequently analysed as one term.

Distance also acts differently on different parts of the pathway. Screening is annual or biennial, so a long trip is tolerable. Radiation is daily for weeks, so the same trip becomes prohibitive. Surveillance after treatment is intermittent, so it tolerates distance again. A single distance-to-care variable averages across three qualitatively different burdens.

The term treatment desert usually describes the absence of a service within a defined travel time. Two things are worth keeping apart in that definition. A service may never have existed in an area, or it may have closed. The second produces a population that had access and lost it, which behaves differently in screening uptake and in trust.

Trial availability follows the same geography and is quantified in Clinical trial representation and the mechanisms of exclusion. A trial was not available at the patient's own institution for 55.6% of patients across 13 studies, with limits of 43.7% to 67.3%⁸. Distance from an academic centre is therefore not only a treatment access variable. It determines whether investigational treatment is an option at all.

8 · Navigation and financial support interventions

Patient navigation is the best-tested access intervention in oncology, and its trial result is more specific than the way it is usually cited.

The Patient Navigation Research Program compared navigation with usual care in 10 521 participants with abnormal screening results and 2105 with a cancer or precancer diagnosis, across multiple centres. The population was 73% from racial and ethnic minority groups, 40% publicly insured and 31% uninsured. There was no benefit in the first 90 days of care. Between 91 and 365 days there was a benefit, with an adjusted hazard ratio of 1.51 for diagnostic resolution, limits 1.23 to 1.84, and 1.43 for treatment initiation, limits 1.10 to 1.86. Benefit was greatest at centres with the longest delays under usual care⁹.

Both halves of that result are informative and only one is usually quoted.

Navigation did nothing in the first 90 days, which is the window in which most patients are resolved. It worked on the patients who were being lost, which is what it was designed to do. And its effect size depended on how badly the receiving system was performing, which means the intervention substitutes for a system function rather than adding a new one⁹.

References

All 9 references

1. Ko NY, Hong S, Winn RA, Calip GS. Association of insurance status and racial disparities with the detection of early-stage breast cancer. JAMA Oncol 2020 6:385-392. PMID 31917398
2. de Souza JA, Yap BJ, Wroblewski K, 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. PMID 27716900
3. Ramsey SD, Bansal A, Fedorenko CR, et al. Financial insolvency as a risk factor for early mortality among patients with cancer. J Clin Oncol 2016 34:980-986. PMID 26811521
4. Dunn MR, Li D, Emerson MA, et al. A latent class assessment of healthcare access factors and disparities in breast cancer care timeliness. PLoS Med 2024 21:e1004500. PMID 39621782
5. Lawson MB, Bissell MCS, Miglioretti DL, et al. Multilevel factors associated with time to biopsy after abnormal screening mammography results by race and ethnicity. JAMA Oncol 2022 8:1115-1126. PMID 35737381
6. Eaglehouse YL, Georg MW, Shriver CD, Zhu K. Racial differences in time to breast cancer surgery and overall survival in the US Military Health System. JAMA Surg 2019 154:e185113. PMID 30673075
7. Fundytus A, Sengar M, Lombe D, et al. Access to cancer medicines deemed essential by oncologists in 82 countries: an international, cross-sectional survey. Lancet Oncol 2021 22:1367-1377. PMID 34560006
8. Unger JM, Vaidya R, Hershman DL, Minasian LM, Fleury ME. Systematic review and meta-analysis of the magnitude of structural, clinical, and physician and patient barriers to cancer clinical trial participation. J Natl Cancer Inst 2019 111:245-255. PMID 30856272
9. Freund KM, Battaglia TA, Calhoun E, et al. Impact of patient navigation on timely cancer care: the Patient Navigation Research Program. J Natl Cancer Inst 2014 106:dju115. PMID 24938303