Financial toxicity, a term that has moved steadily from health economics journals into mainstream oncology conversations, describes more than a large bill. It is a multidimensional experience encompassing material hardship, psychological distress, and coping behaviors driven by cost, such as skipping doses or cutting household spending. The new study, published in Frontiers in Public Health, took a person-centered approach to this problem using latent profile analysis, a statistical technique that detects hidden subgroups within a population rather than averaging everyone into a single score.

The research team, led by Jing Nie, Peijuan Hu, Wei Qian, Xiaoyan Lu and Yulong Tan of the Department of Thoracic Surgery, recruited patients between June and December 2025 using convenience sampling. Of 350 questionnaires distributed, 331 valid responses remained, a 94.6 percent valid response rate. Each patient completed the Comprehensive Score for Financial Toxicity, an 11-item patient-reported outcome measure in which lower scores signal greater perceived financial distress, along with validated Chinese versions of the Perceived Social Support Scale and the Connor-Davidson Resilience Scale. Crucially, the authors note that the COST measure captures subjective financial well-being rather than directly measuring bank accounts, debt or actual expenditure.

The latent profile analysis compared models containing one through five latent classes using the Akaike Information Criterion, the Bayesian Information Criterion, entropy, and two likelihood-ratio tests. The three-class solution emerged as the clear winner: it achieved an entropy of 0.950, indicating excellent separation between groups, while the four- and five-class models failed to show statistically significant improvements and produced unstable, very small classes. Average posterior probabilities, essentially the statistical confidence with which each patient was assigned to a group, ranged from 0.982 to 0.992, extraordinarily high values that lend credibility to the classification.

The three resulting profiles told a stark story. The Severe Financial Toxicity group contained 95 patients, or 28.7 percent of the sample, with a mean COST score of just 11.1 out of a possible 44. The Moderate group, the largest at 39.9 percent, scored 24.3 on average, while the Mild group, comprising 31.4 percent of patients, averaged 35.3, indicating comparatively manageable financial distress. Importantly, the profiles of the 11 items were broadly parallel across classes, meaning the groups differed mainly in overall severity rather than in qualitatively different patterns of struggling. Financial toxicity, at least in this population, appears to be a matter of degree rather than of kind.

To determine what distinguishes patients in the severe group, the team used the three-step R3STEP procedure in Mplus, a method that adjusts for the uncertainty inherent in latent class assignment, and confirmed the results with conventional multinomial logistic regression. The socioeconomic gradient was dramatic. Compared with patients earning more than 8,000 Chinese yuan per month, those earning under 3,000 yuan had nearly fifteen times the odds of belonging to the Severe rather than the Mild profile, an odds ratio of 14.585. Patients paying for their immunotherapy drugs out of pocket, rather than through medical insurance reimbursement, had roughly thirteen times the odds, and those on sick leave or unemployed had five times the odds compared with employed patients.

The psychosocial findings were equally striking. Every point higher on the Perceived Social Support Scale reduced the odds of severe-profile membership by about 8.5 percent, and every point higher on the resilience scale reduced them by about 9 percent. Expressed on a more intuitive 10-point scale, a ten-point increase in social support was associated with odds of severe financial toxicity only 0.411 times those of comparable patients, while a ten-point increase in resilience corresponded to odds of 0.381. Similar, though weaker, associations held for the Moderate versus Mild comparison. The authors were careful to stress that because all measures were collected at a single point in time, causality cannot be established, and resilient or well-supported patients may simply appraise their finances differently rather than being objectively better protected.

The public health implications reach into policy territory. Because lower income, out-of-pocket payment and job loss clustered with the most severe distress, the authors argue that financial toxicity should not be framed as an individual coping failure but as a signal of upstream determinants: insurance design, reimbursement eligibility, and employment disruption during treatment. They suggest that standardized financial-distress screening integrated into oncology pathways, followed by tiered referral to financial counselors, social workers or financial-navigation services, could identify patients who need help before distress compounds. At the same time, they emphasize that psychosocial interventions, from resilience training to peer support, remain untested hypotheses for reducing financial toxicity and require prospective evaluation before any effectiveness can be claimed.

The study’s limitations temper its reach. It was conducted at a single tertiary center, recruited by convenience sampling, and measured subjective distress rather than objective spending, savings or debt. Clinical variables such as cancer stage, drug regimen, treatment line and cumulative cost were not systematically collected, leaving room for residual confounding, and small subgroups, such as the 34 patients relying on charity drug donations or free trials, produced wide confidence intervals. Because financial toxicity has been shown in longitudinal work to fluctuate across treatment phases, even the Mild group may need periodic reassessment.

Nevertheless, the message lands with force. As immune checkpoint inhibitors extend survival for millions worldwide, the lung cancer population, which accounted for roughly 2.48 million new cases and 1.8 million deaths globally in 2022 according to GLOBOCAN, is growing, and with it the population living under prolonged treatment-related financial strain. This study shows that within a single clinic, the financial burden of modern immunotherapy is not shared equally, and that a quick questionnaire, combined with attention to income, insurance coverage and employment status, could flag the patients most at risk. The next step, the authors say, is to test rigorously whether screening and navigation programs can actually convert these statistical signals into relief for patients.

Subject of Research: Latent profiles of financial toxicity and associated socioeconomic and psychosocial factors among lung cancer patients receiving immunotherapy

Article Title: Financial toxicity profiles and associated factors among lung cancer patients receiving immunotherapy: a latent profile analysis

Article References: Nie, J., Hu, P., Qian, W., Lu, X., & Tan, Y. (2026). Financial toxicity profiles and associated factors among lung cancer patients receiving immunotherapy: a latent profile analysis. Frontiers in Public Health, 14, Article 1909233. https://doi.org/10.3389/fpubh.2026.1909233

Image Credits: AI Generated

DOI: 10.3389/fpubh.2026.1909233

Keywords: financial toxicity, lung cancer, immunotherapy, latent profile analysis, social support, psychological resilience, health economics, out-of-pocket payment, patient-reported outcomes, oncology, China, financial navigation