Stigma scales for smoking, COPD, and lung cancer still carry harmful wording
Stigma measurement in smoking, COPD, and lung cancer has advanced enough to produce dozens of instruments, but the field has not yet solved a basic problem: the words inside these scales can reinforce the very bias they are meant to measure. A COSMIN systematic review identified 28 eligible stigma scales from 962 screened articles, then examined their content, language, and psychometric quality. The central finding was blunt. No scale was free of language that could exacerbate stigma.
That matters at the bedside and in research settings. When a questionnaire uses phrasing that blames, moralizes, or reduces people to a diagnosis or behavior, it can distort responses, elevate distress, and blur the boundary between measurement and harm. For conditions already entangled with public judgment, especially tobacco use and lung disease, the wording of an instrument is not a cosmetic issue. It can shape whether patients answer honestly, disengage from care, or feel further isolated.
Most instruments focus on lung cancer, but wording remains a weak link
Among the 28 scales, 17 measured lung cancer stigma, reflecting how heavily this area has driven the field. The item content across the tools largely aligned with established theoretical dimensions of stigma, suggesting that developers have captured the broad social and psychological features of the construct. Those dimensions are clinically relevant because stigma often operates through several channels at once: anticipated judgment, internalized shame, social exclusion, and assumptions about personal responsibility. In respiratory disease, these forces can delay diagnosis, reduce willingness to seek help, and complicate communication about symptoms and treatment.
Yet content coverage alone does not make a scale fit for routine use. A scale can still be conceptually aligned and linguistically problematic at the same time. The review used the Lung Cancer Stigma Communications Assessment Tool to examine whether wording could intensify stigma among participants, and every scale included at least some language with that risk. In practical terms, that means researchers may be quantifying stigma while simultaneously embedding blame into the assessment itself.
One sentence captures the problem: measurement can become part of the message.
The implications extend beyond psychometrics. In smoking-related disease, patients frequently encounter a moral frame that treats illness as a deserved consequence rather than a complex result of addiction, exposure, social determinants, and biology. That framing can be especially damaging in COPD and lung cancer, where stigma may affect both patient experience and clinician behavior. A scale that mirrors the same rhetoric risks reinforcing social judgment instead of isolating the underlying construct with precision.
Psychometric evidence exists, but most scales remain only partly mature
The review found that psychometric testing was uneven. Most scales assessed internal consistency (_n = 25_), construct validity (_n = 20_), and structural validity (_n = 15_). Those are important steps, but they do not complete the validation pathway. Internal consistency only shows that items move together; it does not prove that a tool captures stigma cleanly, performs across settings, or avoids bias in wording. Construct validity and structural validity strengthen confidence, yet a scale can still underperform if its language pressures respondents or if its items fail to function well across populations.
Every scale was categorized as Class B, meaning each has potential for use but requires further research and psychometric testing. That classification is a warning against treating these tools as finished products. In a clinical area where stigma can shape access to care, screening behavior, and psychological burden, incomplete validation is not a minor methodological footnote. It determines whether findings can be trusted when scales are used in intervention studies, prevalence research, or comparative work across countries and patient groups.
The review’s structure also points to a deeper design problem. If the same scale is meant to detect stigma in smokers, people living with COPD, and people with lung cancer, then wording must avoid collapsing distinct experiences into one generic narrative. The stigma attached to active smoking differs from the stigma attached to a cancer diagnosis, and both differ from the burden of chronic respiratory limitation. A robust instrument has to preserve those distinctions without sounding accusatory. That balance is difficult, but it is exactly where measurement quality lives.
Person-first wording is now a methodological requirement, not an editorial preference
The review closes the loop on a point that has become impossible to ignore in patient-reported outcomes: non-stigmatizing, bias-free, and person-first language is not optional when the construct itself involves shame and social judgment. Instruments that fail this test may still be usable in exploratory work, but they are not ready to serve as the backbone for clinical trials or implementation studies aimed at reducing stigma.
Researchers building the next generation of scales will need to do more than translate older items into cleaner prose. They will need to co-design wording with patients, clinicians, and psychometric experts; test items across diverse groups; and show that the instrument measures stigma without injecting blame into the process. For respiratory illness, that challenge is especially urgent because stigma can delay presentation, distort therapeutic conversations, and leave patients less willing to disclose symptoms or seek follow-up.
The review does not suggest that stigma measurement is futile. It shows the field is still using instruments that are only partly mature. The more immediate problem is methodological honesty: a scale can be theoretically sound and still carry language that harms the people it is meant to study. Until a fully validated, person-first instrument exists, researchers will continue to work with tools that are informative but unfinished.