A population-based study on social inequality and barriers to healthcare-seeking with lung cancer symptoms

The study investigates why individuals with potential lung cancer symptoms may delay or avoid contacting a general practitioner and how these barriers vary by social factors, including age, sex, smoking status, and socioeconomic position. Lung cancer remains the leading cause of cancer deaths, and prognosis strongly depends on stage at diagnosis; yet many patients present late. Public campaigns have improved some symptom awareness, but healthcare-seeking is influenced by social context, lifestyle, and socioeconomic status. Prior population studies suggest only about 40% with lung cancer symptoms contact a GP, with even lower rates among current smokers. Given persistent social inequalities in lung cancer outcomes—largely but not solely linked to smoking—understanding barriers across both socioeconomic and smoking strata is essential to inform earlier diagnosis and tailored interventions.

The authors contextualize their work with evidence that social inequalities persist in lung cancer incidence, stage at diagnosis, and survival. Smoking is the major risk factor and contributes substantially to observed inequalities, but socioeconomic determinants also play roles. Screening can improve early detection but has challenges (false positives, psychosocial effects, low participation in high-risk groups), and symptomatic presentation remains important. Previous studies often assessed socioeconomic status without incorporating smoking status, limiting understanding of barriers across risk groups. Qualitative literature indicates stigma, shame, normalization of symptoms, and fear of blame among smokers as reasons for delayed help-seeking. Anticipated barriers studies (e.g., ABC measures) show country differences and associations with delayed intentions; however, evidence based on diagnosed patients may not reflect general population behavior, underscoring the need for population-based data including smoking status.

Design: Population-based cross-sectional survey within the Danish Symptom Cohort (DaSC), 2012. Sampling: 100,000 Danish residents aged ≥20 years randomly selected from the Civil Registration System were invited (June–December 2012). Exclusions (n=4,747): death, unknown address, severe illness (e.g., dementia), language barriers, moved abroad. Eligible: 95,253; respondents: 49,706 (52.2%). For this analysis, respondents aged ≥40 years with complete data were included; after excluding those <40 and 1,517 with missing data (mainly smoking status), final N=35,938. Data collection: Web-based questionnaire with telephone option; non-responders received reminders and phone follow-up. Questionnaire: Report of 44 symptoms (including lung cancer-specific and non-specific), timing, GP contact, and—if no GP contact—barriers to healthcare-seeking. Barriers were predefined items from the validated Awareness and Beliefs about Cancer (ABC) measures: "Being too embarrassed", "Being worried about wasting the doctor's time", "Being worried about what the doctor might find", and "Being too busy"; free-text option for other barriers. Smoking status: never, former, current (self-reported). Lung cancer symptoms analyzed: specific symptoms: prolonged coughing (>4 weeks), shortness of breath, haemoptysis, prolonged hoarseness (>4 weeks); non-specific: weight loss, loss of appetite, tiredness. Register linkage (Statistics Denmark): highest education, marital status, labour market affiliation (working, pension, out of workforce, disability pension), ethnicity (Danish vs immigrants/descendants), plus death/migration during data collection. Analyses: Descriptive statistics for symptom prevalence and GP contact proportions, including combinations of specific and non-specific symptoms. Among those without GP contact, proportions reporting each predefined barrier and no barriers were calculated; due to Danish legislation, haemoptysis barrier results are not reported because of small cell sizes. Associations between barriers (for prolonged coughing, shortness of breath, prolonged hoarseness) and covariates (sex, age group 40–54/55–69/≥70, smoking status, education, marital status, labour market affiliation, ethnicity) were examined using multivariate logistic regression to estimate crude and adjusted odds ratios (95% CI). Adjusted models included age, smoking status, labour market affiliation, educational level, and ethnicity; interactions between sex, age, and smoking status were tested and not found. Software: Stata 16.1; significance p<0.05. Ethics: Participation voluntary with informed consent; no clinical follow-up; data protection approval; ethics committee determined no further approval required.

- Response and sample: Of 95,253 eligible invitees, 49,706 responded (52.2%); analysis included 35,938 respondents aged ≥40 with complete data. Women constituted 53.2%; 20.9% were current smokers; 62.0% working; 4.1% out of workforce. - Symptom prevalence and GP contacts: Prolonged hoarseness was most frequent (8.9%); haemoptysis least frequent (0.1%). GP contact proportions as single symptoms ranged from 31.3% (prolonged hoarseness) to 54.5% (shortness of breath). Reporting an additional specific lung cancer symptom increased GP contact by approximately 5–20% (largest increases for prolonged coughing +11.9% and haemoptysis +19.9%); adding a non-specific symptom generally did not further increase GP contacts except for haemoptysis (small numbers). GP contact proportions were higher in older age groups, former smokers, lower education, and those out of workforce. - Barriers among those with no GP contact: Between 38.2% and 47.7% reported no barriers, depending on symptom. "Being too embarrassed" was least frequent: ~2.3% (prolonged hoarseness) to 3.2% (shortness of breath). For all symptoms, about one in six reported "Being worried about wasting the doctor's time" and "Being too busy". "Being worried about what the doctor might find" was notably frequent for prolonged coughing (15.5%) and shortness of breath (15.1%). - Social inequalities in barriers (adjusted ORs): Older age groups had significantly lower odds of reporting being "too busy" across symptoms and lower odds of worrying about wasting the doctor's time for prolonged coughing (Adj OR ~0.48, 95% CI 0.32–0.71 for ≥70 vs 40–54). Current smokers, compared to never smokers, were more likely to report "worried about what the doctor might find" for prolonged coughing (Adj OR 3.26, 95% CI 2.26–4.69) and shortness of breath (Adj OR 3.40, 95% CI 2.28–5.06), and for shortness of breath were almost five times more likely to report "too embarrassed" (Adj OR 4.73, 95% CI 1.93–11.61). Immigrants/descendants with prolonged coughing had higher odds of reporting "too embarrassed" (Adj OR 3.48, 95% CI 1.49–8.13). Individuals out of the workforce and on pension/disability were substantially less likely to report being "too busy" (e.g., out of workforce: Adj OR 0.33, 95% CI 0.15–0.69 for prolonged coughing; 0.29, 95% CI 0.13–0.62 for shortness of breath). Individuals out of workforce also showed higher odds of emotional barriers (embarrassment and worry about findings) in crude and some adjusted analyses. Overall, almost two-thirds of non-consulters reported at least one predefined barrier, with practical (time) and emotional (worry) barriers prominent.

The study addresses the research question by quantifying barriers to GP contact for lung cancer symptoms in a large, representative general population sample and demonstrating clear social patterning. Findings highlight that practical barriers (e.g., being too busy) are more prominent in younger and employed individuals, while emotional barriers (worry about findings, embarrassment) are more common among current smokers, immigrants, and those out of the workforce. Experiencing multiple specific lung cancer symptoms increases help-seeking, suggesting symptom combinations can trigger appraisal and action, particularly among smokers for whom cough may be normalized. The results underscore how smoking-related stigma, fear, and guilt may deter timely presentation, contributing to persistent inequalities in lung cancer outcomes. These insights are highly relevant for primary care and public health: messaging should reassure that consulting for potential lung cancer symptoms is not a waste of GP time, and tailored outreach should focus on vulnerable groups (current smokers, immigrants, unemployed/out of workforce) to reduce delays and improve timely diagnosis.

Healthcare-seeking for lung cancer symptoms in the general population ranged roughly from one-third to just over half, increasing when additional specific symptoms were present. Among those not consulting, nearly two-thirds reported at least one barrier; the most frequent were being too busy and worrying about wasting the doctor’s time, with worry about what the doctor might find also common for cough and dyspnea. Social inequalities were evident: current smokers were about three times more likely to worry about what the doctor might find and five times more likely to feel embarrassed (for shortness of breath); individuals out of the workforce and immigrants also showed elevated emotional barriers, whereas older age and non-working status were associated with fewer practical (time) barriers. The study suggests interventions should target vulnerable groups with unprejudiced, supportive communication, address stigma, and integrate support across healthcare and community services to facilitate timely GP contact. Future research should examine organizational and relational barriers, the interplay of health literacy, socioeconomic factors, and smoking history, and evaluate individualized, community-based interventions to improve help-seeking and diagnostic pathways.

Potential selection bias may exist if individuals with more symptoms or GP contacts were more (or less) likely to participate; overall response was 52.2%. Recall bias is possible despite a 4-week recall window, potentially underestimating symptom prevalence, overestimating GP contact, and underestimating barriers. Web-based data collection could reduce participation among elderly or digitally limited individuals, though a telephone option was offered. Some subgroups (e.g., immigrants/descendants, haemoptysis reporters) were small, limiting precision and preventing reporting of barrier data for haemoptysis due to Danish disclosure rules. Data were collected in 2012; while overall healthcare-seeking patterns may have evolved, structural inequalities and smoking-related stigma likely persist, affecting generalizability over time. Residual confounding cannot be excluded despite multivariable adjustment.