Attitudes towards lung cancer screening in a population sample

Abstract Background Lung cancer is the leading cause of cancer death worldwide. Routine UK lung cancer screening is not yet available, thus understanding barriers to participation in lung screening could help maximize effectiveness if introduced. Methods Population‐based survey of 1007 adults aged 16 and over in Wales using random quota sampling. Computer‐assisted face‐to‐face interviews included demographic variables (age, gender, smoking, social group), four lung cancer belief statements and three lung screening attitudinal items. Determinants of lung screening attitudes were examined using multivariable regression adjusted for age, gender, social group and previous exposure to lung campaign messages. Results Avoidance of lung screening due to fear of what might be found was statistically significantly associated with negative lung cancer beliefs including fatalism (aOR = 8.8, 95% CI = 5.6‐13.9, P ≤ 0.001), low perceived value of symptomatic presentation (aOR = 2.4, 95% CI = 1.5‐3.9, P ≤ 0.001) and low treatment efficacy (aOR = 0.3, CI = 0.2‐0.7, P ≤ 0.01). Low perceived effectiveness of lung screening was significantly associated with fatalism (aOR = 6.4, 95% CI = 3.5‐11.7, P ≤ 0.001), low perceived value of symptom presentation (aOR = 4.9, 95% CI = 2.7‐8.9, P ≤ 0.001) and low treatment efficacy (aOR = 0.1, 95% CI = 0.1‐0.3, P ≤ 0.001). In contrast, respondents who thought lung screening could reduce cancer deaths had positive beliefs about lung cancer (aOR = 0.4, 95% CI = 0.2‐0.7, P ≤ 0.001) and its treatment (aOR = 6.1, 95% CI = 3.0‐12.6, P ≤ 0.001). Conclusion People with negative beliefs about lung cancer may be more likely to avoid lung screening. Alongside the introduction of effective early detection strategies, interventions are needed to modify public perceptions of lung cancer, particularly for fatalism.


| INTRODUC TI ON
Lung cancer is the leading cause of cancer-related death worldwide. 1 Five-year cancer survival rapidly decreases the later lung cancer is diagnosed, due to limited treatments options. In the UK, when lung cancer is diagnosed at the earliest stage (Stage I), 56% of patients can expect to survive for one year or more, in comparison with 14% of patients diagnosed at the most advanced stage (Stage IV). 2 With 78% of UK non-small-cell lung cancer cases diagnosed in the later stages of disease (Stage III or IV), 2 there is a need to explore strategies to diagnose lung cancer earlier.
Currently, diagnostic testing for suspected lung cancer in the UK requires symptomatic patients to present to a healthcare professional for referral for further investigation. This approach relies on the patient and healthcare professional accurately appraising symptoms, which may be problematic for early diagnosis of lung cancer. 3 Lung cancer symptoms are hard to detect in the early stages, due to misattribution to smoking habit, comorbidities or other benign causes. 4,5 Evidence suggests that low-dose computed tomography (CT) screening is effective in detecting early-stage lung cancer. 6 The US National Lung Screening Trial reported a 20% reduction in lung cancer mortality 6 and is currently the standard of care in the United States. 7 Although not routinely available in the UK, trials are ongoing across Europe to assess the effectiveness of CT lung screening among high-risk groups. [8][9][10][11] In the event that lung screening for high-risk populations is introduced routinely in the UK, it is important to understand the barriers to participation and develop interventions to encourage those who are eligible to engage in lung screening in order to optimize its impact and maintain cost-effectiveness.
Previous studies of attitudes towards lung cancer screening suggest that smokers from socio-economically deprived groups place lower value on the benefits of lung cancer screening, hold fatalistic beliefs about lung cancer as an untreatable disease or report stigma as a barrier to screening participation. 12,13 In addition, emotional barriers such as fear of lung cancer 14 and the belief that the lungs are an untreatable organ 12,15 were reported to deter participation in lung screening trials. However, these studies have been restricted to samples of people over the age of 40. 12,13,[15][16][17][18]  Pre-campaign survey data were collected from a total of 1007 adults. The number of people who declined to participate was not recorded, thus the characteristics of survey decliners are unknown.

| Design/procedure
The survey used random quota sampling based on neighbourhoods classified according to census characteristics. The Omnibus sample is designed to be representative of the adult population resident in Wales aged 16 and over, with Lower Layer Super Output Area (LSOA) as the unit of sampling. Sixty-nine interviewing points throughout Wales were selected with probability proportional to resident population, after stratification by local authority and social group based on occupation. Social group was recorded in four categories using the National Readership Survey grades, based on the occupation of the household's chief income earner: AB (higher and intermediate managerial, administrative and professional), C1 (supervisory, clerical and junior managerial, administrative and professional), C2 (skilled manual workers) and DE (semiskilled and unskilled manual workers, state pensioners, casual and lowest grade workers, and unemployed with state benefits only). Categories were combined to cluster participants by social group: ABC1 participants were considered high socio-economic status, and the C2DE participants were considered low socio-economic status.
Within each sampling point, quota sample controls of age and social group within gender were set for the selection of respondents.
Quotas were set to reflect the individual demographic profile of each selected point. A fresh sample of interviewing locations and individuals was selected for each survey, and no more than one person per household was interviewed. Respondents completed a computerassisted interview in the presence of a trained interviewer. Data were weighted by age group within gender within local authority grouping, so that the sample profiles matched those of people aged 16 years and over in Wales derived from the 2011 Census.

| Measures
Survey measures included demographic characteristics (age, gender, social group), smoking history (smoke up to 20 cigarettes a day, smoke 20 or more cigarettes a day, used to smoke, never smoked), beliefs about lung cancer and attitudes towards lung cancer screening. Prior exposure to lung campaign messages was measured by the following questions: "Have you seen, heard or read any adverts, publicity or other types of information in the last couple of months which focused on the subject of lung cancer?" Response options were "yes," "no" and "don't know/can't remember." A brief description of lung screening so worried about what might be found at lung cancer screening that I would prefer not to go"; "I don't think there is any point going for lung cancer screening because it won't affect the outcome"; and "Lung screening could reduce my chances of dying from cancer." Response options were strongly agree, agree, disagree and strongly disagree. "Don't know" responses were recorded. Responses were recoded for analysis purposes, with strongly agree and agree combined to create "agree," disagree and strongly disagree combined to create "disagree." "Don't know" responses were counted as missing.

| Statistical analysis
Descriptive statistics were used to summarize the demographic characteristics of the sample and to assess missing and "don't know" responses. Chi-square univariable tests were used to examine the influence of smoking history, age, gender, social group and lung cancer beliefs on endorsement of attitudes to individual lung cancer screening items. Multivariable regression modelling was carried out to examine the influence of smoking history and lung cancer beliefs on lung cancer screening attitudes (individual items), adjusting for age, gender, social group combined and prior exposure to lung cancer messages. The significance level was set at P < 0.01 to adjust for multiple testing. To account for nonrepresentativeness, a weight was applied to the data based on age and gender within local authority in Wales.
Most of the sample had never smoked (n = 433, 43%), 286 (28%) used to smoke, 259 (26%) currently smoked up to 20 cigarettes a day and 28 (3%) currently smoked more than 20 cigarettes a day (see  Table 1). For univariable and multivariable analysis purposes, "smoke up to 20 a day" and "smoke over 20 a day" were combined to create a "currently smoke" category.

| Avoidance of lung screening
Fifteen per cent (n = 144) of the sample endorsed avoidance of lung screening due to fear of what might be found (Figure 1). Avoidance of lung screening was statistically significantly associated with fatalism (P ≤ 0.001), low perceived value of symptom presentation (P ≤ 0.001), having negative views about treatment (P ≤ 0.001) and worry about wasting the doctor's time (P ≤ 0.001) (see Table 2).
Associations between lung screening attitudes and age, gender, social group, smoking status and exposure to lung messages were not statistically significant.

| Low perceived effectiveness of lung screening
A total of 8% (n = 78) endorsed low perceived effectiveness of lung screening (see Figure 1). Low perceived effectiveness of lung screening was statistically significantly associated with fatalism (P ≤ 0.001), low perceived value of symptom presentation (P ≤ 0.001), having negative views about treatment (P ≤ 0.001) and worry about wasting the doctor's time (P ≤ 0.001) (see Table 3). Effects of age, gender, smoking status, social group and exposure to lung messages were not statistically significant.

| Lung screening to reduce mortality
Ninety per cent (n = 859) of the sample agreed that lung screening could reduce chances of lung cancer death ( Figure 1). Agreeing that lung screening could reduce chances of dying from cancer was associated with positive lung cancer beliefs reflecting lack of fatalism (P ≤ 0.001) and having positive views about treatment (P ≤ 0.001) (see Table 4). There were no statistically significant effects of any demographic variables.

| Logistic regression
Regression analysis was completed using the weighted data, adjusting for age, gender, social group and previous exposure to lung messages. to being worried about what might be found (see Table 5

| Low perceived effectiveness of lung screening
Negative lung cancer beliefs including fatalism (aOR = 6.4 CI = 3.5-11.7, P ≤ 0.001), low perceived value of symptom presentation (aOR = 4.9, CI = 2.7-8.9, P ≤ 0.001) and having negative views about treatment (aOR = 0.1, CI = 0.1-0.3, P ≤ 0.001) showed a statistically significant association with low perceived effectiveness of lung screening (see Table 5). Smoking status and being worried about wasting the doctor's time were not statistically significantly associated with attitudes towards the efficacy of screening.  and participation in screening. 20,22 The current study suggests that negative beliefs about lung cancer were associated with lung screening avoidance, particularly fatalism, suggesting that those who decline screening would prefer not to know if they have lung cancer, potentially due to fear of treatment and lung cancer death. Our findings are in line with previous research highlighting fear of lung cancer [12][13][14] and fatalism, including beliefs about the treatment for lung cancer 12,15 as barriers to participation in lung cancer screening. It is likely that avoidance of lung screening and negative beliefs may reflect lung cancer stigma, 23 possibly due to the relationship between lung cancer and smoking, and poor lung cancer outcomes.

| D ISCUSS I ON
The absence of an observed association between smoking status and lung cancer screening attitudes in our study contradicts the findings of previous studies, which have highlighted more negative screening attitudes among current smokers. 12,24 In addition, former smokers have been shown to be over-represented in lung cancer screening trials. 25,26 Our contradictory findings are likely to reflect TA B L E 3 Frequencies and univariate analysis for lung screening attitude "I don't think there is any point going for lung cancer screening because it won't affect the outcome"