Factors related to participation in colorectal cancer screening in a Southern European region: a prevalence survey.

Introduction
Neoplasms and cardiovascular diseases are the main causes of death in Spain [1]. Among neoplasms, colorectal cancer is the second leading cause of cancer-related death in men and the third leading cause of cancer-related death in women in the Community of Madrid (CM) [2].
The aim of population-based colorectal cancer screening programmes is to reduce both mortality through early diagnosis and incidence through the detection of preneoplastic lesions [3]. For these programmes to have an impact on mortality, it is recommended that at least 60% of the target population be reached [4]. However, these recommendations may change depending on the results of several ongoing clinical trials [5, 6].
Spain is a European country with a population-based colorectal cancer screening programme. The programme was included in the National Health System’s common portfolio of services in 2014 (Order SSI/2065/2014) and offers a faecal occult blood test (FOBT) every 2 years for men and women aged 50–69 years [7]. Since then, the screening programme has been implemented in each of the Autonomous Communities. The Cancer Strategy of the National Health System, updated in 2021, sets a minimum participation target of 65% [8]. In the European context, the goal was to reach 90% participation by 2025 [9].
The implementation of the CM Colon and Rectal Cancer Screening Programme (PREVECOLON) was consolidated between 2018 and 2019, with the aim of covering the entire target population in 2020 [10]. The test used in the Community of Madrid is the immunochemical test, which shows sensitivity and specificity values that vary slightly depending on the positivity threshold. For 20 µg Hb/g it has a sensitivity of ∼74%–81% and a specificity of 94%–95% [11].
The programme’s monitoring indicators show that in 2023, 40% of the target population was covered and, 5.0% of the test results were positive. Among those who underwent a colonoscopy, 22.9% had a normal colonoscopy [12].
The European Commission, within the framework of Europe’s Beating Cancer Plan [9], has highlighted the importance of addressing existing inequalities throughout the entire process from prevention to treatment of cancer. As one of the main determinants of social inequalities in health [13], gender needs to be accounted for when implementing public health policies to increase the effectiveness of interventions [14]. To achieve gender equality, the different implications of a programme or intervention should be assessed to ensure that it is adapted to the needs of men and women so that both men and women can benefit equally [15].
Two recent systematic reviews [16, 17] examined the sociodemographic and lifestyle factors associated with the lack of participation in colorectal cancer screening programmes. According to a review by Mosquera et al. [17], some studies reported greater participation of women in screening programs, whereas others reported no differences in participation by sex. Unanue-Arza et al.’s review [16] revealed no differences, as did studies with data from Spanish health surveys [18, 19]. The available evidence is therefore not conclusive regarding sex and adherence to colorectal cancer screening. Higher compliance has also been reported in people over 60 years of age [20] and those with higher socioeconomic status [17, 21]. Other factors associated with nonparticipation were being born outside Spain, being obese, smoking, having a sedentary lifestyle [16, 20], and having less contact with health services [18].
Not all studies of participation in colorectal cancer screening reported data disaggregated by sex, and few research teams accounted for sex and performed stratified analyses [16, 17]. Studies that analyse the effects of sex on other variables show differences in risk factors and screening practices between men and women [16, 22]. In addition, some of the results of these studies are contradictory, which makes it difficult to draw conclusions.
The most recent information available on participation in colorectal cancer screening, including an FOBT, in Spain comes from health surveys conducted in 2020 [23]. Although general data from the population-based screening programme in Madrid are available for the past 4 years, there is no information on factors related to compliance. Therefore, updated information on programme participation and its associated determinants are needed. Thus, the objective of the present study was to analyse participation in colorectal cancer screening among the population of the CM from 2020 to 2023, as well as the demographic, socioeconomic, lifestyle and health-related factors associated with participation status.

Methods

Data source and study population
A cross-sectional analytical study was conducted. The source of information was the annual telephone surveys of the Noncommunicable Disease Risk-Factor Surveillance System (SIVFRENT) of the CM. This surveillance system for behavioural disease risk factors and preventive practices is similar to the CDC’s Behavioural Risk Factor Surveillance System (BRFSS). SIVFRENT-A targets the population aged 18–64 years, and SIVFRENT-M targets the population aged 65–79 years.
The sampling frame for these surveys was the CIBELES Population Information System, which includes the population with an individual health card. The annual sample was stratified by sex, age, and geographical area, and within each stratum individuals were randomly selected for interview.
For this study, surveys of the population aged 50–69 years were selected, the target population for screening with an FOBT in CM, and information from 2020 to 2023 was analysed. In 2023, CIBELES database included 1 796 319 individuals aged 50–69 years (52.4% women) and the annual sampling fraction was 0.06%. In previous years, both the number of eligible individuals and the proportion selected were similar.

Variables
The main outcome analysed was whether participants underwent an FOBT for screening purposes at any time in their life. People who had an FOBT were compared with people who had never had the test.
The survey year and the following demographic and socioeconomic variables were analysed: sex, age group, country of birth, educational level, employment status, social class, and marital status. In addition, other variables related to lifestyle were analysed: tobacco use, alcohol use, body mass index (BMI), physical activity at work and during leisure time and fruit and vegetable consumption.

Statistical analysis
A descriptive analysis of the study population was performed. Descriptive analyses of all the variables by outcome were also performed. To calculate the prevalence ratios (PR) and the corresponding 95% confidence intervals (95% CI), a weighting factor was applied, considering the area (Madrid city, metropolitan area or other areas), sex (male or female), and age (50–64 or 65–69) according to the most recent data available. To assess the associations between FOBT compliance and the different independent variables, Poisson regression models with robust variance were used to calculate the crude (cPR) and adjusted (aPR) prevalence ratios for demographic and socioeconomic variables and the corresponding 95% CIs. Interactions between sex and demographic, socioeconomic and lifestyle-related variables were assessed. All analyses were stratified by sex.
Stata 18.0 (StataCorp, College Station, TX, USA) was used for the statistical analysis.

Results

Sample characteristics
The response rates for SIVFRENT-A and SIVFRENT-M ranged from 52.4% in 2023 to 80.9% in 2021. Of the 4116 surveys, 303 were removed: 42 because they did not provide information on test performance and 261 because they were not performed for preventive purposes. There were no sex differences in the reason for testing.
Finally, information from 3813 participants was analysed. Of these, 52.1% were women, and 38.4% were aged between 65 and 69 years.

Table 1 shows the global and sex-stratified distributions in terms of demographic, socioeconomic, and health behaviour characteristics. Notably, the percentages of those born outside Spain and those who were unemployed were greater among women. On the other hand, a greater proportion of men had active employment and were married or living with a partner.

PRs for FOBT compliance

Tables 2 and 3 show the PRs for FOBT compliance for men and women, respectively, together with the cPR. The variables for which no significant differences in FOBT compliance were observed are not shown in the tables.
The overall prevalence of FOBT screening compliance in the population aged 50–69 years was 55.3% (95% CI: 53.6–57.0). Compliance was higher in men (57.8%) than in women (53.1%) (P < .05). For both sexes, 2023 was the year in which the highest percentages were found. In 2021, the greatest differences were observed between men and women.
Considering only those who had an FOBT in the last 2 years, the prevalence of global compliance was 48.5% (95% CI: 46.8–50.2) and was higher in men (50.3%; 95% CI: 47.8–52.8) than in women (46.9%; 95% CI 44.6–49.3), although these differences were not significant (P = .058).

Factors associated with participation
In the crude analyses, male sex was associated with higher FOBT performance (cPR: 1.09; 95% CI: 1.03–1.15). Factors associated with lower compliance with FOBT among men included marital status (being separated/divorced or single), obesity, and smoking. Higher compliance was observed among those who engaged in leisure-time physical activity, were aged 65–69, and were retired (Table 2). Among women, lower FOBT compliance was associated with being underweight, being born outside Spain, having a marital status other than married or cohabiting, and smoking. Factors linked to higher compliance included regular fruit consumption, being aged 60 or older, engaging in physical activity during leisure time and at work, and being retired (Table 3).

Table 4 shows the results of the multivariate analysis for men and women. The assessment of the interaction between sex and other explanatory variables revealed relevant differences that supported the use of stratified analysis, although the interaction P values were greater than the significance threshold.
Among men, being separated or divorced (aPR: 0.80; 95% CI: 0.67–0.94), compared with being married or cohabiting, and being aged 50–59 years (aPR: 0.92; 95% CI: 0.85–0.99), compared with those aged 60–69 years, were associated with a lower likelihood of undergoing an FOBT. Higher compliance was observed among those who engaged in leisure-time physical activity (aPR: 1.13; 95% CI: 1.00–1.27).
Among women, being underweight (aPR: 0.63; 95% CI: 0.40–0.97) compared with those of normal weight; being widowed (aPR: 0.78; 95% CI: 0.66–0.93) compared with those who were married or cohabiting; being born outside Spain (aPR: 0.82; 95% CI: 0.72–0.95); being a current smoker (aPR: 0.84; 95% CI: 0.75–0.95); and being under 60 years of age (aPR: 0.86; 95% CI: 0.79–0.94) compared with those aged 60–69 years were all associated with lower compliance. Among lifestyle variables, regular fruit and vegetable consumption was associated with increased compliance (aPR: 1.52; 95% CI: 1.12–2.07 for ≥5 fruit servings per day, and aPR: 1.35; 95% CI: 1.01–1.82 for 3–4 servings per day, compared with no consumption). Physical activity at work (aPR: 1.15; 95% CI: 1.04–1.26) and engagement in leisure-time physical activity (aPR: 1.13; 95% CI: 1.01–1.27) were also associated with increased compliance.

Discussion
The results of this study show that the proportion of men who reported having undergone an FOBT for colorectal cancer screening in the CM is greater than that of women. In 2023, reported participation in the programme in both sexes reached the minimum standard required for the programme to be cost-effective [24]. Factors associated with FOBT compliance differed by sex. In men, being separated or divorced, as well as being under 60 years were associated with lower compliance. For women, being underweight, widowed, born outside Spain, a current smoker and younger than 60 years of age were associated with lower compliance. No differences in participation were found regarding socioeconomic variables, such as occupational social class or educational level.
When considering only those who reported having undergone the test in 2022 and 2023, we found similar figures to those reported by the Madrid Health Service in its annual reports [12, 25]. The estimated participation in PREVECOLON programme was 49% in 2022 and 40% in 2023. However, in our study, all those who underwent the test were analysed together regardless of the date of the test. This approach was chosen to identify factors associated with the lack of participation, i.e. to target the population that is being left aside from the screening programme. In addition, since our data source was a survey, we did not have information on the results of the test or the performance of a subsequent colonoscopy, so it may be incorrect to consider individuals who were tested >2 years ago as not being screened.
Within the period analysed, a general trend towards an increase in compliance was observed in successive years, although this trend was more notable in 2023. This is likely related to the different strategies used throughout the years to increase coverage and reach the nonparticipating population [12]. The minimum 60% participation proposed for the programme to be cost-effective was reached only in 2023 [24, 26], and this figure is far from the 90% target proposed by the European Commission [9]. Therefore, it is necessary to implement strategies to increase coverage, especially in the groups found to have low compliance.
Previous studies carried out in Spain using data from participants in the population-based colorectal cancer screening programme [21, 22], as well as interviews conducted within this group [27, 28], have shown greater participation among women. However, results from health surveys in Spain do not support this trend [18, 19, 23]. When survey data are disaggregated by region, in autonomous communities that began screening programs before 2010 [23], there is greater participation in women, whereas in those that started screening programs later, there is no clear pattern. In the review by Mosquera et al. [17], which included studies from other European countries as well as Canada, the United States and Australia, overall more women than men underwent screening with FOBT, which differs from the findings of our study. In another study conducted in Finland on factors associated with non-participation in the programme between 2004 and 2016, lower participation was found among men compared with women [29]. By contrast, another study conducted in Czechia based on its national health survey found lower participation among women [30]. Although overall data from screening programmes appear to indicate higher participation among women, the fact that men exhibit a greater incidence and mortality from colorectal cancer [31] may be contributing to increased awareness and participation among men.
In terms of age, in our study, participation rates were higher among both men and women aged 60–69 years compared with those aged 50–59 years. Greater participation of older people has also been reported in other studies carried out in Spain [21, 22] and other European countries [29, 30]. A lower perception of risk among younger individuals or a lack of time due to active employment [32] or caregiving responsibilities could explain the lower participation observed in this age group.
In this study, we also investigated the effects of occupational social class and educational level on participation in screening. However, contrary to findings reported several years ago in studies conducted in other Spanish regions [20, 22], no differences in participation were observed according to the socioeconomic variables examined. This reflects one of the priorities of screening programs in the National Health System, namely, to reduce inequalities in access to programmes [8]. According to the variables we considered, this goal is being achieved in the CM.
Some previous studies have also analysed the effect of marital status on participation in colorectal cancer screening programs. In line with our findings, some [32, 33] reported that being married or living with a partner had a greater protective effect in men than in women. In other studies [34, 35], married people, regardless of sex, participated more than single, divorced and widowed people did. However, studies based on the 2017 Spanish National Health Survey [18, 19] revealed no differences in the performance of the test according to marital status.
Several publications have analysed the effects of living as a couple on people’s health behaviour. Notably, a review [36] that analysed, among other things, the concordance in couples’ lifestyles and changes in habits reported certain associations among these factors, although the evidence was limited. Other studies [37–39] have shown that women can influence their partners’ health-related behaviours to a greater extent than men can, which may explain the observed results. In line with findings of other studies [20, 22], healthy behaviours were also associated with greater participation, but in different ways for men and women, suggesting a different profile for each sex. Although both men and women engage in physical activity during their leisure time, only women show concern about their diet and are non-smokers.
Within the limitations of our study, it should be noted that the results are based on data reported by the respondents and not the results recorded by the PREVECOLON Programme. A major strength of our study is that the use of survey data allows us to analyse variables that are not usually collected in population-based programmes, such as socioeconomic and lifestyle-related variables.
This study shows that the colorectal cancer screening programme has achieved its coverage targets in recent years. However, communication strategies should be intensified for individuals younger than 60 years. The factors influencing participation differ between men and women; therefore, a sex-sensitive approach is essential when designing and implementing strategies to increase participation and promote equity.
Screening recommendations should be reinforced among women, individuals born outside Spain, and widowed women. Among men, efforts should focus on increasing uptake among those not living with a partner.
Additionally, programme data should be systematically disaggregated by sex, as women appear to present a greater number of vulnerability factors that may influence their decision not to participate in screening. This approach would help identify gender-specific barriers and guide more equitable interventions.