Sample
Participants were on average 39.37 years old (Md = 39, Mo = 36, SD = 5.88, range: 30-49). 82.4% of participants indicated to have the Swiss nationality. Overall, 59.9% of participants hailed from the German-speaking region of Switzerland, 29.8% from the French-speaking region, and 10.2% from the Italian-speaking region. These numbers were reflected by participants’ choice of survey language (German: 60.1%; French: 29.8%; Italian: 10%). Cantons Zurich (14.1%), Berne (11.2%), and Vaud (11.1%) were most represented. In terms of screening availability, 52.18% of the participants lived in a canton which offered a systematic screening program at the time of the survey, while 47.82% lived in a canton offering opportunistic screening.
Most participants were married or in a stable relationship (63.8%). 21.7% of the participants were single, 14.5% divorced, separated, or widowed. With regard to their educational background, 8.3% of the participants indicated to have a middle school degree, 62% a professional or high school degree, 15.4% a degree from a university of applied sciences, and 14.2% a degree from a (polytechnic) university. Of all participants, 0.2% indicated not holding any degree. The majority of participants were (self)employed (57.6%). 25.4% of participants were homemakers, 8.8% were unemployed (of which 7.6% in search of employment), 0.7% were students, and finally 0.4% were pensioners.
Participants evaluated their overall health as relatively good: On a scale ranging from 1 (“very poor”) to 5 (“very good”), participants scored an average of 3.74 (
SD = .88, range: 1-5). As expected, having been diagnosed with a chronic disease significantly impacted self-perceived health (F(1, 892) = 129.39,
p ≤ .001, η
2 = .13). Furthermore, 27% of all participants indicated to have had a mammogram in the past. Lastly, 58 participants had been previously diagnosed with breast cancer (2.8%) or with a genetic predisposition to get this disease (3.5%). These participants were removed from the sample for subsequent analyses. See Table
1 for an overview of the sample description.
Table 1
Sample description
Age | – | – | 39.37 | 5.88 | 30-49 |
Marital status | | | – | – | – |
Married/stable relationship | 586 | 63.8 | | | |
Single | 199 | 21.7 | | | |
Divorced/separated/widowed | 133 | 14.5 | | | |
Education | | | – | – | – |
None | 2 | .02 | | | |
Middle school | 76 | 8.3 | | | |
Professional or high school | 569 | 62 | | | |
University of applied sciences | 141 | 15.4 | | | |
University | 130 | 14.1 | | | |
Occupation | | | – | – | – |
(Self)employed | 594 | 57.6 | | | |
Homemaker | 233 | 25.4 | | | |
Unemployed | 81 | 8.8 | | | |
Student | 6 | .7 | | | |
Pensioner | 4 | .4 | | | |
Swiss nationality | | | – | – | – |
Yes | 756 | 82.4 | | | |
No | 162 | 17.6 | | | |
Swiss language region | | | – | – | – |
Swiss-German | 550 | 59.9 | | | |
Swiss-French | 274 | 29.8 | | | |
Swiss-Italian | 94 | 10.2 | | | |
Systematic screening program | | | – | – | – |
Yes | 479 | 52.2 | | | |
No | 439 | 47.8 | | | |
Health status | – | – | 3.74 | .88 | 1-5 |
Genetic predisposition (e.g., BRCA1) | 32 | 3.5 | – | – | – |
Breast cancer diagnosis | 26 | 2.8 | | | |
Past mammogram (N = 860) | 232 | 27 | – | – | – |
30-34 | 23 | 10.1 | | | |
35-39 | 48 | 22.1 | | | |
40-44 | 59 | 29.4 | | | |
45-49 | 102 | 47.7 | | | |
Descriptive statistics
Participants were relatively knowledgeable about breast cancer and mammography screening. On average, women answered 70.8% of the knowledge questions correctly, (Md = 72, Mo = 72, SD = 10.48), with a range between 28% and 96% corresponding to a minimum of seven and a maximum of 24 correct answers. In terms of general knowledge, women answered on average 66.9% of the questions correctly (Md = 66.7, Mo = 67, SD = 14.42, range: 25%-100%). The scores for curability knowledge were higher, with an average score of 78.1% (Md = 75.0, Mo = 88, SD = 17.75, range: 0%-100%). Lastly, with regard to Swiss program knowledge, women scored on average 68.2% (Md = 80.0, Mo = 80, SD = 21.26, range: 0%-100%). Close examination of the individual items revealed that while participants performed well on most questions, notably some questions elicited more erroneous than correct responses. These questions pertained in particular to the risk factors for developing breast cancer. Overall, participants displayed a lack of knowledge of known risk factors, such as being overweight, living in a Western country, and age. More specifically: 54.1% of participants did not know that being overweight is a risk factor for breast cancer; 56.3% of women were not aware that breast cancer is more common in Switzerland than in Africa and Asia. Moreover, 52.9% of participants answered (erroneously) that women without known risk factors rarely develop breast cancer. Further, 57.3% did not know that breast cancer is more prevalent under women aged 65 than 40. This latter finding was emphasized by the result that 42.1% of participants thought that women over 70 rarely develop breast cancer. In addition, 57.6% of women stated that mammograms are pain free. Lastly, participants were unaware of the age-thresholds for organized screening in Switzerland: The majority of participants (64.1%) answered that Swiss programs invite women from the age of 40 onward.
At the same time, participants appeared fearful of breast cancer and over-estimated their risk of getting breast cancer. On average, participants scored a 23.41 on the breast cancer fear scale, with scores ranging between eight and 40 (Md = 24.0, Mo = 24, SD = 7.64). Overall, 13.7% of the participants were characterized by low (score: ≤ 15), 30.2% by moderate (score: 16-23), and 56% by high levels of fear (≥24). When asked about their personal risk of developing breast cancer, participants on average estimated this at 20.45% (Md = 20.0, Mo = 10, SD = 19.06, range: 0-100), equivalent to a chance of 1 in 4.89. Participants’ self-perceived susceptibility to get breast cancer, however, was relatively low with a mean score of 2.24 (Md = 2.20, Mo = 1, SD = .82, range: 1-5).
Participants also appeared to be highly involved with the topic of breast cancer screening (M = 5.40, Md = 5.5, Mo = 7, SD = 1.12, range: 1-7). This implies that these women deem mammography screening of very high relevance to them personally and thus seem to have a very positive attitude towards mammography screening – despite their young age.
Finally, when asked to respond to the statement “I intend to go for mammography screening in the near future”, 87 participants (N = 860) indicated to not have thought about that yet. Among those who responded to the statement, 44.2% indicated to (strongly) disagree, 21.2% to neither agree nor disagree; and 34.5% to (strongly) agree. On average, these respondents scored a 2.90 (Md = 3.0, Mo = 2, SD = 1.33, range: 1-5). When asked at what age they intend to have their (next) screening mammogram (N = 773), 272 women (35.2%) answered they do not plan to have a mammogram at all. The remaining women, indicated an average age for a (next) mammogram of 45.50, with a mode of 50 (Md = 45.0, SD = 5.62, range: 31-80). In total, 56.7% of respondents indicated the intention to have (another) screening mammogram below the age of 50, thus below the lower age threshold for systematic screening in Switzerland (9.6% indicated an age below 40; 24.4% said 40-44; 22.8% answered 45-49). In the Swiss-Italian region the indicated average age was 44.10 (Md = 44.5, Mo = 40, SD = 6.89, range: 32-80), in the Swiss-German region 45.62 (Md = 46.0, Mo = 50, SD = 5.75, range: 32-60), and in the Swiss-French region 45.84 (Md = 46.5, Mo = 50, SD = 4.85, range: 31-55).
Relationships between knowledge, attitudes, risk perceptions, and intentions
In a first step, the strength of the relationships between the variables of interest was tested. No meaningful associations were found between breast cancer knowledge and any of the other variables, including the main outcome variable ‘mammography intentions’. This applied to overall knowledge scores as well as to the three subscales. Bivariate correlation analyses revealed a significant, weak relationship between young women’s fear of breast cancer and their risk perceptions, as such that higher risk perceptions were associated with higher levels of fear (
r(858)
= .24,
p ≤ .001). Perceived susceptibility to get breast cancer, as well, was strongly correlated with breast cancer risk perceptions and fear (
r(858)
= .54,
p ≤ .001); (
r(858)
= .49,
p ≤ .001), respectively). Breast cancer fear and susceptibility, in turn, were moderately correlated with participants’ ego-involvement with breast cancer screening (
r(858)
= .32,
p ≤ .001); (
r(858)
= .30,
p ≤ .001). That is, participants who perceived themselves as fearful and highly susceptible to breast cancer judged mammography as more relevant to them personally and, thus, had a more positive attitude towards screening. Finally, the intention to go for screening in the near future was strongly associated with participants’ ego-involvement (r(771) = .45,
p ≤ .001), their perceived susceptibility (r(771) = .29,
p ≤ .001), and fear (r(771) = .29,
p ≤ .001). The results show an association between participants’ involvement, fear, and perceived susceptibilityand their intention to engage in screening in the near future. In Table
2, an overview of the associations between the different variables can be found.
Table 2
Pearson’s correlation coefficient for the variables studied
Knowledge | – | | | | |
Fear | -.02 | – | | | |
Perceived risk | .01 | .24***
| – | | |
Perceived susceptibility | .04 | .49***
| .54***
| – | |
Ego-involvement | .07 | .32***
| .19 | .30***
| – |
Mammography intention | .02 | .29***
| .19 | .29***
| .45***
|
Subsequently, a hierarchical multiple linear regression analysis was used to test if knowledge, perceived breast cancer risk and susceptibility, breast cancer fear, and ego-involvement can indeed predict participants' intentions to go for mammography screening. Participants’ age, educational levels, and geographical location were controlled for as possible confounder variables. The results of the regression analysis demonstrate that three main variables of interest, as well as two control variables, significantly predict mammography intention, namely: ego-involvement (
β = .34,
p ≤ .001), breast cancer fear (
β = .08,
p ≤ .05), perceived susceptibility (
β = .10,
p ≤ .05), geographical location (Swiss-French group:
β = .15,
p ≤ .001; Swiss-Italian group:
β = .26,
p ≤ .001), and age (
β = .11,
p ≤ .001). Together, these variables explain 32% of the variance (
R
2
= .32, F(9, 772) = 38.78,
p ≤ .001). Breast cancer knowledge, risk perceptions, and educational status did not significantly contribute to the final model (see Table
3).
Table 3
Coefficients variables resulting from hierarchical multiple linear regression analysis (final model)
Constant | -1.27 | .45 | – | -2.85 | .008 | – | – | – |
Covariates |
Age | .025 | .01 | .11***
| 3.69 | .001 | – | – | – |
Education | .075 | .05 | .05 | 1.53 | .174 | – | – | – |
Region | | | | | | – | – | – |
Swiss-Italian | 1.12 | .14 | .26***
| 8.14 | .000 |
Swiss-French | .43 | .09 | .15***
| 4.65 | .000 |
Predictors |
Ego-involvement | .40 | .04 | .34***
| 10.2 | .000 | – | – | – |
Knowledge | -.27 | .39 | -.02 | -.71 | .389 | – | – | – |
Fear | .02 | .01 | .08*
| 2.34 | .013 | – | – | – |
Susceptibility | .15 | .06 | .10*
| 2.36 | .015 | – | – | – |
Risk | .00 | .00 | .03 | .94 | .388 | – | – | – |
Model | – | – | – | – | – | .56 | .32 | .31 |