Factors associated with a positive attitude towards receiving cancer information: a population‐based study in Spain

Abstract

Background

It is necessary to raise a positive attitude towards cancer information to improve disease prevention and control.

Objective

To identify social factors, health characteristics and cancer‐related perceptions and knowledge associated with a positive attitude towards cancer information.

Design

We ran multivariate logistic regression models to analyse population‐based data from OncoBarómetro‐2010, a Spanish representative survey on perceptions and knowledge, related to cancer, conducted among 7938 people aged 18 years or more.

Measurement

Attitudes towards cancer information. A positive attitude includes feeling motivated to keep informed, to have screening tests or to change lifestyle. A negative attitude includes feeling indifference, concern, frustration or fear.

Results

38.3% of the studied population reported having received information related to cancer (within the last 6 months). Among those, 31.5% expressed a negative attitude towards cancer information. People more likely to have a positive attitude towards cancer information (reference category: negative attitude) were people aged 35–74 (ref: aged 18–34) (P < 0.001) and cancer survivors (ref: those who had not had cancer) (OR: 3.05; 95% CI: 1.73–5.38). The likelihood of a positive attitude increased with the level of education (P < 0.001). The variables negatively associated with a positive attitude towards cancer information were poor self‐rated health status (ref: fair) (OR: 0.63; 95% CI: 0.42–0.95) and high self‐perceived risk of developing cancer (ref: low) (OR: 0.75; 95% CI: 0.60–0.92).

Conclusions

These findings have potential to inform programmes designed to promote cancer prevention behaviours. Policies should target population groups with low socio‐economic groups, those with poor self‐rated health and individuals with high self‐perceived risk of cancer. Further, in order to increase knowledge of cancer symptoms, we need to focus on individuals with unhealthy lifestyles.

Introduction

Cancer is a leading cause of mortality worldwide. In 2008, this disease was responsible for 7.6 million deaths, which account for 13% of all global fatalities.1 It is estimated that, in that same year, almost 170 million years of disability‐free life were lost to cancer.2 According to the World Health Organization, over 30% of cancer fatalities could be avoided by modifying or avoiding the leading behavioural, dietary and environmental risk factors associated with the disease.1 In addition, cancer mortality may be greatly reduced through early detection and prompt treatment.3

Thus, as healthy lifestyle promotion and cancer screening are the fundamental pillars of the fight against the disease, effective delivery of health communication plays a key role in cancer prevention efforts. Yet, disseminating accurate information is not enough, it is essential that health messages inspire in the recipient a positive attitude towards both prevention as well as early diagnosis.

Previous research has identified several sociodemographic characteristics and preventive behaviours of individuals receiving cancer information. Their findings indicate that both active search (cancer information seeking) and passive exposure (cancer information scanning) to cancer information are associated with being female, having a high socio‐economic status, leading a healthy lifestyle and a higher adherence to screening recommendations.4, 5, 6 Further, studies examining the role of psychosocial factors on health promotion concluded that preventive behaviours are associated with greater cancer knowledge levels,7, 8, 9 as well as with the ability to successfully cope with any threats perceived as part of health messages.10

Similarly, cancer information avoidance has been linked to cancer fear, perceived cancer severity, fatalism and low response–efficacy beliefs.11 Further, individuals reporting fear of cancer and holding fatalistic beliefs are less likely to adhere to cancer screening guidelines12, 13 to follow health promotion recommendations14 and to seek medical guidance promptly when first noticing a cancer symptom.15

Despite this evidence illustrating the importance of a positive (or negative) attitude towards cancer information on an individual's preventive behaviours,9 empirical research on the specific factors determining such attitudes is scarce. Moreover, we are currently unaware of the Spanish population's attitude towards cancer information seeking and/or scanning and the factors contributing to such attitudes. Therefore, the aim of this work was to identify factors associated with a positive vs. a negative attitude towards cancer information among the population of Spain.

Methods

Study participants

For this population‐based study, we used data from OncoBarómetro, a national survey on perceptions, knowledge and attitudes related to cancer, conducted among 7938 people aged 18 years or more, living in Spain. This survey was carried out by the Cancer Observatory of the Spanish Association against Cancer (AECC) during the last 2 months of 2010. The sample is representative of the population of the country as a whole as well as of the autonomous regions. Disproportionate stratified sampling was used, and a sample weighting factor was applied to the analysis.

A multistage cluster sampling technique was applied as follows: primary sampling units were selected from the municipalities, and secondary units were selected from the census tracks, using proportional random sampling. The last units (individuals) were selected by sex and age quotas, using random‐route procedure. For an allowable sampling error of 1.12 and a 95% confidence level, the sample size was 7938. The questionnaire was designed by the AECC and the Center for Sociological Research (CIS). The fieldwork, which consisted of face‐to‐face interviews, was done by the CIS during the months of November and December of 2010. The data file is available on the website: http://www.cis.es/cis/opencm/ES/1_encuestas/estudios/ver.jsp?estudio=12124).

Measures

To analyse the differences related to respondents’ attitudes towards cancer information, we included several relevant variables. The dependent variable was derived from the answers to the following questions: ‘Have you received, seen or read information about cancer in the last six months?’ if ‘yes’, participants were then asked ‘How do you react to information about cancer?’ which had five possible answers: ‘feeling motivated to keep informed’; ‘motivated to have screening tests’; ‘motivated to change my lifestyle’; ‘with indifference’; and ‘with concern, frustration or fear’. These answers were dichotomized into either positive attitude (‘feeling motivated to keep informed’, ‘motivated to have screening tests’, ‘motivated to change my lifestyle’) or negative attitude (‘with indifference’ and ‘with concern, frustration or fear’). This is an exploratory measure which could be tested for validity and reliability in future studies.

The independent variables were selected based on their association with behaviours related to cancer.7, 9, 16, 17 They included demographic, socio‐economic and health characteristics, as well as cancer‐related perceptions and knowledge variables. The demographic characteristics included sex, age (grouped into four groups: 18–34, 35–54, 55–74 and 75 or older) and autonomous region of residence. The socio‐economic variable selected was education level classified into four groups (less than primary education, primary, secondary or vocational training, and higher). The health variables considered were self‐rated health status in the last 12 months (grouped into three categories: ‘poor or very poor’, ‘fair’ and ‘good or very good’), self‐perceived lifestyle (‘unhealthy’ vs. ‘healthy’) and having or having ever had cancer (yes/no).

Cancer‐related perceptions and knowledge were assessed with the following variables: (1) self‐perceived risk of developing cancer, coded into three categories (‘very low or low’, ‘high or very high’ and ‘don't know/no answer’); (2) ranking of risk factors in terms of contribution to the development of cancer (diet, sun exposure, family history of cancer, weight, pollution, radiation exposure (X‐ray and other radiation), tobacco, stress, alcohol). Possible answers ranged from 1 to 10, with 10 indicating the highest contribution. Variable values for each risk factor were then dichotomized into low (<7) and high contribution (7–10) to the development of cancer; (3) knowledge of symptoms or signs of cancer (yes/no). Other variables analysed were (1) fear of cancer, evaluated with the question: What disease do you fear the most? Respondents had to choose a single option from a list of seven diseases. Answers to this question were coded into: cancer is the most feared disease (yes/no); (2) perceived severity, evaluated with the question: In Spain, do you believe that cancer is ‘very serious’, ‘serious’ or ‘little serious’ disease? (3) expectations of a definite cancer treatment refer to a near future (10 years) evaluated with the question ‘Do you believe that in 10 years humanity will be able to cure cancer?’ (yes/no).

Statistical analysis

We first examined the characteristics of the studied population according to the different variables selected. To identify the factors associated with a positive vs. a negative attitude towards cancer information, we fitted multivariate logistic regression models. After the univariate analysis, we tested for potential collinearity among the independent variables with the Spearman's correlation coefficient. To reduce collinearity, a Spearman coefficient of a maximum of 0.6 was the cut‐off point for inclusion in the model.18 Next, we performed a stepwise forward analysis. Models fit to the data were compared using the Akaike's information criterion and Schwarz Bayesian information criterion methods. Based on these criteria, the model with the smallest value is considered better. After the independent variables were selected for the model, we examined possible interactions between them. All analyses were performed with weighted data according to the OncoBarometro Survey weights. Analyses were computed using IBM SPSS version 18, New York, USA and Stata version 11, Texas, USA.

Results

Close to two‐fifths (38.3% of the weighted sample) of the studied population reported having received information related to cancer in the last 6 months. A higher proportion of women than men reported receiving cancer information (40.7 vs. 36.8%, respectively); and of the age groups, those between 34 and 54 years of age were also more likely to receive cancer information. Individuals with a higher educational level, those reporting knowing symptoms or signs of the disease, people assigning high importance to certain risk factors to cancer onset, respondents with a high perceived risk for cancer and those who have it or have had it were also more likely to receive cancer information in the last 6 months (Table 1).

Table 1.

Receipt of information in the last 6 months, by sample characteristics

Variables	Received informationa						P b
	Total sample		No		Yes
	n	%	n	%	n	%
Demographic characteristics
Sex
Men	3884	49.1	2455	63.2	1429	36.8	<0.001
Women	4033	50.9	2392	59.3	1641	40.7
Age
18–34 years old	2347	29.7	1409	60.0	938	40.0	<0.001
35–54 years old	2931	37.0	1598	54.5	1333	45.5
55–74 years old	1953	24.7	1304	66.8	649	33.2
75 years old or older	683	8.6	534	78.2	149	21.8
Socioeconomic characteristics
Education level
<Primary	612	7.7	495	80.9	117	19.1	<0.001
Primary	1639	20.7	1197	73.0	442	27.0
Secondary/Vocational training	4148	52.4	2456	59.2	1692	40.8
University	1496	18.9	686	45.9	810	54.1
Don't know/No answer	22	0.3	13	59.1	9	40.9
Health characteristics
Self‐rated health status
Poor/Very poor	575	7.3	370	64.3	205	35.7	<0.001
Fair	1897	23.9	1224	64.5	673	35.5
Good/Very good	5413	68.4	3238	59.8	2175	40.2
Don't know/No answer	31	0.4	14	45.2	17	54.8
Lifestyle
Unhealthy	1212	15.3	736	60.7	476	39.3	0.08
Healthy	6605	83.4	4043	61.2	2562	38.8
Don't know/No answer	100	1.3	67	67.0	33	33.0
Have or Have had cancer
Yes	302	3.8	127	42.1	175	57.9	<0.001
No	7615	96.2	4720	61.9	2895	38.1
Perceptions and knowledge about cancer
Perceived risk of developing cancer
Very low/Low	3134	39.8	1983	63.3	1151	36.7	<0.001
High/Very high	2983	38.0	1611	54.0	1372	46.0
Don't know/No answer	1800	22.2	1253	69.6	547	30.4
Knowledge about symptoms
Yes	4002	50.5	2070	51.7	1932	48.3	<0.001
No	3894	49.2	2764	71.0	1130	29.0
Total	7917	100	4846	61.2	3070	38.3

^aNo answer = 22.

^bStatistical differences between groups by Wald chi‐square test.

Among those reporting receiving cancer information, 31.5% expressed a negative attitude towards cancer information; that is, they reacted with indifference or with concern, frustration or fear. Table 2 describes attitudes towards cancer information by demographic and socio‐economic characteristics, self‐rated health, and individuals’ cancer‐related perceptions and knowledge.

Table 2.

Frequency distribution of attitudes towards cancer information received

Variables	Attitudes towards cancer information						P a
	Subsample		Positive		Negative
	n	%	n	%	n	%
Demographic characteristics
Sex
Women	1341	53.0	940	70.1	401	29.9	0.075
Men	1187	46.9	791	66.6	396	33.4
Age
18–34 years old	787	31.1	498	63.3	289	36.7	<0.001
35–54 years old	1131	44.7	798	70.6	333	29.4
55–74 years old	514	20.3	379	73.7	135	26.3
75 years old or older	97	3.8	57	58.8	40	41.2
Socioeconomic characteristics
Education level
<Primary	79	3.1	39	49.4	40	50.6	<0.001
Primary	353	14.0	229	64.9	124	35.1
Secondary/Vocational training	1424	56.3	947	66.5	477	33.5
University	673	26.6	517	76.8	156	23.2
Health characteristics
Self‐rated health status
Poor/Very poor	155	6.1	90	58.1	65	41.9	0.004
Fair	536	21.2	352	65.7	184	34.3
Good/Very good	1838	72.7	1290	70.2	548	29.8
Lifestyle
Unhealthy	379	15.0	217	57.3	162	42.7	<0.001
Healthy	2149	85.0	1514	70.4	635	29.6
Have or have had cancer
Yes	119	4.7	100	84.0	19	16.0	0.001
No	2409	95.3	1631	67.7	778	32.3
Perceptions and knowledge about cancer
Perceived risk of developing cancer
Low/Very low	981	38.8	709	72.3	272	27.7	0.002
High/Very high	1159	45.8	780	67.3	379	32.7
Don't know/No answer	389	15.4	242	62.2	147	37.8
Knowledge about symptoms
Yes	1622	64.2	1165	71.8	457	28.2	<0.001
No	907	35.9	567	62.5	340	37.5
High importance attributed to cancer risk factors
Diet and nutrition	1440	57.0	1040	72.2	400	27.8	<0.001
Weight	676	26.7	510	75.4	166	24.6	<0.001
Exposure to X‐ray/radiation	1595	63.1	1134	71.2	461	28.8	0.002
Alcohol	1864	73.7	1310	70.3	554	29.7	<0.001
Total	2528	100	1731	68.5	797	31.5

^aStatistical differences between groups by Wald chi‐square test.

Multivariate analysis revealed how attitudes towards cancer information vary according to demographic, socio‐economic and health characteristics, and cancer‐related perceptions and knowledge (Table 3). People aged 35–54 and 55–74 were more likely to have a positive attitude towards cancer information than younger people (aged 18–34) (P < 0.001). However, no differences were found between men and women after controlling for potential confounders – age, socio‐economic and health characteristics, and cancer‐related perceptions and knowledge. The likelihood of having a positive attitude increased with the level of education (P < 0.001). Regarding health characteristics, the likelihood of having a positive attitude decreased in people with poor/very poor self‐rated health status [odds ratio (OR): 0.63; 95% confidence interval (CI): 0.42–0.95] (reference: very good/good). In contrast, the opposite was true when comparing individuals by cancer status: cancer survivors were more likely to have a positive attitude towards cancer information than those who had not had the disease (OR: 3.05; 95% CI: 1.73–5.38).

Table 3.

Variables associated to a positive attitude towards receiving cancer information

Variables	Univariate models			Multivariate modelsa
	OR	95% CI		OR	95% CI
		Lower	Upper		Lower	Upper
Demographic characteristics
Sex
Men (ref: women)	0.85	0.71	1.02	0.91	0.76	1.11
Age (ref: 18–34)
35–54 years old	1.39	1.13	1.70	1.40	1.13	1.74
55–74 years old	1.63	1.27	2.11	2.05	1.52	2.76
75 years old or older	0.83	0.53	1.30	1.22	0.72	2.06
Socioeconomic characteristics
Education level (ref: <primary)
Primary	1.89	1.12	3.19	1.67	0.96	2.93
Secondary/Vocational training	2.03	1.25	3.30	2.27	1.30	3.96
University	3.38	2.04	5.61	3.32	1.84	6.00
Health characteristics
Self‐perceived health status (ref: very good/good)
Fair	0.81	0.65	1.00	0.85	0.67	1.09
Poor/Very poor	0.59	0.41	0.84	0.63	0.42	0.95
Have or have had cancer (ref: no)
Yes	2.51	1.49	4.23	3.05	1.73	5.38
Perceptions and knowledge on cancer
Perceived risk of cancer (ref: very low/low)
High/Very high	0.79	0.65	0.96	0.75	0.60	0.92
Don't know/No answer	0.63	0.49	0.82	0.66	0.50	0.87
Importance attributed to cancer risk factors (ref: low importance <7)
Diet and nutrition
High importance (7–10)	1.49	1.25	1.78	1.28	1.05	1.55
Weight
High importance (7–10)	1.61	1.30	1.99	1.36	1.08	1.72
Don't know/No answer	1.17	0.81	1.71	1.09	0.72	1.63
Exposure X‐ray/radiation
High importance (7–10)	1.38	1.14	1.67	1.31	1.07	1.61
Don't know/No answer	0.97	0.67	1.40	0.95	0.63	1.44
Alcohol
High importance (7–10)	1.50	1.24	1.81	1.37	1.11	1.68
Interaction: Lifestyle × Knowledge symptoms
Healthy lifestyle
Knowledge about symptoms (ref: no)
Yes	1.72	1.41	2.10	1.50	1.21	1.87
Unhealthy lifestyle
Knowledge about symptoms (ref: no)
Yes	0.84	0.54	1.33	0.72	0.44	1.17

OR, odds ratio; CI, confidence interval; ref. reference.

^aAdjusted for autonomous region of residence.

People who perceived they had a high risk of developing cancer were less likely to have a positive attitude towards cancer information than people who perceived they had a low risk (OR: 0.75; 95% CI: 0.60–0.92). On the other hand, individuals who gave a high importance to cancer risk factors were more likely to react positively to cancer information. Interestingly, we detected an interaction between knowledge of symptoms or signs of cancer and perceived lifestyle. Being familiar with cancer symptoms or signs increased the likelihood of a positive attitude towards information only among those who reported a healthy lifestyle (OR: 1.50; 95% CI: 1.21–1.87) but not among those perceiving their lifestyle as unhealthy. Several variables, cancer being the most feared disease, perceiving cancer as a very serious health problem in Spain, birthplace and identifying tobacco, family history or sun exposure as important risk factors for cancer, were not associated with attitude towards cancer information.

Variables such as fear of cancer, perceived severity and expectations of a definite cancer treatment were non‐significant and did not remain in the regression model.

Discussion

Almost 40% (38.8%) of the weighted sample reported receiving, seeing or reading some cancer‐related information in the last 6 months. The proportion of people exposed to this information is greater among those in the middle age category (35–54 years of age), and it increases with educational level. It is also directly associated with knowledge of symptoms or signs of cancer, the importance assigned to risk factors such as diet and nutrition, weight, exposure X‐ray and alcohol, and high/very high self‐perceived risk for developing cancer. Cancer‐related information was received with a positive attitude by 68.5% of people surveyed; that is, it encouraged them to keep informed, to improve their lifestyle or to get screened. Likelihood of such positive respo‐nse is directly associated with being between the ages of 35 and 54 (vs. other age groups), with higher educational levels and with attributing an important role in the development of cancer to certain risk factors. In contrast, positive attitudes are less likely among those who rated their health as poor or very poor and among those who perceive being at high risk for developing the disease. Among those with healthy lifestyle, reporting familiarity with cancer symptoms or signs increases the chances of having a positive attitude towards cancer information, but this is not the case among those reporting unhealthy lifestyle.

Our findings, placed in the context of previous results in the literature, underscore the great variability of findings regarding exposure to cancer information. Whereas our population‐based study found that close to 40% of respondents had received information in the last 6 months, Kelly and colleagues5 in USA reported that 90% of their sample had been exposed to scanned information about healthy lifestyles and/or cancer screening tests. In contrast, Tortolero‐Luna and colleagues6 in Puerto Rico estimated that 28.1% of the sample had ever engaged in cancer information‐seeking practices. The variability in estimates may be related to the different context where the studies were conducted, with type of access to the information examined in recent studies (scanning/seeking/either), information content (healthy lifestyle/cancer information/cancer screening) as well as the time period (ever/in the last year/in the last 6 months).

Our findings regarding the characteristics of information recipients support the results of previous studies.4, 5, 6 The majority of individuals reporting receiving cancer information are women, between 35 and 54 years of age and with high educational level. They tend to be individuals who have or have had cancer or who have a high or very high perceived risk of developing cancer.

Recently published research points out that lower access to cancer‐related information among lower socio‐economic status groups may lead to wider health inequalities.19 Currently, net of incidence rates, lower socio‐economic status groups present higher cancer mortality rates and lower survival.20 Improving information access to these groups may contribute to the reduction of these inequalities although it is also necessary to encourage a positive attitude towards cancer information.

In our study, 32% of people who received cancer information expressed a negative attitude towards it. This proportion is similar to the group of adults that reported avoiding cancer information as described by Miles and colleagues.11 Notwithstanding, the most important contribution of our work is the ability to document, using population‐based data, that individuals reporting a high perceived risk of cancer were less likely to have a positive attitude towards cancer information. There is solid scientific evidence in this regard establishing that such association between the perceived threat (risk and severity) and the response to it is contingent on the coping abilities of the individuals.21, 22, 23 As in our study, the self‐perception of high risk of cancer does not seem to promote a positive attitude towards cancer information, one may suggest that the Spanish population is exposed to information regarding a threat (cancer) but lacking corresponding information on coping strategies which would allow individuals to develop an adaptive response. It is worth noting that our results show that individuals that assign high importance to certain cancer risk factors (weight, diet, radiation, alcohol) are more likely to have a positive attitude towards information. As supported by previous researches, these results may indicate that better cancer knowledge would influence lifestyle behaviour, prevention and screening decisions.9, 12 No such association was found with other risk factors such as tobacco or sun exposure, however. This is possibly due to the widespread knowledge of these behaviours as cancer risk factors and, thus, becoming poor discriminators of attitudes towards information about the risk associated with them.

A novel contribution to the literature is our result that being familiar with cancer symptoms or signs is related to a positive attitude towards information but only among those reporting a healthy lifestyle. Previous studies24 highlight the importance of adapting cancer information to sociodemographic characteristics and education level of the target population when promoting preventive behaviours. However, our results suggest a new interaction effect, worth examining in greater detail in future qualitative research aimed to enhance our understanding of the relationships between knowledge of symptoms of cancer and lifestyles. An in‐depth analysis conducted with people with and without healthy lifestyles could help to disentangle the perceptions and attitudes of both groups towards cancer information. In regard to health status, our work shows that poor self‐rated health is associated with lower likelihood of having a positive attitude towards cancer information, which fits nicely with recent findings16 linking poor self‐rated health with lower participation in screening programmes in England.

Finally, it is important to point out that our results suggest that those with higher socio‐economic status are more likely to report a positive attitude towards cancer information. Recent studies provide solid evidence on socio‐economic determinants of cancer,20 disparities in information access19 and social inequalities in screening rates.16, 25 Our work contributes to this literature by underscoring current social cancer disparities and their implications regarding cancer prevention and control.

Our study has several notable limitations. Approximately 60% of the weighted sample was excluded from the analysis because they reported that they had not received, seen or read any cancer‐related information in the last 6 months. The dependent variable, attitude towards received information, refers to individuals’ exposure to any type of cancer information. This measure has not been validated yet, but allowed us to have an approach to information difficult to find. However, this variable captures real‐world conditions under which people come into contact with information and how, in turn, this contact may affect cancer‐related behaviours. In other words, it suggests an impact of information exposure on cancer, but it does not allow us to associate such impact with a specific message content, message source or access method (passive vs. active). The variable used in the parent study, the OncoBarómetro, refers to ‘receiving, seeing or reading information’ specifically related to cancer in the last 6 months, but it does not allow us to discriminate between active information seeking and passive exposure to the information. It is important to underline that the independent variables regarding health status, cancer‐related perceptions and knowledge, and type of lifestyle are based on self‐reported data. Of note, self‐rated health status has been shown to be an indicator of health with high validity and strong association with mortality and healthcare services utilization.16, 26, 27 Similarly, the variable self‐perceived cancer risk allows us to appraise the level of vulnerability that the person feels in regard to the disease, which may be one of the mediators between information/knowledge and actual preventive behaviours. Still, self‐reported data on life‐style may be biased due to social desirability, especially given that interviews are face to face. However, any measurement error resulting from socially desirable responses would bias the degree of association towards the null value.

Unfortunately, the survey was not able to include objective measures on lifestyle, cancer‐related knowledge, variables related to severity, response–efficacy, self‐efficacy, health literacy or response cost of acting positively towards the information, for example, the negative cost derived from the fear to do a screening or the fear of feeling vulnerable to the disease. Independent variables related to perceptions and knowledge on cancer have not been validated in previous studies, and there is a great heterogeneity on how to measure these variables in the literature.9, 28, 29 Our analysis failed to detect any associations between the attitude towards information and perceiving cancer as a serious/very serious health problem in Spain, perceiving cancer as the most feared disease or expecting a definite treatment for cancer in the near future. The reason for the lack of said associations may reside in how these three variables were defined. For instance, instead of asking whether respondents consider cancer to be a serious disease, they are asked whether they consider it to be a serious disease in Spain. Similarly, the question about the fear of cancer is asked in relation to other diseases, instead of in absolute terms. Lastly, the question about respondent's expectations of a definite cancer treatment refers to a near future (10 years). Not surprisingly, then, these variables have little impact on attitudes towards information in comparison with perceived risk of cancer which directly relates to individual's vulnerability or exposure to the disease. Unfortunately, the data set did not include variables evaluating self‐efficacy or response cost, which may partly explain the association found between perceived risk and a negative attitude towards cancer information. Despite the limitations of this study, two main strengths of this study should be highlighted: on one hand, no previous article has analysed attitudes towards cancer information; and on the other hand, the OncoBarometro is a representative sample of the Spanish population.

Conclusions

Based on the results from this population‐based study, in order to improve the attitude towards cancer information, policies should target population groups more likely to respond to cancer information with indifference, frustration or fear, that is, low socio‐economic groups, those with poor self‐rated health and individuals with high perceived risk of cancer. The use of local media in neighbourhoods with low socio‐economic status and including accessible cancer information in health centres could be a way to target this population. Further, in order for knowledge of cancer symptoms/signs to translate into a better attitude towards information, we need to focus on individuals with unhealthy lifestyles as, in this particular group, better cancer knowledge does not lead to better attitude towards cancer‐related information.

Source on funding

This study has been funded by the Scientific Foundation of the Spanish Association against Cancer, Madrid, Spain.

Conflict of interest

The authors have no conflict of interests.