A Study of Trends in Beliefs and Attitudes Toward Cancer

Transcription

1 J Canc Educ (2010) 25: DOI /s z A Study of Trends in Beliefs and Attitudes Toward Cancer Eva Schernhammer & Gerald Haidinger & Thomas Waldhör & Roberto Vargas & Christian Vutuc Published online: 30 April 2010 # Springer 2010 Abstract Understanding trends in population perceptions of cancer is essential in identifying groups that may benefit from more targeted strategies in cancer prevention and detection. We report the prevalence of knowledge and attitudes with respect to various aspects of cancer in two independent national probability samples of 4,073 Austrian adults aged 15 years or above, and we describe the association of these variables with age, gender, marital status, and education. In-home interviews were conducted in September 1995 and in August Perceptions of cancer causes, positive personal experiences with cancer cures, and knowledge of cancer cures and causes generally improved across all groups. Those with less education were less likely to believe that cancer could be cured in 1995, a difference that resolved in Less encouragingly, education-based disparities persisted for factors such as E. Schernhammer Department of Medicine, Channing Laboratory, Brigham and Women s Hospital and Harvard Medical School, Boston, MA 02115, USA E. Schernhammer : G. Haidinger : T. Waldhör : C. Vutuc Department of Epidemiology, Centre of Public Health, Medical University of Vienna, Vienna, Austria R. Vargas Division of General Internal Medicine and Health Services Research, David Geffen School of Medicine at UCLA, Los Angeles, CA 90024, USA E. Schernhammer (*) Channing Laboratory, 181 Longwood Avenue, Boston, MA 02115, USA eva.schernhammer@channing.harvard.edu knowledge of the importance of early detection in cancer cure and experiences with cancer cures. Differences in other subgroups, although existent, were only modest. Prevention education strategies should consider targeted approaches that aim to reduce disparities in cancer perception among those with less formal education. Keywords Cancer. Knowledge. Attitude. Behavior Introduction Cancer remains one of the main killers both in Austria [1] and worldwide [2, 3]. Much of the world's cancer burden can be attributed to environmental [4], lifestyle [5], and genetic factors [6] and the interplay of these factors, and it has been argued that a majority of all cancers could be prevented if cancer risk behavior would reflect knowledge of cancer risk factors [7]. To change peoples' behavior towards cancer requires understanding their beliefs and factors that influence these beliefs. Several psycho-social models of health beliefs aim to explain people's health behaviors by relating them to their beliefs. Essentially, these models are based on constructs thought to be necessary prerequisites for people's readiness to act [8 10], such as (besides cost) perceived susceptibility and severity of illness. Other factors with strong influence on a person's cancer preventive potential are poverty [11], socioeconomic status [12, 13], and access to health care [14]. To evaluate factors that influence beliefs related to cancer control over time, we studied a large, representative sample of Austrians. The country's health system includes statutory health insurance with close to 100% population coverage. Austria is one of the most affluent countries in the world [15].

2 212 J Canc Educ (2010) 25: Methods The data analyzed were extracted from the population-based cross-sectional study Attitudes Towards Cancer conducted twice, in September 1995 and in August Each of the surveys, which were funded by the Austrian Cancer Society, covered all parts of Austria (population, 8.17 million; 6.85 million 15 years of age; 1.77 million men and 1.92 million women years in 2005). In cooperation with a public opinion survey institute, a quota sample of Austrians aged 15 years and over was pre-selected, which was representative with regard to age, sex, occupational status, and area of residence (source: Population Census 1991). The aim of this study was to survey knowledge, attitudes, and behavior regarding various aspects of cancer. Selection criteria, sample size, methods, and questionnaires were identical in 1995 and 2005 [16 19]. Study Population For both surveys, in cooperation with a public opinion survey institute, a quota sample of 2,400 Austrians aged 15 years and over was selected and invited in writing to participate in a survey to study beliefs and attitudes about cancer, its risk factors, and treatment. The sample comprised 0.04% of the population 15 years of age and was representative in terms of age, sex, occupational status, and area of residence (source: Population Census 2001). Survey Instrument In September 1995 and August 2005, respondents were visited in their homes by trained interviewers. The interview was face-to-face, using a standardized questionnaire. Of the 2,400 planned interviews, only 86.4% and 83.3%, respectively, were completed because the subjects were not at home or refused interview. The remaining persons met the inclusion criteria regarding representativeness. For validation purposes, 30% of the interviews were repeated by telephone, and no significant differences to the original face-to-face interview were found. Specifically, the surveys queried perception of cancer cure ( Do you believe that cancer can be cured? and Do you know someone who was cured from cancer? ) and factors contributing to cure ( Do you believe that is important in cancer cure? ), with 5 on a rating scale from 1 to 5 considered as most important. Beliefs about cancer were assessed via an openended question querying the main causes for the development of cancer, allowing for multiple entries. Besides eliciting specific information on beliefs and attitudes toward cancer genesis and prevention, data on sociodemographic variables concerning age, gender, marital status, residence, education, and income were also collected. While information on other demographics was complete, a relatively large proportion of participants chose not disclose information about their employment status and income. Statistical Analysis The sample of respondents was weighted by age ( poststratified ) to the Austrian census population in Frequencies and 95% confidence limits were calculated by the procedure SURVEYFREQ in SAS. The chi-square distribution was used to test the significance of the difference between the two time periods (1995 and 2005). P values were two-sided, and P values of 0.05 were considered statistically significant. All statistical analyses were performed using the SAS 8.2 statistical package (SAS Institute, Cary, North Carolina). Results We had valid data for 2,073 participants enrolled in the 1995 survey and 2,000 participants of the 2005 survey. Demographic factors were largely comparable between the two surveys. The differences were modest and reflected expected trends in demographics. For example, in 2005, participants were slightly more likely to live in single households and less likely to be married than in General education appeared to improve over time, with fewer people reporting only primary school (21.5% in 2005 vs. 29.5% in 1995) and more reporting having completed school years beyond primary school in 2005 as compared with that from In addition, the unemployment rate rose slightly between 1995 and 2005 (1.6 vs. 4.9%), whereas fewer people reported a household income in the lowest bracket, with largest income gains seen among women. Overall, we observed a trend towards a more positive attitude regarding the curability of cancer over time. The proportion of participants reporting knowing someone who was cured of cancer rose from 51.2% (1995) to 55.7% (2005), an increase which was paralleled by more participants believing that cancer could be cured most of the time (57.2% in 1995; 64.1% in 2005). When asked about the main factors contributing to cancer cure, participants ranked early detection, type of cancer, and physical constitution highest, and this ranking remained fairly stable between the studies (Table 1). However, the importance of an individual's psychological constitution (51.0% vs. 44.1%) or the treating physician's role (43.3% vs. 37%) was considered less important in 2005.

3 J Canc Educ (2010) 25: Table 1 Summary statistics comparing 1995 with 2005 individual beliefs regarding potential for cancer cure and causes for developing cancer, by gender Women Men Characteristic or belief (%) 1995 n=1, n=1,041 P for difference a 1995 n= n=959 P for difference a Can cancer be cured? Always, almost always, or frequently Rarely or never Don't know Do you know someone personally who was cured of cancer? Yes 53.9 c 58.5 c c 52.6 c 0.05 No 45.4 c 41.5 c c 47.4 c 0.07 Which of the following is most important in curing cancer? b Early detection Psychological constitution c c c c <0.001 Physical constitution c c Type of cancer c c 1.0 Attending physician Alternative medicine c c c c 0.02 In your opinion, what are the main causes of cancer? d Smoking c c c c 0.02 Environment c < c 0.10 Diet Genetics c c < c c <0.001 Alcohol c c Psyche c c c c 0.33 Stress < Lifestyle (including exercise) < <0.001 Too much sun exposure c c 0.06 Drugs c c 0.02 Working conditions 2.6 c c Pesticide exposure Carcinogens, viruses, bacteria Immune system Ozone Other < <0.001 Don't know < a Difference between the two surveys (1995 vs. 2005); χ 2 statistics b Percentage rating belief or behavior with 5 on a scale from 1 to 5 c Significant difference between women and men in respective surveys (p 0.05) d Self-reported; multiple entries possible When asked to list the main causes for cancer, participants ranked smoking and the environment highest in 1995, followed by diet, genetics, alcohol, and stress, closely followed by lifestyle and too much sun exposure. In 2005, this ranking remained mostly unchanged, with the exception of genetics rising from rank 4 to rank 2 and lifestyle, which in 2005 was ranked higher than stress. Next, we analyzed these attitudes and how they changed over time by subgroups. In addition to age and gender, we stratified on all the demographics that changed over time including employment status, household income, and education. We found only modest differences when we stratified by age groups (15 to 34 vs. 35+years). Overall, younger

4 214 J Canc Educ (2010) 25: people placed more value on sun exposure and smoking as cancer risk factors than did older participants (both surveys). In 1995, younger people felt more strongly that the environment constituted a major cancer risk factor than did their older counterparts, a difference which was less pronounced in While men and women had similar perceptions about the curability of cancer, fewer men reported knowing someone who was cured of cancer, a difference that remained fairly stable over time. There were also notable differences in the proportion of women who considered psychological constitution as an important factor in curing cancer, compared Table 2 Comparing 1995 with 2005 believes about a potential cancer cure and causes for developing cancer, by education (men and women combined) Primary school Secondary school, university Characteristic or belief (%) 1995 n=1, n=1,366 P for difference a 1995 n= n=634 P for difference a Can cancer be cured? Always, almost always, 54.3 b 62.7 < b or frequently Rarely or never 34.9 b 28.1 < b Don't know 10.8 b b Do you know someone personally who was cured of cancer? Yes 48.7 b 53.6 b b 59.8 b 0.28 No 50.8 b 46.4 b b 40.2 b 0.39 Which of the following is most important in curing cancer? c Early detection b a Physical constitution Psychological constitution Type of cancer Attending physician b < b Alternative medicine In your opinion, what are the main causes of cancer? d Smoking Environment Diet b b Genetics b b < b b <0.001 Alcohol b b 0.29 Stress b b Too much sun exposure b b Psyche b b 0.45 Lifestyle (including exercise) b b < b b <0.001 Drugs Working conditions Pesticide exposures Carcinogens, viruses, bacteria Immune system Ozone < Other < <0.001 Don't know b b a Difference between the two surveys; χ 2 statistics b Significant difference between primary school vs. secondary school/university in respective surveys (p 0.05) c Percentage rating belief or behavior with 5 on a scale from 1 to 5 d Self-reported; multiple entries possible

5 J Canc Educ (2010) 25: with men. Comparing beliefs by gender over time, women were also more likely than men to believe that alternative medicine was important in curing cancer (Table 1). The proportion of those who thought that physical constitution was important in curing cancer decreased in men but not in women. The percentage of women who regarded alcohol consumption as a main cancer risk factor, finally, was larger in 2005 than in 1995, a difference that we did not observe among the men. Next, we stratified on education (primary school and/or apprenticeship vs. secondary school with or without diploma and/or university degree; Table 2 and Fig. 1). We found marked differences in perceived beliefs and attitudes toward cancer (Table 2). Optimism in terms of cure potential was higher among those better educated in 1995 (64.3% thought that cancer could be cured most of the time vs. 54.3% among the less educated), which was reflected also in the proportion of people who knew someone who was cured of cancer. These differences diminished over time; 64.2% with primary schooling vs. 62.7% with secondary schooling thought that cancer could be cured most of the time in In sum, the attitude of lesseducated Austrians became more optimistic, whereas the attitude of those with higher education remained fairly stable. In addition, a larger proportion of the highereducated Austrians reported personally knowing someone who had been cured from cancer in 2005 (59.8%) than of the lower education group (53.6%). While early detection was ranked first in terms of importance in curing cancer by both subgroups, those with higher education levels showed marked gains between 1995 Fig. 1 Differences in cancer beliefs and experiences over time and by education and 2005, whereas no such difference was seen in participants with primary education only (Fig. 1). Regarding the main causes for cancer, distinct differences according to education also emerged. Most importantly, less-educated participants (primary education only) placed less value on lifestyle than did participants with higher educational levels. Psychological strain or stress was rated less frequently as a major cancer risk factor by those less educated in both 1995 and Only 24.9% (1995) and 32.7% (2005) of the less educated thought that genetics was a major risk factor, whereas 32.1% (1995) and 42.3% (2005), respectively, of those with higher education thought so. In 2005, fewer participants with only primary education than those with higher educational levels considered diet a main cancer risk factor (32.8 vs. 37.1%). Overall, while differences between less and more educated people in beliefs about smoking (Table 2) appear to have decreased over the decade, certain differences in perception regarding the importance of alcohol consumption, genetics, and lifestyle in the development of tumors, for example, persisted. Discussion We studied time trends of beliefs and attitudes toward cancer and cancer risk factors in the Austrian population, where everyone benefits from a mandatory health insurance system providing a wide range of free screening and other preventive services. To our knowledge, ours is the first study to examine trends in beliefs and attitudes toward cancer over time, with two independent assessments in 1995 and Overall, we observed an improvement in the perception of main risk factors for cancer, with the most frequently listed factors believed to be main causes for cancer being in accordance with major scientifically recognized risk factors for cancer as investigated by several prior European surveys [20 22]. However, while a disparity in cancer cure belief by education group appeared to have resolved over time, there was a persistent disparity in experience and knowledge about cancer cure by education in our study. Few studies have examined perceived cancer risk, and they tended to find that people are generally aware of cancer risk factors but fail to change their behavior accordingly. Similar to our findings, a study in New Zealand suggests that people's awareness of their personal cancer risk was adequate, likely reflecting prior efforts from public awareness programs [23]. In Spain, individuals with higher education had cancer risk beliefs similar to those with primary or less than primary school education, but tended to score lower in cancer risk behavior [20].

6 216 J Canc Educ (2010) 25: Cognitive abilities including the use of numerical information [24] have been suggested as potential reasons for the observed discrepancy between cancer risk perception and taken actions. Our findings of education-based disparities in cancer beliefs are in line with previous reports from the USA. A telephone survey of 957 cancer-free US adults revealed that a relatively large proportion of this sample held inaccurate cancer beliefs, a disparity that was much more strongly pronounced among less-educated participants [25]. In studying colorectal cancer knowledge, perceptions, and behavior among 100 African-Americans aged 50 years [26], it was found that barriers to screening were strongly inversely correlated with post-secondary education. Poor literacy was an even stronger negative predictor of cervical screening than education in a multiethnic cohort of women in Chicago [27]. Cotugna et al. [28], finally, report that education was the single most important factor to impact cancer prevention knowledge and attitude in their large US national probability sample (n=22,043). Limitations of our study include the large number of persons who chose not to disclose information about their employment status and income as well as the limited number of factors that was evaluated. No information on actual risk behavior as opposed to risk beliefs was available. Finally, self-report of cancer risk factors could be considered less valuable than ranking within a preselected set of cancer risk factors. A major strength of this study is the large sample size and close to complete representativeness of people in Austria, an affluent country with equal access to care for close to 100% of the population. In conclusion, while overall encouraging, our results support the need for continued monitoring of beliefs and attitudes toward cancer in Austria. A more detailed assessment of disparities is needed for the development of targeted approaches. Prevention education strategies should consider targeted approaches that aim to reduce disparities in cancer perception among those with less formal education. Acknowledgements We are indebted to all Austrians who participated in this survey for providing their data. We would also like to thank the Austrian Cancer Society (Prof. Paul Sevelda, President and Prof. Michael Micksche, past Vice-President) for funding the population-based crosssectional study on Attitudes Towards Cancer in Reference 1. Vutuc C, Waldhoer T, Haidinger G (2004) Cancer mortality in Austria: Wien Klin Wochenschr 116(19 20): International Agency for Research on Cancer (WHO) (2003) Cancer incidence in 5 continents. In: Parkin DM, Whelan SL, Ferlay J, Teppo L, Thomas DB (eds) IARC Scientific Publications 3. Boyle P, Ferlay J (2005) Cancer incidence and mortality in Europe, Ann Oncol 16(3): Belpomme D, Irigaray P, Sasco AJ et al (2007) The growing incidence of cancer: role of lifestyle and screening detection (Review). Int J Oncol 30(5): Kolonel LN, Altshuler D, Henderson BE (2004) The multiethnic cohort study: exploring genes, lifestyle and cancer risk. Nat Rev Cancer 4(7): Doll R, Peto R (1981) The causes of cancer: quantitative estimates of avoidable risks of cancer in the United States today. J Natl Cancer Inst 66(6): Colditz GA, Atwood KA, Emmons K et al (2000) Harvard report on cancer prevention. Cancer Causes Control 4(11): Janz NK, Becker MH (1984) The Health Belief Model: a decade later. Health Educ Q 11(1): Ajzen I (2001) Nature and operation of attitudes. Annu Rev Psychol 52: Glanz K, Rimer BK, Lewis FM(2002) Health behavior and health education: theory, research and practice. Wiley & Sons, San Francisco 11. Wilkes G, Freeman H, Prout M (1994) Cancer and poverty: breaking the cycle. Semin Oncol Nurs 10(2): Freeman HP, Chu KC (2005) Determinants of cancer disparities: barriers to cancer screening, diagnosis, and treatment. Surg Oncol Clin N Am 14(4): , v 13. Segnan N (1997) Socioeconomic status and cancer screening. IARC Sci Publ 138: Bezold C (2006) Disparities, cancer surveillance, and next steps. J Cancer Educ 21(1 Suppl):S61 S World Bank (2007) GNI per capita World Bank 16. Haidinger G, Waldhoer T, Vutuc C (1998) The prevalence of smoking in Austria. Prev Med 27(1): Vutuc C, Haidinger G, Waldhoer T (1998) Prevalence of selfreported screening mammography and impact on breast cancer mortality in Austria. Wien Klin Wochenschr 110(13 14): Vutuc C, Waldhoer T, Haidinger G (2006) Breast cancer trends: opportunistic screening in Austria versus controlled screening in Finland and Sweden. Eur J Cancer Prev 15(4): Vutuc C, Waldhoer T, Sevelda P et al (2007) Self-reported opportunistic screening mammography in Austria, 2005 vs Breast Care (in press) 20. Garcia M, Fernandez E, Borras JM et al (2005) Cancer risk perceptions in an urban Mediterranean population. Int J Cancer 117(1): La Vecchia C, Anelli M, Zuccato E (1999) Perceived risk of cancer in Italy. Eur J Cancer Prev 8(5): La Vecchia C, Anelli M, Zuccato E (2000) Perceived risk of cancer in population samples from 5 european countries. Int J Cancer 86(5): Trevena J, Reeder A (2007) Perceptions of New Zealand adults about reducing their risk of getting cancer. N Z Med J 120(1258):U Klein WM, Stefanek ME (2007) Cancer risk elicitation and communication: lessons from the psychology of risk perception. CA Cancer J Clin 57(3): Stein K, Zhao L, Crammer C et al (2007) Prevalence and sociodemographic correlates of beliefs regarding cancer risks. Cancer 110(5): Green PM, Kelly BA (2004) Colorectal cancer knowledge, perceptions, and behaviors in African Americans. Cancer Nurs 27(3): ; quiz Lindau ST, Tomori C, Lyons T et al (2002) The association of health literacy with cervical cancer prevention knowledge and health behaviors in a multiethnic cohort of women. Am J Obstet Gynecol 186(5): Cotugna N, Subar AF, Heimendinger J et al (1992) Nutrition and cancer prevention knowledge, beliefs, attitudes, and practices: the 1987 National Health Interview Survey. J Am Diet Assoc 92(8):