Health Education Research, Vol. 17, No. 2, 239-252,
April 2002
© 2002 Oxford University Press
Tailored versus general information on early detection of cancer: a comparison of the reactions of Dutch adults and the impact on attitudes and behaviors
Department of Health Education and Promotion, Maastricht University, PO Box 616, 6200 MD Maastricht and
1 Department of Social Sciences, Open University Netherlands, PO Box 2960, 6401 DL Heerlen, The Netherlands
 |
Abstract |
|---|
 Top Abstract IntroductionInterventionMethodsResultsDiscussionReferences |
|---|
This article describes the process evaluation of two interventions (tailored or general information) to encourage people to behave in a manner conducive to the early detection of cancer. A total of 1040 Dutch adults participated in the study. Tailored information is adapted to the characteristics, needs and interests of the individual, and hence has more personal relevance and contains less redundant information. It was therefore hypothesized that tailored information would be evaluated better than general information, does not increase worries, influences attitude and behavior change, and that no differences in subgroups would be found in the reactions to the tailored information. The results showed that both kinds of information were read by a large majority of the participants. Recipients of the tailored information saved and discussed the information with others more often, and they were more appreciative of its format and content. Tailored information resulted significantly more often in changes in attitudes and behavior after the intervention. It is concluded that information adapted to individuals' characteristics is an accepted method of health education in early detection of cancer.
|
Introduction |
|---|
|
TopAbstractIntroductionInterventionMethodsResultsDiscussionReferences |
|---|
The earlier cancer is detected, the better the chances for successful treatment and survival (Ruwaard and Kramers, 1997
; American Cancer Society, 2000). Encouraging behaviors associated with early detection of cancer allows treatment to take place, and may contribute to reducing cancer morbidity and mortality (Tabar et al., 1985). Early detection of cancer encompasses various forms of behavior, such as passive or active self-detection behavior, help-seeking behavior and participation in screening programs. Passive detection behavior may be described as becoming aware of cancer symptoms, while no concrete actions have to be undertaken. This assumes correct knowledge and interpretation of cancer symptoms. An example is being attentive to changes in bowel habits or unusual bleeding. Active detection refers to concrete actions to detect cancer symptoms, such as breast self-examination (BSE). Among the general public knowledge of symptoms and detection methods is insufficient (Berman and Wandersman, 1991; Bostick, 1993; Lechner and De Vries, 1996; Deliens, 1998; Sheik and Ogden, 1998). Some forms of early detection behavior, such as looking out for cancer symptoms, are hardly engaged in, especially by certain subgroups such as young males (Lechner, 1998; De Nooijer et al., submitted). In contrast, performing BSE and participating in screening programs are more common within target groups (Cope, 1992; Bostick, 1993; Friedman, 1994; Budden, 1995; Wardle et al., 1995; Lechner et al., 1997; Persson et al., 1997; De Nooijer et al., submitted). Various subgroups can be distinguished among the general population, based on cancer incidence (some cancers occur more frequently among specific age groups or gender) (Visser et al., 2000), early detection behaviors [some forms of behavior are relevant for specific subgroups, e.g. BSE for women and testicular self-examination (TSE) for men] (De Nooijer et al., submitted) and the social psychological correlates of early detection behaviors (e.g. some people have higher self-efficacy expectation regarding early detection than others) (De Nooijer, et al., submitted). Hence, it is not easy to give one single clear general message about the early detection of cancer in health education material that suits the needs of all these groups.
Interventions to stimulate individuals to engage in actions conducive to early detection of cancer mainly focus on participating in screening programs or performing BSE or TSE [see, e.g. (Friman, 1986; Rakowski et al., 1998)]. Interventions to stimulate individuals to be attentive to cancer symptoms in their own body (passive detection) and to respond appropriately to the detection of a possible cancer symptom are rare. One example was the European Week Against Cancer in 1998 (Dutch Cancer Society, 1998), which focused on men and cancer. The intervention included a brochure, encouraging men to look out for symptoms of common forms of cancer among males. Unfortunately, it was not tested on its format and impact. It is often assumed that these types of interventions increase worries about or fear of cancer. However, various studies on encouraging people to adopt early detection behaviors were found not to increase fear (Best et al., 1996; Weist, 1996; Mansson et al., 1999).
Recent new approaches in health education allow for the adaptation of health education messages to the individual level. Using computerized tailoring respondents receive health information adapted to their characteristics, needs and interests. This results in personally relevant information and less redundant information. According to Petty and Cacioppo's Elaboration Likelihood Model, personally relevant information is more likely to be thoughtfully considered via the `central route' (Petty and Cacioppo, 1986). This is information processing that works by engaging in a subject thoughtfully and in a manner that will result in enduring attitude change. It is therefore considered to be more effective in changing attitudes than general information (Brug et al., 1996; Dijkstra et al., 1999). Subsequently, an enduring attitude is likely to influence behavior (Eagly and Chaiken, 1990). Throughout the last decade, computerized tailored health interventions on various forms of health behavior have been developed, applied and evaluated. These interventions have proven to be effective in changing intentions and behaviors for a number of forms of health behavior, such as physical activity (Bull et al., 1999), smoking (Strecher et al., 1994; Dijkstra et al., 1998), dietary habits (Campbell, 1994; Brug et al., 1996, 1999; De Bourdeaudhuij and Brug, 2000), mammography (Skinner, 1994; Rakowski et al., 1998) and weight loss (Kreuter et al., 1999). Tailored interventions can be distinguished from traditional interventions in that the former are intended for a particular person rather than a group and are based on individual-level factors related to health or behavior outcome (Kreuter and Skinner, 2000). Compared to expensive and time-consuming interpersonal communication, computerized tailoring enables us to reach large target groups at relatively low cost (De Vries and Brug, 1999; Velicer and Prochaska, 1999; Kreuter and Skinner, 2000). An example is a computer-tailored smoking cessation intervention, implemented by the Foundation for Smoking and Health. In 2000, 15 000 individuals received a tailored letter about smoking cessation, which cost about $5 per letter. In comparison, approximately 200 individuals participated in group sessions to stop smoking, which cost about $100 per person (M. Willemsen, pers. commun.).
Dijkstra and De Vries identified why tailored messages are likely to increase the effectiveness of the information (Dijkstra and De Vries, 1999). Firstly, the message is adapted to prior knowledge. Secondly, the message contains information that is relevant only to the individual in question and thus contains more extensive information. Thirdly, since the message is adapted to the recipient's characteristics and intention to change their behavior, defensiveness to the message, which may lead to counter argumentation, is limited. Fourthly, the message is adapted to the phase of the behavior change process and can therefore offer individual behavioral goals.
So far, no tailored interventions have been developed to stimulate people to adopt early detection behaviors regarding cancer, with the exception of mammography (Skinner, 1994; Rakowski et al., 1998). The present paper describes a tailored intervention aimed at motivating the general public to engage in behavior conducive to the early detection of cancer, and evaluates this intervention on its effects on attitudes, passive detection and help-seeking behavior. The process evaluation was conducted in order to analyze how this tailored intervention was assessed compared to the current information on early detection of the Dutch Cancer Society in terms of use, format and content. Tailored information is adjusted to the characteristics of the recipients and the information is designed to be more personally relevant to all readers. The present study evaluates therefore four hypotheses. First, tailored information will be more positively evaluated than the general information regarding use (reading, saving and discussing), format aspects (e.g. personally relevant, interesting) and content aspects (e.g. does the information convince the recipient why s/he should engage in early detection behaviors). Second, tailored information will not result in more worries about a particular complaint or diagnosis, since it discusses personal worries. Third, tailored information will result in a greater influence on attitudes and behaviors of respondents than general information. Fourth, tailored information will not be evaluated differently among subgroups of the recipients of the tailored intervention.
|
Intervention |
|---|
|
TopAbstractIntroductionInterventionMethodsResultsDiscussionReferences |
|---|
Computerized tailored feedback requires three inter-related elements (De Vries and Brug, 1999
; Dijkstra and De Vries, 1999): a screening instrument that enables the researchers to tailor the information to the individual's characteristics, a message source file that provides the feedback messages, and a computer program that connects the screening instrument with the relevant messages and generates the personal feedback. This program consisted mainly of logical (`if–then') statements (Kreuter et al., 2000).
The screening instrument
The screening instrument consisted of a written questionnaire. The first part of the questionnaire was developed to assess participants' knowledge of cancer symptoms [based on the Knowledge of Cancer Warning Signs Inventory (Berman and Wandersman, 1991)] and forms of early detection behavior (i.e. looking out for cancer symptoms, BSE, TSE, help-seeking behavior and participation in screening programs).
The second part of the screening instrument focused on determinants of early detection behaviors. For this part, the Attitude–Social influence–self-Efficacy (ASE) model (De Vries et al., 1995, 1988) served as a theoretical framework. This model has proven to be useful in explaining passive detection behavior and help-seeking behavior (De Nooijer et al., submitted). Attitudes were assessed by asking respondents about the advantages (e.g. important, useful, detection as soon as possible, certainty about health) and disadvantages (e.g. frightening, unpleasant, scary when something is detected) of looking out for cancer symptoms (passive detection) and seeking help. Anticipated regret and moral obligation were assessed for passive detection behavior and help-seeking behavior. Concerning social influence, the social norms of the partner, family, friends and the family physician with respect to looking out for and seeking help for cancer symptoms were assessed. Furthermore, possible modeling influences were assessed by asking how many people in the direct environment of the participant were attentive to cancer symptoms and were likely to seek help in the case of cancer symptoms. Self-efficacy was assessed using different concepts: the difficulty of looking out for and seeking help in the case of cancer symptoms, the ability to look out for and seek help in the case of cancer symptoms, and the ability to seek help in four different situations (when people doubt as to whether the symptom is a cancer symptom, when people expect the physician to think the complaint is not serious, when people do not expect the complaint to be serious and when people experience fear). We also assessed two general concepts that might influence passive detection and help-seeking behavior, i.e. chronic fear of cancer, based on a questionnaire by Jepson and Chaiken (Jepson and Chaiken, 1990) and fatalism.
We asked respondents about demographic variables (gender and age) to facilitate tailoring based on epidemiological data, about their level of education, marital status, cancer history and whether they knew someone with cancer in their direct environment, to provide information for the analyses.
The message source file
The message source file consisted of 319 feedback messages that could be combined into 1.8x1024 different tailored information letters. Each letter consisted of approximately five to eight pages. Positive feedback was given on correct baseline situations, to reinforce the states desired. For instance, respondents who were alert to a cancer symptom received information that it was good to be attentive to cancer symptoms. If the baseline situation could be improved, respondents received additional information.
The tailored letter consisted of nine sections. The first section included an introduction on the early detection of cancer to explain why it is important that cancer be detected at an early stage. Information about 5-year survival rates with and without metastasis was given for the most common types of cancer for both men and women. Furthermore, this section provided information about risk perception, because people might think that cancer is something that would never happen to them (De Nooijer, et al., 2001). For this risk perception, role-model stories had been included, adapted to various target groups (for women: a woman who talked about breast cancer; for young men: a young man who talked about testicular cancer; and for men aged 35+: a man who talked about cancer of the colon).
The second section stated the main goal of the letter, and contained feedback about knowledge of cancer symptoms and feedback about respondents' behavior (looking out for cancer symptoms and consulting a physician) for each separate cancer symptom. Feedback differed between respondents that correctly and incorrectly identified a symptom as a cancer symptom. Both groups received information on each specific symptom, to ensure a basic level of knowledge. Feedback on attentiveness to cancer symptoms was different for various levels of attentiveness (always, often/sometimes or seldom/never). The first two groups of respondents were encouraged to keep doing so, while the last group received information on the importance of being attentive to the specific symptom. Similar feedback was given for help-seeking behavior for each possible cancer symptom. A distinction was made in the feedback between people who sought medical help within the appropriate time and people that delayed seeking medical help (in accordance with recommendations from the Dutch Cancer Society) (Dutch Cancer Society, 1996).
The third, fourth and fifth sections discussed the determinants of passive detection behavior and help-seeking behavior, including attitudes, social influences and self-efficacy. For instance, respondents with low self-efficacy received information on how they could deal with different situations. These sections also addressed fear of cancer and a fatalistic attitude toward cancer. People who were extremely frightened received information to put this fear into perspective, whereas people who denied all fears or worries were warned that fear or worries can sometimes be useful. Feedback on existing fear is important, because negative affective responses might influence compliance (Millar, 1995). The information emphasized that early detection behaviors may also have positive sides, such as the relief of fear and more certainty about someone's health.
Sections six and seven were only sent to women, and discussed BSE and participation in screening programs for breast and cervical cancer. Feedback on BSE was provided by asking respondents about their intention to carry out BSE and included guidelines on the correct manner of self-examination. It also included tips on how to remember to carry out a BSE every month. Messages were framed negatively, since this was found to be effective in promoting BSE (Meyerowitz, 1987). An example of a negatively framed message was `If you do not carry out BSE, you could miss the opportunity of detecting cancer in its early stages'. Feedback was given to the target groups for breast and cervix screening programs based on the intention to participate the next time the female respondent received an invitation. The advantages of participation were stressed and explained to women who were to receive the first invitation within the 2 years following the research study, and they were encouraged to participate upon receipt of the invitation.
Section eight was only sent to men aged 18–35 and dealt with TSE. Feedback was given on their intention to carry out TSE. Since most of the respondents in this target group were unfamiliar with TSE, they all received information on how to carry out TSE. Messages on TSE were positively framed, because people with low involvement with the issue appeared to agree more often with its positive aspects (Maheswaran, 1990). For instance `If you carry out TSE, you will know whether your testicles are healthy'.
Section nine was the closing section, which included a reference to the Dutch Cancer Society for further questions about cancer.
|
Methods |
|---|
|
TopAbstractIntroductionInterventionMethodsResultsDiscussionReferences |
|---|
Study design and procedures
Data was obtained from Dutch adults by means of written questionnaires in 1999. A total of 1855 participants were recruited via local door-to-door newspapers and a national newspaper in the Netherlands. Participation was on a voluntary basis. Participants were randomly assigned to the tailored information group (n = 619), the general information group (n = 618) or a control group (n = 618). Participants received the baseline questionnaire (which is the screening questionnaire for the tailored information group), together with information about the study and an informed consent form. Three weeks after they returned the first questionnaire, participants received either the tailored information, the general information or no information (in the control group). Again, 3 weeks later, all participants received the second questionnaire.
The general information, which was a current brochure of the Dutch Cancer Society, dealt with such issues as the importance of early detection, fear of cancer, cancer symptoms, screening programs for breast and cervical cancer, early detection of skin cancer, and primary prevention of cancer (Dutch Cancer Society, 1991). The control group was not included in the analysis of the process evaluation and therefore not included in this study. Approval for the study had been obtained by the university's medical ethics commission.
The questionnaire
A written questionnaire was used for the evaluation of the intervention. Use was assessed by asking whether the information had been read, kept and discussed with others. These questions could be answered on a dichotomous scale with `yes' or `no'.
The format of the information was assessed using six questions (Cronbach's 
= 0.68), such as `Was the information attractively presented?' or `How understandable was the information?'. Answers were given on one-sided four-point scales, varying from `not at all' (1) to `very much' (4).
The content of the information was assessed using five questions (Cronbach's = 0.83), e.g. `After reading the letter/folder, did you have sufficient information on how you can contribute to early detection of cancer?'. Answers were given on one-sided four-point scales, varying from `insufficient' (1) to `very sufficient' (4).
The extent to which the information resulted in worries about a particular complaint or diagnosis was assessed using two questions: `After reading the letter/folder, were you worried about a particular complaint (diagnosis)?'. Answers were given on one-sided four-point scales, varying from `not at all worried' (1) to `very worried' (4).
One question measured the evaluation of the information in terms of a score on a 1 (`very bad') to 10 (`very good') scale.
For the tailored information group only, three additional questions were asked to evaluate the tailored information on BSE and TSE. These questions assessed whether the instructions on how to perform BSE or TSE were `very clear' (4) to `unclear' (1), whether respondents agreed with the importance of BSE or TSE [`very much' (4) to `not at all' (1)] and how they evaluated the information on BSE or TSE in terms of a score (1–10).
Attitudes toward early detection behaviors was assessed by asking about the advantages and disadvantages of passive detection and help-seeking behavior. Advantages were assessed using four items for passive detection (Cronbach's = 0.67) and four items for help-seeking behavior (Cronbach's = 0.58), e.g. `Do you think looking out for cancer symptoms is useful? [very useful (4) to not at all useful (1)]'. Disadvantages were assessed using three items for passive detection (Cronbach's = 0.73) and five items for help-seeking behavior (Cronbach's = 0.76), e.g. `Would you be hesitant about visiting a family physician if you detected a possible cancer symptom? [`very much' (4) to `not at all' (1)].
Passive detection behavior, i.e. becoming aware of cancer symptoms without having undertaken concrete action, was assessed using 14 questions regarding 14 cancer symptoms (i.e. blood in stool, coughing up blood, indigestion, swallowing problems, sores that do not heal, appearance of warts, appearance of moles, changes in a mole, thickenings or lumps, unusual weight loss, nagging cough or hoarseness, changes in bowel habits, blood in urine, urinary problems) for instance, `How often do you look out for...?'. Questions had to be answered on a five-point scale (`always' to `never').
Help-seeking behavior was assessed using 14 items concerning the 14 cancer symptoms, e.g. `After what period of time would you consult a physician if you detected the following cancer symptom?'. Answers were given on a six-point scale (ranging from `within 1 week' to `never'). Data was re-coded into `appropriate help-seeking behavior' (1) (i.e. seeking medical help within 1 week for blood in stool, coughing up blood, the appearance of warts or new moles or changes in a mole, thickenings or lumps, blood in urine and seeking medical help within 4 weeks for indigestion or swallowing problems, sores that do not heal, unusual weight loss, nagging cough or hoarseness, changes in bowel habits or urinary problems) or `inappropriate help-seeking behavior' (0) (all other answers) in accordance with recommendations from the Dutch Cancer Society (Dutch Cancer Society, 1996).
Statistical analysis
Analyses were carried out for all respondents who were not suffering from cancer at the time of data collection. Differences between the tailored information group and the general information group at baseline in demographic variables (gender, age, educational level, marital status, knowing someone with cancer, cancer history) were tested with 
2 tests or analyses of variance. Respondents' reactions to the information received and their opinion about the information were tested using 2 tests or t-tests for independent samples. Analyses of co-variance were performed to study the effects of the information on attitude and passive detection behavior 3 weeks after the intervention (T1) between the two study groups. The co-variate included was the attitude or the behavior prior to the intervention (assessed in the screening questionnaire at T0). Help-seeking behavior 3 weeks after the intervention was analyzed using 2 tests and logistic regression analysis, since this is a dichotomous variable (yes or no appropriate help seeking). Separate analyses were conducted on the tailored information group to study differences in subgroups in the evaluation of the information, using logistic and multiple linear regression analyses. All analyses were carried out with SPSS (P < 0.05).
|
Results |
|---|
|
TopAbstractIntroductionInterventionMethodsResultsDiscussionReferences |
|---|
Respondents
A total of 1189 screening questionnaires (96%) were completed and returned. Within 3 weeks all respondents that had returned the first questionnaire received either the tailored information or the general information, followed by the second questionnaire after 3 weeks. The response to the second questionnaire was 83% (n = 516) for the tailored information group and 85% (n = 524) for the general information group.
The majority of the respondents were female (80%). The average age was 46 years (SD = 13). Most respondents had a spouse or partner (76%), 24% were single or widowed. Thirty-five percent of the respondents had lower-level education (including primary school or basic vocational school), 36% had medium-level education (including secondary vocational school or high school) and 29% had higher-level education (including higher vocational school or university).
Ten percent of the respondents reported that they had had cancer at some time in their lives. Almost all respondents (92%) knew someone with cancer in their direct environment.
There were no differences between the tailored information group and the general information group in gender, age, educational level and marital status. In both study groups, respondents had the same extent of experience with cancer in their direct environment or had suffered from cancer at some time in their lives.
Evaluation of the information
Table I shows that the tailored information and the general information was read by the majority of the respondents. Respondents who received tailored information saved the information [2(982,1) = 14.12, P < 0.001] and discussed the information with others [2(983,1) = 18.37, P < 0.001] significantly more often compared to respondents who received general information.
|
In general, respondents were very positive about the format of both the tailored information and the general information (Table II
). t-tests showed significant differences in the total format score (t = 8.08, d.f. = 979, P < 0.001), and in the extent to which the information was interesting (t = 4.86 d.f. = 977, P < 0.001), new (t = 7.19, d.f. = 980, P < 0.001) and attractively presented (t = 6.97, d.f. = 969, P < 0.001); respondents of the tailored information group gave more positive scores on these aspects than respondents of the general information group. Moreover, respondents in the tailored information group felt significantly more personally addressed by the information they had received than respondents in the general information group (t = 7.85, d.f. = 980, P < 0.001).
|
With regard to the content of the information, the responses in the tailored information group were significantly more positive than those in the general information group concerning the total content score (t = 4.30, d.f. = 972, P < 0.001), and the advantages (t = 4.61, d.f. = 973, P < 0.001) and disadvantages (t = 3.71, d.f. = 963, P < 0.001) of looking out for cancer symptoms. In addition, respondents that had read the tailored information knew more about what one can do to contribute to the early detection of cancer than those that had read the general information (t = 4.53, d.f. = 978, P < 0.001). Neither the tailored information nor the general information resulted in worries about a particular complaint or diagnosis. Overall, the tailored information was evaluated more positively than the general information (8 versus 7) (t = 4.73, d.f. = 973, P < 0.001).
The tailored information addressed BSE and TSE, while this was not specifically addressed in the general information. Therefore, only recipients of the tailored information were asked to answer an additional set of questions regarding the tailored information on BSE and TSE. The female respondents reported that the instructions in the tailored information on how to carry out a BSE were clear (mean score 3.14, SD = 0.65). Moreover, they reported understanding the importance of BSE after reading the tailored information (mean score 3.13, SD = 0.73). On average, the tailored information on BSE was awarded a score of 8 (SD = 1.15) by the target group.
The male population younger than 35 was asked to answer questions regarding TSE, since they are the target group for TSE. The number of respondents in the high-risk group for testicular cancer was very limited, only 14 respondents received the information on TSE. About half of these 14 respondents said that the instructions on how to perform a TSE were clear, the other half reported that the instructions were not so clear (mean score 2.57, SD = 0.85). Most respondents (n = 9) agreed with the importance of TSE (mean score 2.64, SD = 0.63). This part of the tailored information was awarded a score of 7 (SD = 1.49) in the evaluation.
Influence of information on attitudes, passive detection behavior and help-seeking behavior
Three weeks after the intervention, the effects of tailored versus general information on attitudes toward passive detection and the effects of passive detection itself were tested using analyses of co-variance. Attitude and behavior at T0 were included as co-variates to adjust for pre-existing attitudes and behaviors. Table III shows that the tailored information had a significant effect on the perception of the advantages of passive detection. Furthermore, significant effects were found for passive detection of all cancer symptoms. This indicates that recipients of the tailored information perceived more advantages of passive detection and paid significantly more attention to cancer symptoms at T1 than recipients of the general information.
|
Co-variance analysis was carried out on attitudes toward help seeking and
2 tests were performed to test differences in help-seeking behavior at T1 (Table IV). At T1, the tailored information was found to have an effect on the perception of the advantages and disadvantages of help seeking for cancer symptoms, as well as on actual help seeking regarding the majority of cancer symptoms. The recipients of the tailored intervention had more positive attitudes toward help seeking and were more likely to respond appropriately to the detection of these particular cancer symptoms than recipients of the general information. Additionally, logistic regression analyses were performed to adjust for pre-existing scores on help-seeking behavior. Results were similar for help seeking for all cancer symptoms, with the exception of help-seeking behavior for thickenings or lumps, which did not differ between respondents of the tailored information group and respondents of the general information group when using 2 tests, but did differ when using logistic regression (OR = 1.44, 95% CI = 1.054–1.980, P < 0.05).
|
Differences among subgroups in the evaluation of the tailored information
Separate analyses were performed in the tailored information group to test whether various subgroups, based on demographic differences or behavioral differences, influenced the percentage of readers that actually used the material, the evaluation of the format and the evaluation of the content.
Logistic regression analysis was carried out to assess the influence of demographics and behavior on reading the information, saving the information and discussing the information with others. The tailored information was read by older respondents more often (OR = 1.04, 95% CI = 1.009–1.080, P < 0.05) and discussed by older respondents more often (OR = 0.98, 95% CI = 0.967–0.996, P < 0.05) than by younger respondents. The information was discussed more often by respondents with a partner (OR = 1.87, 95% CI = 1.225–2.873, P < 0.01). Other subgroups did not differ in terms of level of use.
Multiple regression analysis was carried out in order to see whether demographic variables and early detection behaviors influenced the evaluation of the tailored information with respect to the format. No significant factors could be identified that were related to differences in the evaluations of the format of the tailored intervention.
Multiple regression analysis was also carried out to indicate which variables were significantly related to the evaluation of the content of the tailored information. Women (ß = 0.125, P < 0.01), older respondents (ß = 0.179, P < 0.001), respondents who knew someone with cancer in their direct environment (ß = 0.133, P < 0.01), respondents who were attentive to cancer symptoms (ß = 0.097, P < 0.05) and respondents who sought help within the appropriate time (ß = 0.167, P < 0.001) were significantly more positive about the content of the tailored information.
|
Discussion |
|---|
|
TopAbstractIntroductionInterventionMethodsResultsDiscussionReferences |
|---|
This paper has described the process evaluation of a newly developed tailored intervention designed to stimulate people to behave in a manner which promotes the early detection of cancer. Our first hypothesis dealt with the evaluation of the information. With respect to use, both types of information were read by the majority of the recipients, but the tailored information was saved and discussed with others more often than the general information. This result was also found in previous studies of others (Campbell, 1994; Skinner, 1994; Brug et al., 1996
; De Bourdeaudhuij and Brug, 2000) and confirmed in a review on tailored printed communications (Skinner et al., 1999). This high level of use provides an important basis in McGuire's assumption that exposure to the message and attention to the message are the first steps for communication to be effective (McGuire, 1989). Further, an active reaction to the information, such as saving and discussing, suggests that the recipient thinks and talks about the information, which is more likely to lead to a greater elaboration of the information (Kreuter et al., 2000).
Additionally, the tailored information was more positively evaluated than the general information with respect to aspects of format, e.g. it made respondents feel personally addressed, and was considered interesting, new and attractively presented. With respect to aspects of content of the tailored information, the information on the advantages and disadvantages of looking out for cancer symptoms and on what respondents can do to contribute to early detection was viewed as being more sufficient by the recipients of the tailored information. Also other studies found that tailored information was more interesting, personally relevant and new compared to general information (Brug et al., 1998; De Bourdeaudhuij and Brug, 2000). Personal relevance and credibility were two factors consistently identified as better in the tailored information compared to the non-tailored information (Skinner et al., 1999). Credibility in our study was not found to be significantly better. This might be explained by the fact that the non-tailored information was an original brochure of the Dutch Cancer Society, which is considered to be a highly credible organization in the Netherlands.
The second hypothesis addressed increased worries about complaints or diagnosis as a result of the tailored information. No differences were found between the tailored and the general information regarding worries about a particular complaint or diagnosis, which is consistent with other studies (Best et al., 1996; Weist, 1996; Mansson et al., 1999). This is probably due to the fact that both the tailored information and the general information stressed that fear can be reduced by actively responding to newly detected symptoms and that a physician may help relieve the fear away.
Our third hypothesis stated that, more than general information, tailored information influences changes in attitudes toward passive detection and help seeking, as well as passive detection and help-seeking behaviors. Regarding attitude aspects, this was true for three out of four attitude concepts. This indicates that the tailored information, which was based on the ASE model and thus addressed the individual's present attitude, was able to establish more positive attitudes. The ASE model seems a helpful model for developing feedback messages. Furthermore, the effects of the information on passive detection behaviors in particular, but also on the majority of seeking help for cancer symptoms, were due to the tailored information. This may be explained by the fact that the respondents of the tailored information felt significantly more personally addressed than other respondents. According to Petty and Cacioppo, feeling personally addressed is likely to trigger someone's motivation to thoughtfully consider and process the information (Petty and Cacioppo, 1986). Moreover, personally relevant information is more likely to result in changes in attitude and behavior. Compared to the respondents in the general information group, the respondents in the tailored information group also reported that the advantages and disadvantages were addressed more sufficiently, which may have resulted in more favorable attitudes toward passive detection behavior.
We also hypothesized that there would be no differences between subgroups within the tailored information group regarding aspects of use, format and content. Indeed, we found that there were no differences between subgroups in the evaluation of the format aspects. This indicates that all respondents were positive about format aspects, which was considered to address respondents personally and offer understandable and interesting information, independent of demographic or behavioral differences. In other words, the format of the tailored information was sufficiently adapted to the characteristics of the target group.
Regarding use and content, we did find a number of differences in the evaluation between subgroups. Older respondents were more likely to read and discuss the tailored information. This might be due to the fact that cancer is more likely to occur in older people than in younger people, and thus information on cancer may be more interesting and personally relevant to older respondents. The differences between subgroups in the evaluation of the content might be explained by the fact that the form of communication of the information was tailored to the individual level, i.e. personally relevant to the recipient, rather than the content. An example of the first is that the information on the possible cancer symptoms was adjusted to prior knowledge, while in the latter all respondents received information on the possible cancer symptoms. Alternatively, tailored information might be effective when it fits between behavioral characteristics and the content of the message, but may not work when the tailored information is not well-matched with the characteristics of the recipient, as suggested by Kreuter et al. (Kreuter et al., 2000).
The specific section on TSE was evaluated only by very few people in the target group, as a result of which it is not possible to draw conclusions with confidence. An explanation for the less positive evaluation of the TSE part might be that young and male respondents gave lower scores in their evaluations on several other aspects as well. This was found in the results of the multiple regression analysis on the evaluation of the general part of the tailored information, where men and younger people were less positive. For TSE, the target group consisted of the young, male respondents, which is also the group that evaluated the information less positively. It is possible that this subgroup does not like the information provided (content aspects), that they do not like the format of the information (format aspects) and that they are not receptive to health information about early detection of cancer at all. These issues should be further studied.
A limitation of the study is the low percentage of male participants. Therefore, the results can be less easily generalized to men than women, among those who are interested in information on early detection of cancer. Young men, the high-risk population for testicular cancer, were under represented in our study sample. Future studies should pay attention to the selection of the study group and possibly to the dissemination of the health education information. Further research should explore how to interest this group in health education about early detection of cancer or about health issues in general and which channels should be used to reach them.
In summary, tailored information seems to be a highly appreciated form of communication about early detection of cancer, both in format and in content. Tailored information also contributes effectively to people's attitudes and early detection behavior compared to general information. Further research is needed to explore the impact of tailored information on behavior concerning the early detection of cancer in the long term.
|
Acknowledgments |
|---|
This study was supported by a grant from the Dutch Cancer Society.
|
References |
|---|
|
TopAbstractIntroductionInterventionMethodsResultsDiscussionReferences |
|---|
American Cancer Society (2000) Cancer Facts and Figures 2000. American Cancer Society, Atlanta, GA.
Berman, S. H. and Wandersman, A. (1991) Measuring knowledge of cancer. Social Science and Medicine, 32, 1245–1255.
Best, D. L., Davis, S. W., Vaz, R. M. and Kaiser, M. (1996) Testicular cancer education: a comparison of teaching methods. American Journal of Health Behavior, 20, 229–241.
Bostick, R. M., Sprafka, J. M., Virnig, B. A. and Potter, J. D. (1993) Knowledge, attitudes, and personal practices regarding prevention and early detection of cancer. Preventive Medicine, 22, 65–85.[Web of Science][Medline]
Brug, J., Glanz, K., Van Assema, P., Kok, G. and Van Breukelen, G. J. P. (1998) The impact of computer-tailored feedback and iterative feedback on fat, fruit, and vegetable intake. Health Education and Behavior, 25, 517–531.
Brug, J., Campbell, M. and Assema, P. (1999) The application and impact of computer-generated personalized nutrition education: a review of the literature. Patient Education and Counseling, 36, 145–156.[Web of Science][Medline]
Brug, J., Steenhuis, I. H. M., Van Assema, P. and De Vries, H. (1996) The impact of a computer-tailored nutrition intervention. Preventive Medicine, 25, 236–242.[Web of Science][Medline]
Budden, L. (1995) Young women's breast self-examination knowledge and practice. Journal of Community Health Nursing, 12, 23–32.[Medline]
Bull, F. C., Kreuter, M. W. and Scharff, D. P. (1999) Effects of tailored, personalized and general health messages on physical activity. Patient Education and Counseling, 36, 181–192.[Web of Science][Medline]
Campbell, M. K., DeVellis, B. M., Strecher, V. J., Ammerman, A. S., DeVellis, R. F. and Sandler, R. S. (1994) Improving dietary behavior: the effectiveness of tailored messages in primary care settings. American Journal of Public Health, 84, 783–787.
Cope, D. G. (1992) Self-esteem and the practice of breast self-examination. Western Journal of Nursing Research, 14, 618–631.
De Bourdeaudhuij, I. and Brug, J. (2000) Tailoring dietary feedback to reduce fat intake: an intervention at the family level. Health Education Research, 15, 449–462.
De Nooijer, J., Lechner, L. and De Vries, H. (2001) A qualitative study on detecting cancer symptoms and seeking medical help: an application of Andersen's model of total patient delay. Patient Education and Counseling, 42, 145–157.[Web of Science][Medline]
De Vries, H., Backbier, E., Kok, G. J. and Dijkstra, M. (1995) Measuring the impact of social influences on smoking onset in a longitudinal study: an integration of social psychological approaches. Journal of Applied Social Psychology, 25, 237–257.[Web of Science]
De Vries, H. and Brug, J. (1999) Computer-tailored interventions motivating people to adopt health promoting behaviours: introduction to a new approach. Patient Education and Counseling, 36, 99–105.[Web of Science][Medline]
De Vries, H., Dijkstra, M. and Kuhlman, P. (1988) Self-efficacy: the third factor besides attitude and subjective norm as a predictor of behavioral intentions. Health Education Research, 3, 273–282.
Deliens, L. (1998) Sociale Determinanten van Kennis van Kankerpreventie in Vlaanderen [Social Determinants of Knowledge of Cancer Prevention], Universiteit Maastricht, Maastricht.
Dijkstra, A. and De Vries, H. (1999) The development of computer-generated tailored interventions. Patient Education and Counseling, 36, 193–203.[Web of Science][Medline]
Dijkstra, A., De Vries, H. and Roijackers, J. (1999) Targeting smokers with low readiness to change with tailored and nontailored self-help materials. Preventive Medicine, 28, 203–211.[Web of Science][Medline]
Dijkstra, A., De Vries, H. and Roijackers, J. (1998) Long-term effectiveness of computer-generated tailored feedback in smoking cessation. Health Education Research, 13, 207–214.
Dutch Cancer Society (1991) Wees Wijzer, van Doorlopen met Klachten Wordt U Zeker niet Beter [Be wise: Living with Complaints won't Make You any Better]. Dutch Cancer Society, Amsterdam.
Dutch Cancer Society (1996) Vroege Ontdekking van Kanker...Wat U Zelf Kunt Doen [Early Detection of Cancer...What You Can Do Yourself]. Dutch Cancer Society, Amsterdam.
Dutch Cancer Society (1998) Mijnheer, Sommige Kleinigheden Kunt U Maar Beter Niet Negeren [You Would Do Better Not To Ignore Some Lower Parts, Sir]. Dutch Cancer Society, Amsterdam.
Eagly, A. H. and Chaiken, S. (1990) The Psychology of Attitudes. Hartcourt, Fort Worth, TX.
Friedman, L. C., Nelson, D. V., Webb, J. A., Hoffman, L. P. and Baer, P. E. (1994) Dispositional optimism, self-efficacy, and health beliefs as predictors of breast self-examination. American Journal of Preventive Medicine, 10, 130–135.[Web of Science][Medline]
Friman, P. C., Finney, J. W., Glasscock, S. G., Weigel, J. W. and Christophersen, E. R. (1986) Testicular self-examination: validation of a training strategy for early cancer detection. Journal of Applied Behavior Analysis, 19, 87–92.[Web of Science][Medline]
Jepson, C. and Chaiken, S. (1990) Chronic issue-specific fear inhibits systematic processing of persuasive communications. Journal of Social Behavior and Personality, 5, 61–84.
Kreuter, M., Farrell, D., Olevitch, L. and Brennan, L. (2000) Tailoring Health Messages. Customizing Communication with Computer Technology. Lawrence Erlbaum, Mahwah, NJ.
Kreuter, M. W., Bull, F. C., Clark, E. M. and Oswald, D. L. (1999) Understanding how people process health information: A comparison of tailored and non-tailored weight-loss materials. Health Psychology, 18, 487–494.[Web of Science][Medline]
Kreuter, M. W., Oswald, D. L., Bull, F. C. and Clark, E. M. (2000) Are tailored health education materials always more effective than non-tailored materials? Health Education Research, 15, 305–315.
Kreuter, M. W. and Skinner, C. S. (2000) Tailoring: what's in a name? Health Education Research, 15, 1–4.
Lechner, L., De Nooijer, J. and De Vries, H. (1998) Vroege Ontdekking van Kanker: Een Gedrags-en Determinantenanalyse onder de Nederlandse Bevolking [Early Detection of Cancer: A Behavior and Determinant Analysis among the Dutch Population]. Universiteit Maastricht, Maastricht.
Lechner, L. and De Vries, H. (1996) Vroege Ontdekking van Kanker [Early Detection of Cancer]. Maastricht University, Maastricht.
Lechner, L., De Vries, H. and Offermans, N. (1997) Participation in a breast cancer screening program: Influence of past behavior and determinants on future screening participation. Preventive Medicine, 26, 473–482.[Web of Science][Medline]
Maheswaran, D. and Meyers-Levy, J. (1990) The influence of message framing and issue involvement. Journal of Marketing Research, 27, 361–367.
Mansson, J., Marklunc, B., Bengtsson, C. and Fridlund, B. (1999) Evaluation of an educational programme for the early detection of cancer. Patient Education and Counseling, 37, 231–242.[Web of Science][Medline]
McGuire, W. J. (1989) Theoretical foundations of campaigns. In Atkin, E. andAtkin, C. K. (eds), Public Communication Campaigns. Sage, Newbury Park, CA, pp. 43–67.
Meyerowitz, B. E. and Chaiken, S. (1987) The effect of message framing on breast self-examination attitudes, intentions, and behavior. Journal of Personality and Social Psychology, 52, 500–510.[Web of Science][Medline]
Millar, M. G. and Millar, K. (1995) Negative affective consequences of thinking about disease detection behaviors. Health Psychology, 14, 141–146.[Web of Science][Medline]
Persson, K., Svensson, P. and Ek, A. C. (1997) Breast self-examination: an analysis of self-reported practice. Journal of Advanced Nursing, 25, 886–892.[Web of Science][Medline]
Petty, R. E. and Cacioppo, J. T. (1986) The elaboration likelihood model of persuasion. In Berkowitz, L. (ed.), Advances in Experimental Social Psychology. Academic Press, London, vol. 19.
Rakowski, W., Ehrich, B., Goldstein, M., Rimer, B. K., Pearlman, D. N., Clark, M. A., Velicer, W. F. and Woolverton, H. (1998) Increasing mammography among women aged 40–74 by use of a stage-matched, tailored intervention. Preventive Medicine, 27, 748–756.[Web of Science][Medline]
Ruwaard, D. and Kramers, P. G. N. (1997) Volksgezondheid Toekomst Verkenning: De Gezondheidstoestand van de Nederlandse Bevolking in de Periode 1950–2010 [Public health Status and Forecasts: The Health Status of the Dutch Population over the Period 1950–2010]. Elsevier/De Tijdstroom, Den Haag.
Sheik, I. and Ogden, J. (1998) The role of knowledge and beliefs in help seeking behaviour for cancer: a quantitative and qualitative approach. Patient Education and Counseling, 35, 35–42.[Web of Science][Medline]
Skinner, C. S., Campbell, M. K., Rimer, B. K., Curry, S. and Prochaska, J. O. (1999) How effective is tailored print communication? Annals of Behavioral Medicine, 21, 290–298.[Web of Science][Medline]
Skinner, C. S., Strecher, V. J. and Hospers, H. (1994) Physicians' recommendations for mammography: do tailored messages make a difference? American Journal of Public Health, 84, 43–48.
Strecher, V. J., Kreuter, M. W., Boer, D. D., Kobrin, S. C., Hospers, H. J. and Skinner, C. S. (1994) The effects of computer tailored smoking cessation messages in family practice settings. Journal of Family Practice, 39, 262–270.
Tabar, L., Fagerber, C. J., Gad, A., Baldetorp, L., Holmberg, L. H., Grontoft, O., Ljungquist, U., Lundstrom, B., Manson, J. C. and Eklund, G. (1985) Reduction in mortality from breast cancer after mass screening with mammography. Randomised trial from the Breast Cancer Screening Working Group of the Swedish National Board of Health and Welfare. The Lancet, i, 829–832.
Velicer, W. F. and Prochaska, J. O. (1999) An expert intervention for smoking cessation. Patient Education and Counseling, 36, 119–129.[Web of Science][Medline]
Visser, O., Coebergh, J. W. W., Schouten, L. J. and Van Dijck, J. A. A. M. (2000) Incidence of Cancer in the Netherlands 1996. Comprehensive Cancer Centers, Utrecht.
Wardle, J., Steptoe, A., Smith, H., Groll-Knapp, E., Koller, M., Smith, D. and Brodziak, A. (1995) Breast self-examination: Attitudes and practices among young women in Europe. European Journal of Cancer Prevention, 4, 61–68.[Web of Science][Medline]
Weist, M. D. and Finney, J. W. (1996) Training in early cancer detection and anxiety in adolescent males: a preliminary report. Developmental and Behavioral Pediatrics, 17, 98–99.
Received on October 28, 2000; accepted on May 9, 2001
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  R. Tobias, A. Brugger, and H.-J. Mosler Developing Strategies for Waste Reduction by Means of Tailored Interventions in Santiago de Cuba Environment and Behavior, November 1, 2009; 41(6): 836 - 865. [Abstract] [PDF]
|
|
|
|
 |
|
 C.C. Burgess, A.M. Bish, H.S. Hunter, P. Salkovskis, M. Michell, P. Whelehan, and A.J. Ramirez Promoting early presentation of breast cancer: development of a psycho-educational intervention Chronic Illness, March 1, 2008; 4(1): 13 - 27. [Abstract] [PDF]
|
|
|
|
 |
|
 J. de Nooijer, L. Lechner, M. Candel, and H. de Vries A Randomized Controlled Study of Short-term and Long-term Effects of Tailored Information versus General Information on Intention and Behavior Related to Early Detection of Cancer Cancer Epidemiol. Biomarkers Prev., November 1, 2002; 11(11): 1489 - 1491. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||