Health Education Research Advance Access originally published online on April 2, 2008
Health Education Research 2008 23(3):440-453; doi:10.1093/her/cyn011
Health information styles among participants in a prostate cancer screening informed decision-making intervention
Health Communication Program, RTI International, Research Triangle Park, NC 27709, USA
* Correspondence to: Pamela A. Williams-Piehota. Health Communication Program, RTI International, 3040 Cornwallis Road, Research Triangle Park, NC 27709, USA. E-mail: ppiehota{at}rti.org
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionConflict of interest statementAcknowledgementsReferences |
|---|
The purpose of this study was to assess the usefulness of a health information styles segmentation strategy in understanding audience subgroups. We examined the health information styles of men participating in a prostate cancer screening informed decision-making (IDM) intervention and assessed intervention effects on men with distinct health information styles. We classified participants into three health information style groups based on their independence in health decision making (independent versus doctor dependent) and engagement in health enhancement (active versus passive): independent active (IA), doctor-dependent active (DDA) and passive. We developed profiles of men in these groups: IAs tended to be white and more highly educated and to have greater baseline prostate cancer knowledge; DDAs were older, less healthy and more likely to have visited a doctor in the past year and passives tended to be younger, not to have had a recent prostate-specific antigen test and to have lower self-efficacy related to communication with doctors and less positive interactions with doctors. All groups significantly increased their prostate cancer knowledge after the intervention, but passives exhibited the greatest increase in knowledge at a 6-month follow-up. The health information styles segmentation strategy used in this study offers a viable framework for segmenting audiences.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflict of interest statementAcknowledgementsReferences |
|---|
There has been a trend toward greater patient involvement in decisions about their health. Wider access to consumer health information, especially through the Internet, and increased recognition that in many cases clinical decisions are not ‘one size fits all’ have contributed to the increased emphasis on patients’ informed decision making (IDM) [1–3]. IDM is particularly important in the face of uncertainty about optimal screening and treatment choices, which is the case for prostate cancer. Prostate-specific antigen (PSA) screening is controversial because of uncertainty about whether it reduces mortality and whether the potential benefits outweigh the harms, which include unnecessary worry and the side effects of common treatments. The U.S. Preventive Services Task Force concluded that the evidence is insufficient to recommend for or against routine PSA screening [4, 5]. Given the uncertainties about PSA screening, many medical associations recommend an IDM process so that men can make well-informed decisions that reflect their personal values and preferences [5–7].
IDM involves patients obtaining up-to-date and scientifically accurate health information in a timely fashion and in a form they can understand. In order to reach different audiences and optimize their understanding of health information, researchers have examined various individual needs and preferences related to health information, including preferred channels [8], media usage [9], trusted sources [10], types of information sought [10, 11] and barriers to health information seeking [12, 13].
Given the well-recognized importance of audience segmentation for health education and communication interventions, such as IDM interventions, it is important to identify meaningful audience segments based on shared characteristics generally associated with the outcomes of interest [14, 15]. Audience segmentation strategies allow health educators and others to tailor or target health information to the identified needs and preferences of individuals or specific subgroups to make them more effective [16, 17]. Message tailoring has been identified as one method of influencing various health behaviors (e.g. dietary behavior [18] and mammography utilization [19]). The effectiveness of tailored messages is believed to be due, in part, to the increased scrutiny that recipients give these messages [20], which thereby increases the likelihood of a subsequent behavior change [21].
One type of audience segmentation strategy is to develop typologies of people [15], such as typologies based on health preferences, styles and behaviors (e.g. [22]). For example, Miller et al. [23] identified dispositional differences in the way individuals seek health information and in the way they respond cognitively and emotionally to potentially threatening health information. As another example, Shim et al. [24] used the National Cancer Institute's Health Information National Trends Survey (HINTS) data to develop a typology based on individuals’ cancer-related scanning (i.e. gathering information incidentally from the media and other environmental sources) and seeking behaviors. Both scanning and seeking were significantly related to cancer knowledge and PSA screening.
Maibach et al. [25] developed a typology that is particularly germane to IDM because it taps into patients’ interaction styles with health care providers, health decision-making preferences and health information preferences. Information tailored to these preferences should be especially useful for the patient and result in improved health-related decision making. Maibach conducted a segmentation analysis of the U.S. adult population to define meaningful categories of consumers in terms of their orientation to health information and decision making. The study resulted in the development and validation of a brief screening instrument that identified four distinct groups based on differences in degree of engagement in health enhancement (active versus passive) and degree of independence in health decision making (independent versus doctor dependent): independent active (IA), doctor-dependent active (DDA), independent passive (IP) and doctor-dependent passive (DDP). Both IAs and DDAs value health information. Yet, these groups are distinguished from one another in that IAs have high self-efficacy for understanding health information and largely prefer to make health decisions for themselves, while DDAs have lower self-efficacy and are more likely to depend on their doctor to make decisions about their health. Both IPs and DDPs tend not to seek health information. However, DDPs are more receptive to health information from their doctors than IPs, who have little communication with their doctor [25]. Audience segmentation according to such health information styles could guide the development of targeted health education and communication interventions that will effectively reach and resonate with these distinct audiences.
This paper expands on research of Maibach et al. by applying the health information styles segmentation strategy to a sample of men participating in a community-based, prostate cancer screening IDM intervention to determine its relevance and usefulness with this sample. We examine the profiles of men with distinct health information styles in terms of their sociodemographic characteristics; knowledge, attitudes and behaviors related to prostate cancer; interactions with health care providers; and self-efficacy related to communication with providers. In addition, we explore the effects of the intervention among men with differing health information styles. This study contributes to our understanding of how populations differ in terms of their orientation to health information and health decision making so that health communication messages can be targeted appropriately and provides insight into the effects of an IDM intervention on participants with differing health information styles.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflict of interest statementAcknowledgementsReferences |
|---|
Study design and procedures
Designed to provide men with the information and skills they need to make informed decisions about PSA screening and to promote discussion and shared decision making with their physicians, the
45-min intervention involved an oral presentation by a physician followed by a question and answer session, a 20-min video and print materials, including a 3-fold brochure, a 4 x 6-inch poster and a shirt-pocket decision aid. The following key intervention messages reflected the clinical evidence available at the time the interventions were developed (September 2004) [4]. (i) There are two types of prostate cancers: slow growing and fast growing. (ii) A problem with the PSA test is that it leads some men with a slow-growing prostate cancer to get treatment that they do not need. (iii) About half of all men who get treatment for prostate cancer will have permanent side effects. (iv) Men should decide whether they feel the PSA test is right for them and talk with their doctors. The video and print materials were developed for this project using a systematic formative research process. The video conveys the experiences of four men who make different decisions about PSA screening. The men explain why they made their decision, the consequences of their decision (including treatment side effects) and their thoughts about their decision after some time had passed. The video also models PSA discussions with a physician. The print materials contain information about the PSA test and treatment, potential negative side effects of treatment and issues to consider in deciding about screening. The intervention components were delivered as a package; the individual pieces were not intended to be freestanding, particularly given the importance of the interactive exchange between the physician and participants in the community settings. Using a snowball sampling approach, we recruited community-based organizations (e.g. faith based, fraternal, fitness and recreation and senior centers) in two North Carolina metropolitan areas that had the capacity to host the intervention sessions. The organizations received $250 for each session they hosted. Community sites advertised the intervention sessions through informal channels and invited members of their organizations and the community to attend the sessions. Most participants were organizational members. Thus, we delivered the interventions to a convenience sample of men in the organizations, with a total of 20 sessions held between September 2004 and February 2005 (with sessions ranging in size from 10 to 30 participants each).
We administered the baseline survey as men gathered at the organizations for the intervention session. We administered a follow-up survey by mail so that participants received the questionnaire 6 months after attending the intervention session. Non-respondents received telephone reminder calls and were given the option of completing the survey by telephone. Participants who completed the 6-month survey were compensated with $10.
The full study design involved a comparison of two versions of the PSA screening IDM intervention—one presenting prostate cancer only and the other presenting prostate cancer information framed in the context of broader men's health issues. The focus of this study was on the health information styles rather than how the PSA message was framed; thus, we combined men in the two intervention groups. Research findings not touching on health information styles, but instead comparing the effects of the two versions of the IDM intervention on PSA knowledge and screening decisions, appear separately (L. McCormack C. M. Bann, P. Williams-Piehota, D. Driscoll, C. Soloe, J. Poehlman, T. -M. Kuo, K. N. Lohr, S. L. Sheridan, C. E. Golin, R. Harris and S. Cykert: in preparation).
Participants
The sample for this study was 319 men who participated in the intervention and for whom we obtained baseline data about health information style. (A total of 361 men completed the baseline survey; however, 42 men were excluded from this study because health information style data were incomplete.) To be eligible for the study, men had to be between 40 and 80 years of age and not previously diagnosed with prostate cancer. Of the men who completed the baseline survey, 78% (248/319) completed the 6-month survey.
We compared the characteristics of men who completed the 6-month survey with the characteristics of men who had dropped out of the study. Health information styles did not differ significantly, but a few sociodemographic differences emerged between the men who did and did not complete the 6-month survey. Men who dropped out were younger [mean (standard deviation) age: 55 (11) versus 63 (11), P < 0.01], less educated (high school or less: 25 versus 13%, P < 0.05), more likely to be black (41 versus 27%, P < 0.05) and less likely to have a personal doctor (76 versus 88%, P < 0.01) than men who completed the 6-month survey.
Measures
Health information style group
The baseline survey included 10 items that make up the Porter Novelli© Health Information Seeking Segmentation Scale (Table I) [25]. The constructs addressed in the segmentation scale include self-efficacy for health information, prevention orientation, doctor dependence, proactive style with doctor and perceived importance of health information. Validity testing with a nationally representative sample of the US adult population demonstrated strong construct and criterion validity [25]. Cronbach's alpha for the scale was 0.71. Based on responses to the series of 10 questions that address these constructs, participants can be classified into four mutually exclusive groups that describe their health information style—IA, DDA, IP and DDP [25]. Of the 319 men surveyed, 41% were classified as IA and 47% as DDA. Because of the small numbers of participants categorized as IP and DDP, these two groups were combined into a single passive category (12%) for the purposes of analyses. Table I illustrates how individuals falling into each of the health information style groups rated (high or low) in terms of their self-efficacy for health information, prevention orientation, doctor dependence, proactive style with doctor and perceived importance of health information. In the current study, health information style is used as both a predictor and an outcome. When we examine intervention effects on knowledge, health information style is used as a predictor variable. We also examine predictors of health information style, in which case health information style is the outcome variable.
|
Prostate cancer knowledge
We measured knowledge at baseline and 6-month follow-up and assessed change in knowledge over time. The knowledge items used for this study measured participants’ answers to 10 questions about the following pieces of information based on conclusions drawn from the U.S. Preventive Services Task Force [4]: (i) understanding of key intervention messages: most prostate cancers are slow growing; (ii) men can live long, normal lives with slow-growing prostate cancer; (iii) men can have prostate cancer even if they have a normal PSA level; (iv) a PSA test can find both slow- and fast-growing prostate cancers; (v) a high PSA test can be caused by prostate cancer and/or an enlarged prostate; (vi) doctors agree that men over the age of 70 do not need a PSA test; (vii) there is uncertainty about which types of treatment work best; (viii) treatment side effects are fairly common; (ix) common side effects and (x) men are more likely to die from heart attack and stroke than prostate cancer. We coded responses to each of the above items as correct or incorrect (‘don’t know’ responses were coded as incorrect) and computed a knowledge index score that ranged from 0 to 10. The Cronbach's alpha coefficient for this index was 0.68 at baseline and 0.70 at the 6-month follow-up.
Beyond the two key measures explained above, sociodemographics, attitudes, health behaviors and health status were also measured. A number of attitudinal measures were included. We tapped into perceived risk with the question ‘How likely do you think it is that you will develop prostate cancer in the future? Would you say your chance of getting prostate cancer in your lifetime is ... ’ (1 = ‘very low’; 5 = ‘very high’). The extent of agreement with the statement ‘If my health worsens, it's a matter of fate’ (1 = strongly agree; 5 = strongly disagree) was used to assess fatalistic attitudes. Plans to get a PSA test in the future were measured with the item ‘Do you plan to get a PSA test in the next 12 months?’ (yes/no). To assess men's preferences related to participation in the PSA screening decision—from more doctor dependent to more autonomous—we adapted a measure previously developed and tested by Degner et al. [26–28]. We asked men ‘How much would you like to be involved in the decision about whether or not to get a PSA test?’ with response options of leave all decisions to my doctor/doctor make the final decision but seriously consider my opinion/doctor and I share responsibility for the decision/I make the final decision after seriously considering my doctor's opinion/I make the final decision. Health behavior measures included ‘How much attention do you usually pay to information about health or medical topics?’ (1 = none; 2 = a little/some; 3 = a lot) and ‘When did you have your most recent PSA test to check for prostate cancer?’ with response options of <6 months ago/6–12 months ago/1–3 years ago/>3 years ago/I never had one. Health status was assessed with the question ‘In general, would you say your health is ... ’ with responses rated on a five-point scale (1= ‘excellent’; 5= ‘poor’).
Data analysis
Descriptive analyses
We conducted chi-square tests and t-tests to examine the characteristics of the sample overall and to test for significant differences among the three health information style groups—IA, DDA and passives. We examined differences in sociodemographics, knowledge about PSA screening, attitudes and health behaviors. Then, we assessed differences among the groups in their reactions to the information conveyed during the presentation.
Multivariate analyses
To determine the relationship between sociodemographic and other variables and the health information groups, we conducted a multinomial logistic regression model [29] with the variables from the bivariate analyses (i.e. chi-square tests and t-tests) as predictors of health information group. For these analyses, the passive group was used as the reference category (i.e. the IA and DDA groups were each compared with the passive group). By including all the variables in the same model, we could determine which variables remain significant when accounting for other possible predictors.
Next, we explored differences among the health information style groups in prostate cancer knowledge and changes in knowledge over time (using the 10-item knowledge index). We used a generalized linear model to examine predictors of knowledge at baseline and 6-month follow-up. We interacted time and health information style group in the model to assess whether changes in knowledge over time differed across the three groups.
We included nearly identical predictors in our two multivariate analyses. In the analysis predicting health information style, we included variables that prior research has shown to be associated with seeking cancer information to see if these relationships were evident in our sample and for the segmentation strategy examined in this study, including health status, having a personal doctor and perceived risk [30–32]. In the analysis using health information style as a predictor of change in knowledge, we controlled for the effects of these same variables.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflict of interest statementAcknowledgementsReferences |
|---|
Sociodemographic characteristics
The study population had a mean age of 64 years (SD = 10.82) and earned nearly $30,000 per person a year on average. The sample tended to be white (70%), have a college degree or higher (58%) and have health insurance (95%) and a personal doctor or nurse (85%). Although 85% of participants reported excellent or very good health status (85%), 28% said they had been diagnosed with some kind of cancer other than prostate.
Results of t-tests and chi-square tests comparing the groups allowed us to characterize the three groups by sociodemographic characteristics (Table II). IAs were more likely to be white (than passives, P < 0.05), have more formal education (than DDAs and passives, P < 0.001) and have higher incomes (than passives, P < 0.05). DDAs were older (than passives and IAs, P < 0.01), more likely to have visited the doctor in the past year (than passives, P < 0.01) and more likely to have Medicare and/or Medicaid coverage (than IAs, P < 0.05). DDAs self-reported as being less healthy with a higher percentage reporting that they had been diagnosed with cancer other than prostate cancer (than IAs, P <0 .05) and with heart attack or stroke (than passives, P < 0.05) (not shown in table). Passives were less likely to have a personal doctor (than DDAs and IAs, P < 0.01) (not shown in table), were younger (than IAs, P < 0.01) and less likely to have health insurance (than DDAs and IAs, P < 0.05).
|
Knowledge, attitudes and behaviors
As shown in Table III, at baseline IAs were more knowledgeable about prostate cancer than the other two groups (P < 0.05), and the DDAs were more knowledgeable than the passives (P < 0.05). Correspondingly, higher percentages of both IAs and DDAs than passives (P < 0.001) reported paying more attention to health topics. Passives had a more fatalistic attitude about their health than did IAs and DDAs (P < 0.05). Further, compared with IAs and DDAs, lower percentages of passives reported having had a recent PSA test (P < 0.001) and planning to get a PSA test in the next year (P < 0.01). Not surprisingly, a higher percentage of DDAs than IAs (P < 0.001) preferred greater involvement of their doctor in the PSA test decision. Compared with the other two groups, passives were less self-efficacious about communicating effectively with doctors (P < 0.01) and reported that doctors and other health care professionals less frequently listened carefully to them, explained things in a way they could understand and spent enough time with them (P < 0.05).
|
Predictors of health information style
Results from the multinomial logistic regression model predicting health information group indicated that IAs were more likely than passives to be older (P = 0.058), have higher education (P < 0.01), have a personal doctor or nurse (P < 0.05) and have higher self-efficacy (P < 0.001) (Table IV). In the comparisons between DDAs and passives, DDAs were more likely to be older (P < 0.001), have a personal doctor (P < 0.01) and have higher self-efficacy (P < 0.0001).
|
Prior PSA information and reactions to the intervention
Consistent with their higher prior knowledge levels about PSA screening, IAs reported having more information about screening before the intervention than both DDAs and passives (P < 0.05) (Table V). In terms of sources of PSA information, greater percentages of IAs and DDAs than passives reported getting screening information from their doctors (P < 0.001). IAs were more likely than both DDAs and passives to report getting information from newspapers, magazines or newsletters (P < 0.05). They were also more likely than DDAs and passives to obtain screening information from the Internet (P < 0.01). IAs reported that less of the intervention information was new to them than did passives and DDAs (P < 0.05). Compared with passives, higher percentages of both IAs and DDAs reported agreeing with the information presented (P < 0.05) and being more likely to share the information with others (P < 0.05). DDAs were more likely than passives (P < 0.01) to trust the intervention information.
|
Predictors of prostate cancer knowledge
Table VI presents the results of the generalized linear model predicting prostate cancer knowledge at baseline and 6-month follow-up. Participants who were married (P < 0.05) and had a college education (P < 0.01) received significantly higher knowledge scores. The overall interaction between time and health information style was significant (P < 0.05), indicating that participants in the different health information style groups gained and retained differential amounts of knowledge from the intervention. These findings are consistent with the findings from the larger study examining the effects of the two versions of the intervention: higher knowledge scores were significantly associated with higher education and being married and the two intervention sites experienced significant increases in knowledge over time (McCormack et al., in preparation).
|
For easier interpretation of the interaction effect and because coefficients from the interaction cannot be individually interpreted, Fig. 1 presents the model-adjusted mean knowledge scores for each health intervention style group over time. Although passives began the study with the lowest mean knowledge scores, they exhibited the greatest improvement in knowledge following the intervention. Passives had significantly greater increases in model-adjusted mean knowledge scores from baseline to the 6-month follow-up than IAs (P = 0.01) and greater increases than DDAs, although they were not significantly greater (P = 0.06).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflict of interest statementAcknowledgementsReferences |
|---|
This study had two primary purposes. First, we explored the utility and applicability of a previously developed health information styles segmentation strategy [25] for a study with men ages 40–80 who participated in a community-based, prostate cancer screening IDM intervention. The goal was to assess how helpful the segmentation strategy is in understanding audience subgroups so that messages can be targeted appropriately. Second, we examined whether the PSA intervention had differential effects on men depending on their health information style according to the segmentation strategy.
Application of the health information styles segmentation strategy
The health information segmentation strategy used in this study has been applied previously to a nationally representative sample. Maibach et al. estimated the population prevalence of each health information style in 1999 and 2003, respectively, as follows: IAs (30.6 and 27.1%), DDAs (22.9 and 20.7%), DDPs (26.3 and 20.0%) and IPs (20.2 and 32.1%) [25]. Compared with the national estimates for population prevalence of each segment, our study was skewed toward IAs (41%) and DDAs (47%), likely in large part because of the self-selection process for participating in the intervention. The age eligibility requirements for our study (aged 40–80) also may explain the lower proportion of passives (12% for DDPs and IPs combined) relative to the national study. The national study found that IPs were the youngest health information segment with a mean age of 38 years. Thus, many passives would have been ineligible for our intervention based on their age. The sociodemographic profiles for IAs and DDAs in our study were generally consistent with findings from the national study.
When we applied the health information styles segmentation strategy to our study population, we identified audience segments with distinct and meaningful profiles in terms of sociodemographic, psychosocial, health and behavioral characteristics. IAs tended to be white, to be more highly educated and to have greater baseline prostate cancer knowledge (this may be because they are more highly educated in general). DDAs were older, less healthy and more likely to have visited a doctor in the past year (perhaps because they are older and less healthy). Passives tended to be younger, not to have had a PSA test and to have lower self-efficacy related to communication with doctors and less positive interactions with doctors. These profiles are generally consistent with HINTS and other research, which found that individuals with higher income and educational levels [8, 31], health insurance [33] and chronic health conditions [8] tend to be more active health information seekers. This study does not allow us to determine why individuals falling into different health information style categories have specific profiles. Future research could explore questions such as why passives tend to report less positive interactions with doctors (e.g. is it because they are less confident in their ability to understand health information or because they place less importance on health information?).
The segmentation strategy is useful because it identifies audience segments that may respond to different amounts and types of information and different message strategies and who can be reached most effectively through different communication channels. As a result, public health educators can plan and implement interventions more successfully, increasing the likelihood that people will be reached and informed [34]. For example, health information styles can be assessed with the brief 10-item assessment and then information tailored to this style can be provided to the individual whether it be in the form of educational materials or patient-provider communication in a clinical setting. Because a central component of IDM revolves around patients obtaining and understanding scientifically accurate information, passives who have low self-efficacy for health information are a group of particular importance. Discovering characteristics associated with passives can assist educators in targeting information to them in a manner that enhances their receptivity and comprehension using appropriate channels. For instance, we found that passives have lower utilization of medical appointments and thus fewer opportunities to intervene in that setting. Therefore, familiar and trusted community settings, like the ones utilized in this intervention, may be one way to reach them.
The health information styles segmentation strategy is distinct from yet not in conflict with other health information seeking typologies. It is based on self-efficacy, attitudes and interactions with physicians, which makes it particularly useful for IDM interventions. Therefore, because the information seeking and scanning behavior typology of Shim et al. [24] is based solely on engagement in these two behaviors and monitoring and blunting coping styles of Miller et al. [23] tap more into preferences for types of and amounts of information and emotional reactions to information, these typologies could be seen as complementary and could potentially be used in conjunction with the health information styles segmentation strategy.
Intervention effects
Our findings indicate that the IDM intervention had a meaningful and sustained positive impact on men's knowledge about prostate cancer, particularly among passives. Passives had the least knowledge about prostate cancer before the intervention, yet exhibited the greatest knowledge at the 6-month follow-up, and therefore the greatest knowledge gain overall. These results are encouraging because they indicate that this type of community-based intervention can be successful with a hard-to-reach population (i.e. passives). It is plausible that passives benefited more from the intervention because they had less knowledge about prostate cancer to begin with. Following the intervention, passives may have been more likely to attend to related information in the media, for example, thus accounting for the sustained increase in knowledge. Whereas passives were much more likely than other groups to say that the intervention information was new to them, they were also more skeptical about the information (less likely to agree with the information, trust the information, say they would share the information with others).
Study limitations
The present study had several limitations. First, almost 90% of our study population was classified as actives (IAs and DDAs). Given the small numbers of men classified as IPs and DPs, we had to combine these groups into a single passive category for the purpose of analyses. As a result, we were unable to explore possible differences in profiles between men having these two different passive health information styles. In particular, we could not examine the main difference between these two groups—the extent of their dependence on their doctor for medical decision making and health information—and its implications on PSA-related knowledge, attitudes and behaviors. We were also unable to assess the differential intervention effects for IPs and DDPs.
Second, our sample was limited to participants who attended a community-based intervention and responded to a 6-month follow-up survey, and they tended to be white, have a college degree or more, be insured and have a personal doctor. To further our understanding of how IDM interventions affect knowledge and other outcomes across participants having different health information styles, it will be beneficial to conduct studies with more inclusive samples. Further, men who were younger, less educated, black and who did not have a personal doctor were more likely to drop out of the study by the 6-month follow-up. This differential dropout could have skewed the findings from our analyses predicting prostate cancer knowledge. However, all these variables were controlled for in our analysis and, of them, education was the only significant predictor.
Although the community-based IDM intervention demonstrated positive effects on participants’ knowledge, the study design does not allow us to determine which specific elements or characteristics of this multifaceted intervention positively influenced participants’ knowledge (i.e. physician presentation, question and answer session, video and print materials). The intervention was designed to be provided as a whole with the different elements reinforcing and complementing each other. However, based on structured observations at the community intervention sessions, the physician presentation appeared to be the most engaging component of the intervention and the video also drew significant interest.
Implications for interventions and future research
Even with these considerations, our study findings provide insight into interventions for audiences with differing orientations to health information. First, we emphasize that a community-based intervention can successfully reach passives—men who may otherwise be less likely to seek out such information—and improve their knowledge. The community-based nature of the intervention, which was conducted within familiar and trusted faith-based and fraternal organizations and potentially among friends and acquaintances, may have enabled us to reach passives better than other intervention models (e.g. in provider settings). Our findings show some support for this hypothesis; for instance, passives are just as likely as others to get health information from family members, friends or coworkers, but less likely to get it from health care professionals, Web sites or brochures. The findings underscore the importance of understanding audiences’ health information styles when developing and implementing public health interventions. The health information styles segmentation strategy used for this study offers a viable framework for segmenting audiences and developing health communication messages and campaigns.
An intriguing area of research related to this study is whether and how an individual's health information style changes over time and as a result of IDM and other interventions. Because we did not assess health information styles other than at baseline, we cannot determine if the intervention influenced participants’ health information style—that is whether they moved from a more passive to a more active style or vice versa. To gain an understanding of health information styles over time—that is whether an individual's style tends to be static or changes with age or in response to health concerns or other factors—longitudinal studies are needed. An important focus for such research will be on evaluating the impact of IDM and other interventions. By assessing participants’ health information styles before and after an intervention and incorporating long-term follow-up, evaluations can identify intervention strategies that help consumers to become more actively engaged with health information and decision making in accordance with their individual preferences.
|
Conflict of interest statement |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflict of interest statementAcknowledgementsReferences |
|---|
None declared.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflict of interest statementAcknowledgementsReferences |
|---|
This publication was made possible through the Centers for Disease Control and Prevention and the Association for Prevention Teaching and Research Cooperative Agreement, No. U50/CCU300860, Project TS-0845. The findings and conclusions in this publication are those of the authors and do not necessarily represent the views of the Centers for Disease Control and Prevention or the Association for Prevention Teaching and Research.
Copies of the intervention materials used in this research are available on request to Lauren McCormack, PhD, MSPH, RTI International, 3040 Cornwallis Road, PO Box 12194, RTP, NC 27709 or E-mail: Lmac{at}rti.org.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflict of interest statementAcknowledgementsReferences |
|---|
1. Briss P, Rimer B, Reilley B, et al. Promoting informed decisions about cancer screening in communities and healthcare systems. Am J Prev Med (2004) 26:67–80.[CrossRef][Web of Science][Medline]
2. Institute of Medicine. Crossing the Quality Chasm: A New Health System for the 21st Century (2001) Washington, DC: National Academy Press.
3. Rimer BK, Briss PA, Zeller PK, et al. Informed decision making: what is its role in cancer screening? Cancer (2004) 101(Suppl. 5):1214–28.[CrossRef][Web of Science][Medline]
4. Harris R, Lohr KN. Screening for prostate cancer: an update of the evidence for the U.S. Preventive Services Task Force. Ann Intern Med (2002) 137:917–29.
5. U.S. Preventive Services Task Force. Screening for prostate cancer: recommendation and rational. Ann Intern Med (2002) 137:915–33.
6. Chan ECY, Vernon SW, O'Donnell FT, et al. Informed consent for cancer screening with prostate-specific antigen: how well are men getting the message? Am J Public Health (2003) 93:779–85.
7. Gwede CK, McDermott RJ. Prostate cancer screening decision making under controversy: implications for health promotion practice. Health Promot Pract (2006) 7:134–46.
8. Tu HT, Hargraves JL. Seeking health care information: most consumers still on the sidelines. Issue Brief Cent Stud Health Syst Change, (2003) 61:1–4.[Medline]
9. Rodgers S, Chen Q, Duffy M, et al. Media usage as health segmentation variables. J Health Commun (2007) 12:105–19.[CrossRef][Web of Science][Medline]
10. National Cancer Institute. Health Information National Trends Survey (HINTS). Available at: http://hints.cancer.gov/hints/. Accessed: 6 March 2008.
11. Squiers L, Finney Rutten LJ, Treiman K, et al. Cancer patients’ information needs across the cancer care continuum: evidence from the cancer information service. J Health Commun (2005) 10(Suppl. 1):15–34.[CrossRef][Web of Science][Medline]
12. Leydon GM, Boulton M, Moynihan C, et al. Cancer patients’ information needs and information seeking behavior: in-depth interview study. Br Med J (2000) 320:909–13.
13. Matthews AK, Sellergren SA, Manfredi C, et al. Factors influencing medical information seeking among African American cancer patients. J Health Commun (2002) 7:205–19.[CrossRef][Web of Science][Medline]
14. Maibach E, Maxfield A, Ladin K, et al. Translating health psychology into effective health communication: the American Healthstyles Audience Segmentation Project. J Health Psychol (1996) 1:261–77.[Abstract]
15. Slater MD. Theory and method in health audience segmentation. J Health Commun (1996) 1:267–83.[CrossRef][Medline]
16. Kreuter MW, Strecher VJ, Glassman B. One size does not fit all: the case for tailoring print materials. Ann Behav Med (1999) 21:276–83.[Web of Science][Medline]
17. Kreuter MW, Oswald DL, Bull FC, et al. Are tailored health education materials always more effective than non-tailored materials? Health Educ Res (2000) 15:305–15.
18. Brug J, Steenhuis I, Van Assema P, et al. Computer-tailored nutrition education: differences between two interventions. Health Educ Res (1999) 14:249–56.
19. Williams-Piehota P, Pizarro J, Schneider TR, et al. Matching health messages to monitor-blunter coping styles to motivate screening mammography. Health Psychol (2005) 24:58–67.[CrossRef][Web of Science][Medline]
20. Updegraff JA, Sherman DK, Luyster FS, et al. Understanding how tailored messages work: the effects of message quality and congruency on perceptions of health messages. J Exp Soc Psychol (2006) 43:249–57.[CrossRef][Web of Science]
21. Petty RE, Wegener DT. Attitude change: multiple roles for persuasion variables. In: The Handbook of Social Psychology—Gilbert D, Fiske S, Lindsay G, eds. (1998) 4th ed. New York: McGraw Hill. 323–90.
22. Navarro FH, Wilkins ST. A new perspective on consumer health Web use: "valuegraphic" profiles of health information seekers. Manag Care Q (2001) 9:35–43.[Medline]
23. Miller SM, Shoda Y, Hurley K. Applying cognitive-social theory to health-protective behavior: breast self-examination in cancer screening. Psychol Bull (1996) 119:70–94.[CrossRef][Web of Science][Medline]
24. Shim M, Kelly B, Hornik R. Cancer information scanning and seeking behavior is associated with knowledge, lifestyle choices, and screening. J Health Commun (2006) 11(Suppl. 1):157–72.[CrossRef][Web of Science][Medline]
25. Maibach EW, Weber D, Massett H, et al. Understanding consumers’ health information preferences: development and validation of a brief screening instrument. J Health Commun (2006) 11:717–36.[CrossRef][Web of Science][Medline]
26. Degner LF, Sloan JA. Decision making during serious illness: what role do patients really want to play? J Clin Epidemiol (1992) 45:941–50.[CrossRef][Web of Science][Medline]
27. Degner LF, Sloan JA, Venkatesh P. The Control Preferences Scale. Can J Nurs Res (1997) 29:21–43.[Medline]
28. Degner LF, Russell CA. Preferences for treatment control in adults with cancer. Res Nurs Health (1988) 11:367–74.[Web of Science][Medline]
29. Long JS. Regression Models for Categorical and Limited Dependent Variables (1997) Thousand Oaks, CA: Sage Publications.
30. Hay J, Coups E, Ford J. Predictors of perceived risk for colon cancer in a national probability sample in the United States. J Health Commun (2006) 11(Suppl. 1):71–92.[CrossRef][Web of Science][Medline]
31. Rutten LJ, Squiers L, Hesse B. Cancer-related information seeking: hints from the 2003 Health Information National Trends Survey (HINTS). J Health Commun (2006) 11(Suppl 1):147–56.[CrossRef][Web of Science][Medline]
32. Rutten LJ, Augustson E, Wanke K. Factors associated with patients’ perceptions of health care providers’ communication behavior. J Health Commun (2006) 11(Suppl. 1):135–46.[CrossRef][Web of Science][Medline]
33. Treiman K, Squiers L. CIS Research Agenda: Overview of Relevant Research (2005) National Cancer Institute. NIH Pub. No. 05–5309.
34. Slater MD, Flora JA. Health lifestyles: audience segmentation analysis for public health interventions. Health Educ Q (1991) 18:221–33.[Web of Science][Medline]
Received on April 27, 2007; accepted on January 23, 2008
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||