Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Oct 1.
Published in final edited form as: Health Educ Behav. 2017 Dec 28;45(5):790–799. doi: 10.1177/1090198117744242

Perceptions of Cancer Risk/Efficacy and Cancer-related Risk Behaviors: Results from the HCHS/SOL Sociocultural Ancillary Study

Rina S Fox 1,2, Sarah D Mills 2, Scott C Roesch 2,3, Daniela Sotres-Alvarez 4, Patricia Gonzalez 5, Venera Bekteshi 6, Jianwen Cai 4, David W Lounsbury 7, Gregory A Talavera 8, Frank J Penedo 1, Vanessa L Malcarne 2,3
PMCID: PMC6035116  NIHMSID: NIHMS978595  PMID: 29284295

Abstract

This study evaluated the associations among perceived risk, perceived efficacy, and engagement in six cancer-related risk behaviors in a population-based Hispanic/Latino sample. Interviews were conducted with 5,313 Hispanic/Latino adults as part of the Hispanic Community Health Study/Study of Latinos (HCHS/SOL) Sociocultural Ancillary Study. Participants were recruited from the study’s four field centers (Bronx, NY; Chicago, IL; Miami, FL; San Diego, CA) between February 2010 and June 2011. Perceived risk and perceived efficacy were assessed with questions drawn from the Health Interview National Trends Survey. More than half of the sample endorsed perceived risk of cancer associated with the six evaluated behaviors, as well as general perceived efficacy for preventing cancer. Adjusted logistic regression analyses demonstrated significant differences across Hispanic/Latino background groups for perceived risk associated with high consumption of alcohol and saturated fat, low consumption of fruits and vegetables, and insufficient exercise, but not for smoking or low consumption of fiber. Differences were also found for the belief, “It seems like everything causes cancer,” but not for other perceived efficacy items. Perceived cancer risk and perceived efficacy for preventing cancer were neither independently nor interactively associated with engagement in cancer-related risk behaviors after controlling for sociodemographic covariates. Results suggest perceptions of risk and efficacy with regard to cancer vary across Hispanic/Latino background groups, and therefore background group differences should be considered in prevention efforts. Perceived risk and perceived efficacy were not related to cancer-related risk behaviors among Hispanics/Latinos. Further work is needed to evaluate determinants of cancer-related risk in this population.

Keywords: Hispanic/Latino, cancer, perceived risk, perceived efficacy, health behaviors, HINTS

Cancer is the leading cause of mortality among Hispanics/Latinos (American Cancer Society [ACS], 2014), and tends to be diagnosed at more advanced stages among Hispanics/Latinos as compared to non-Hispanic/Latino whites (ACS, 2015). Given the rapid growth of the U.S. Hispanic/Latino population (Humes, Jones, & Ramirez, 2011; Passel, Cohn, & Lopez, 2011), understanding determinants of cancer-related risk behaviors is of significance to the public health of this growing population.

Lifestyle behaviors have been identified as major contributors to preventable cancers in the United States (U.S.; ACS, 2014). Several theoretical models including the Transtheoretical Model (Prochaska & DiClemente, 1983), the Social Cognitive Theory (Bandura, 1986), the Health Belief Model (Rosenstock, 1974, 2005), and the Theory of Reasoned Action (Ajzen, 1991; Ajzen & Fishbein, 1980) help explain why individuals engage in cancer-related risk behaviors. Two common constructs across these models are perceived risk and perceived efficacy. Within the context of cancer, perceived risk refers to the degree to which individuals believe they are susceptible to cancer. Perceived efficacy refers to the degree to which individuals believe they are capable of diminishing the likelihood of cancer. The Risk Perception Attitude (RPA) model postulates that perceived efficacy moderates the relationship between perceived risk and cancer-related risk behavior (Rimal, 2001; Rimal & Real, 2003). Specifically, the model hypothesizes that perceived efficacy will have a weak association with behaviors perceived to be low risk, but a strong association with behaviors perceived to be more risky for cancer.

Several studies conducted across multiple countries have evaluated the relationship between perceived risk and cancer-related risk behaviors; however, few have been conducted with large, national samples. Studies that have used an observational design on a national scale have examined diet (Harnack, Block, Subar, Lane, & Brand, 1997; Hawkins, Berkowitz, & Peipins, 2010; Kristal, Hedderson, Patterson, & Neuhauser, 2001; Patterson, Kristal, Lynch, & White, 1995; Patterson, Kristal, & White, 1996; Sullivan et al., 2010; van Assema, Pieterse, Kok, Eriksen, & de Vries, 1993; Watters & Satia, 2009; Watters, Satia, & Galanko, 2007), physical activity (Hawkins et al., 2010), sun exposure (Hamilton et al., 2012; Sullivan et al., 2010; van Assema et al., 1993), alcohol consumption (van Assema et al., 1993), and smoking (Hawkins et al., 2010; van Assema et al., 1993). Some studies have found support for an association between risk perceptions and behavior, while others have not. None of these studies, however, have examined the relationship between cancer-risk perceptions and cancer-risk behaviors in a large and diverse sample of Hispanics/Latinos.

Perceived efficacy, or the belief that cancer risk is modifiable, has been labeled in the literature in multiple ways. For example, Cameron (2008) operationalized this belief as “perception of personal control over prevention,” and found no association between control beliefs and skin cancer prevention behaviors among university students. Conversely, Niederdeppe and Levy (2007) operationalized the belief that cancer risk is modifiable as “fatalism,” and reported that fatalistic beliefs about cancer prevention were associated with less exercise, less consumption of fruits and vegetables, and more smoking in a U.S. population-based sample. However, although some research has explored self-efficacy in the general population, the relationship between perceived efficacy and health behaviors has not been examined specifically among Hispanics/Latinos at the population level.

A primary limitation of most studies examining Hispanics/Latinos in the U.S. is that they categorize Hispanics/Latinos as a single ethnic group. However, there are important health-related differences across Hispanic/Latino background groups. For example, women of Dominican background have been shown to be more likely to receive mammograms, and individuals of Puerto Rican, Central American, and South American background have been shown to be less likely to engage in colorectal cancer screening, as compared to other Hispanic/Latino background groups (Gorin & Heck, 2005). Additionally, individuals of Cuban background have demonstrated greater knowledge of mammography guidelines than those of Mexican background, and greater knowledge of pap smear guidelines than those of Puerto Rican background (Ramirez, Suarez, Laufman, Barroso, & Chalela, 2000). Accordingly, it has been recommended that studies evaluate Hispanic/Latino background groups distinctively, rather than as a homogenous population, to capture such differences (Weinick, Jacobs, Stone, Ortega, & Burstin, 2004; Zsembik & Fennell, 2005).

The present study evaluated the associations among perceived risk, perceived efficacy, and cancer-related risk behaviors in a population-based Hispanic/Latino sample. Six lifestyle-based cancer-related risk behaviors were examined: smoking, saturated fat intake, fruit and vegetable intake, fiber intake, alcohol consumption, and physical activity. The specific aims were to 1) describe perceptions of which behaviors are risky for cancer, and perceived efficacy for preventing cancer; 2) compare Hispanic/Latino background groups (i.e., Central American, Cuban, Dominican, Mexican, Puerto Rican, South American, More than one race/Other) on the perceptions of which behaviors are risky for cancer and perceived efficacy for preventing cancer; and 3) examine the association between perceived risk and perceived efficacy with cancer-related risk behaviors, both independently and in interaction. It was hypothesized that both perceived risk and perceived efficacy would independently be associated with engagement in cancer-related risk behaviors, and that perceived efficacy would moderate the relationship between perceived risk and cancer-related risk behavior in accordance with the RPA model.

Methods

Study Design

The current study was a cross-sectional, population-based study conducted with 5,313 persons enrolled in the Hispanic Community Health Study/Study of Latinos (HCHS/SOL) Sociocultural Ancillary Study (SCAS). HCHS/SOL is the largest prospective, population-based cohort of Hispanic/Latino adults living in the U.S. (N = 16,415) assessing the prevalence and incidence of chronic illness, as well as risk and protective factors thereof. The SCAS examined the relationship of sociocultural and psychosocial factors to the prevalence of cardiovascular disease and metabolic syndrome among Hispanics/Latinos.

Participants

Participants were representative of the broader HCHS/SOL study, as the households involved with the parent study were randomly sampled for the SCAS to form three waves and years of recruitment (Gallo et al., 2014; LaVange et al., 2010), and were recruited from four field centers (Bronx, NY; Chicago, IL; Miami, FL; San Diego, CA). The HCHS/SOL methodology for the parent study (LaVange et al., 2010; Sorlie et al., 2010) and the SCAS (Gallo et al., 2014), have been published elsewhere. Participants included immigrants from various countries in Latin America, with varying numbers of years living in the U.S., second and third generation U.S.-born individuals, and individuals identifying Puerto Rico as their background group.

Procedures

All participants in the HCHS/SOL completed an in-person baseline clinic visit and yearly follow-up telephone assessments, and are currently participating in a second clinic visit (October 2014 – September 2017). The SCAS invited individuals from the parent study who had completed the baseline visit to participate; of the 7,321 individuals who were invited 5,313 (72.6%) participated between February 2010 and June 2011. Individuals who enrolled and provided informed consent completed a separate interviewer-administered standardized assessment. Perceptions of which behaviors are risky for cancer and general perceived efficacy for preventing cancer were assessed at this appointment. The HCHS/SOL and SCAS were approved by Institutional Review Boards at all participating institutions.

Measures

Perceived risk

Participants completed questions from the Health Interview National Trends Survey (HINTS; Nelson et al., 2004) to evaluate the degree to which individuals believed certain behaviors are associated with cancer risk. HINTS questionanires were developed according to scientific validity and reliability criteria and have previously been evaluated for psychometric rigor (Cantor, Covell, Davis, Park, & Rizzo, 2005). Specifically, participants were asked “Do you think that _______ increase(s) a person’s chance of getting cancer?” Items evaluated in the present study addressed: 1) smoking; 2) eating a high fat diet; 3) not eating many fruits and vegetables; 4) not eating much fiber; 5) drinking a lot of alcoholic beverages; and 6) not getting much exercise. Responses marked as “a lot” or “a little” were coded to indicate agreement, while those marked as “not at all” or “no opinion” were coded to indicate disagreement, consistent with previous research (Lykins et al., 2008).

Perceived efficacy

Questions from the HINTS (Nelson et al., 2004) were also used to assess beliefs about whether or not a person can influence cancer risk. Participants were asked to rate their agreement with the following statements: “It seems like everything causes cancer;” “There’s not much you can do to lower your chances of getting cancer;” and “There are so many different recommendations about preventing cancer, it’s hard to know which ones to follow.” Responses marked as “strongly agree” or “agree” were coded to indicate agreement, while those marked as “disagree” or “strongly disagree” were coded to indicate disagreement, following previous research (Lykins et al., 2008).

Cancer-related risk behaviors

Smoking

In accordance with prior research (Daviglus et al., 2012), participants who reported currently smoking on some or all days were categorized as current smokers. Those who reported smoking fewer than 100 cigarettes in their lifetime and those who reported smoking 100 or more cigarettes during their lifetime but were not currently smoking were categorized as nonsmokers/former smokers.

Saturated fat intake

Intake of saturated fatty acids, fruits and vegetables, and fiber were evaluated via two separate 24-hour dietary recalls, administered six weeks apart. See Siega-Riz et al. (2014) for a comprehensive description of the dietary data collection for the HCHS/SOL. Participants were assigned a score of one to five to reflect their gender-specific quintile of daily saturated fat intake. Individuals who fell into the lowest 40th nutritional percentile were characterized as healthy, while those in the highest 60th percentile were characterized as unhealthy (Daviglus et al., 2012).

Fruit and vegetable intake

Participants were classified as adherent or nonadherent to the 5-A-Day for Better Health guidelines, based on the program sponsored by the National Cancer Institute and the Produce for Better Health Foundation (Daviglus et al., 2012).

Fiber intake

Participants were assigned a score of one to five reflecting their gender-specific quintile of daily fiber intake. Individuals who fell into the highest 40th nutritional percentile were characterized as having a healthy intake of fiber, while those in the lowest 60th percentile were characterized as unhealthy (Daviglus et al., 2012).

Alcohol consumption

Participants who reported currently consuming alcoholic beverages on some days or daily were categorized as drinkers, while those who reported drinking no alcohol were categorized as nondrinkers.

Physical activity

Based on responses to the Global Physical Activity Questionnaire (GPAQ; Bull, Maslin, & Armstrong, 2009) individuals who reported engaging in a minimum of 150 minutes of moderate-intensity activity or 75 minutes of vigorous-intensity activity per week, or the equivalent combination of moderate and vigorous activity, in episodes of at least 10 minutes, were classified as active, while those who engaged in less activity were classified as inactive. The GPAQ has been shown to have acceptable psychometric properties (Bull et al., 2009).

Covariates

Sociodemographic variables

Sociodemographic variables including age, sex, income, education, and years spent living in the U.S. were collected via self-report.

Acculturation

Acculturation was measured with a modified version of the Short Acculturation Scale for Hispanics (SASH; Marin, Sabogal, Marin, Otero-Sabogal, & Perez-Stable, 1987). For the present analysis, the four-item social subscale and the five-item language subscale were used. Subscale scores range from one to five, with higher scores indicating greater acculturation to the U.S. Internal consistency reliability for the current data was acceptable (Language: α = .93; Social: α = .72).

Statistical Analyses

For all analyses inferential statistics accounted for the complex survey design and sample weights to produce weighted population estimates (LaVange et al., 2010). To describe the perceptions of which behaviors are risky for cancer, and perceived efficacy for preventing cancer (Aim1), descriptive statistics were calculated for all study variables. To compare Hispanic/Latino background groups on the perceptions of which behaviors are risky for cancer and perceived efficacy for preventing cancer (Aim2), a series of logistic regression analyses were conducted to evaluate potential differences across background groups (independent variable) for each of the six perceived risk and three perceived efficacy variables (dependent variables). To control for confounding of study site with Hispanic/Latino background group, because people of specific Hispanic/Latino backgrounds tend to concentrate in specific geographic areas, models were fit using a background group-by-center variable, with levels corresponding to the ten combinations of background group and center that had at least 100 participants according to unweighted counts. To examine perceived risk and perceived efficacy as correlates of cancer-related risk behaviors, both independently and in interaction (Aim3), logistic regression models were fit for each of the six cancer-related risk behaviors, in which each behavior was regressed simultaneously onto the associated perceived risk variable, one of the three perceived efficacy variables, and the interaction of the perceived risk and perceived efficacy variables. The logistic regression models that were run to address the second and third study aims controlled for age, sex, income, education, years spent living in the U.S., and the SASH language and social subscale scores. To account for multiple testing, a Bonferroni correction (α = .001) was utilized in the analyses for these aims. Full information maximum likelihood (FIML) was used to address missing data. This approach has been shown to be preferable to listwise deletion and multiple imputation, and has demonstrated production of unbiased parameter estimates and standard errors under various missing data conditions (Enders, 2010; Enders & Bandalos, 2001). All analyses were conducted in IBM Statistical Package for the Social Sciences (SPSS) Version 20.0 and MPlus version 7.2 using complex survey procedures.

Results

Sociodemographic Characteristics

Average scores on the SASH language (M = 2.1, SD = 1.2) and social (M = 2.2, SD = 0.6) subscales indicated that participants identified more strongly with the Hispanic/Latino culture than the U.S. culture. The majority was categorized as unhealthy with regard to consumption of fiber (77.1%) and fruits and vegetables (96.0%), while the majority was categorized as healthy with regard to physical activity (64.6%), consumption of saturated fat (60.9%), and smoking (79.3%). Half was categorized as unhealthy (49.6%) with regard to alcohol consumption (Table 1).

Table 1.

Percentage of engagement in cancer-related risk behaviors, perceived cancer risk and efficacy, and sociodemographic characteristics in the HCHS/SOL Sociocultural Ancillary Study

Unweighted
n 		Weighted percent or mean (SD)
Engagement in Cancer-related Risk Behaviors
 Alcohol use: Drinker 2,425 49.6%
 Exercise: Inactive 1,994 35.4%
 Fiber: Unhealthy 3,880 77.1%
 5-A-Day: Non-adherent 5,005 96.0%
 Saturated fat: Unhealthy 1,680 39.1%
 Cigarette use: Smoker 974 20.7%
High Perceived Cancer Risk
 Alcohol 4,190 79.9%
 Exercise 2,993 56.5%
 Fiber 3,433 62.3%
 Fruits and Vegetables 3,384 62.6%
 Saturated fat 4,204 78.6%
 Smoking 5,120 97.7%
High Perceived Efficacy to Avoid Cancer
 Everything 3,032 60.5%
 Not Much 2,985 59.7%
 Confusing 1,468 29.2%
Sex
 Women 3,299 54.9%
Hispanic/Latino Background Group
 Central American 553 7.6%
 Cuban 775 20.3%
 Dominican 534 11.7%
 Mexican 2080 36.5%
 Puerto Rican 880 15.8%
 South American 350 4.8%
 More than one race/Other 137 3.3%
Annual Household Income
 < $10,000 888 17.7%
 $10,001–15,000 988 20.1%
 $15,001–20,000 685 13.5%
 $20,001–25,000 568 11.3%
 $25,001-$29,999 386 7.5%
 $30,000-$40,000 623 12.9%
 $40,001-$50,000 299 6.4%
 $50,001-$75,000 257 5.6%
 $75,001-$100,000 105 2.8%
 > $100,000 73 2.2%
Education
 < High school/GED 1,923 32.5%
 High school/GED 1,383 28.0%
 > High school/GED 1,998 39.4%
US Born
 Not born in 50 US states 4,393 78.1%
Language of Interview
 Spanish 4,296 75.4%
Years in United States among Foreign Born 4,384 17.3 (13.9)
Age, years 5,313 42.5 (15.0)
SASH Social Score 5,118 2.2 (0.6)
SASH Language Score 5,306 2.1 (1.2)
Open in a new tab

Note. Variations in total sample size across variables were due to missing data; GED = General Education Development Test; US = United States; SASH = Short Acculturation Scale for Hispanics.

Perceived risk and perceived efficacy

For each of the evaluated cancer-related risk behaviors, more than half agreed that the behavior was associated with increased risk for cancer (Table 1). The smallest percentage (56.5%) held this belief for the relationship between insufficient exercise and cancer risk, whereas nearly all (97.7%) did so for the relationship between smoking and cancer risk. Approximately seventy percent agreed that “There are so many different recommendations about preventing cancer, it’s hard to know which ones to follow,” demonstrating lower perceived efficacy, and approximately 40 percent agreed that “It seems like everything causes cancer,” and “There’s not much you can do to lower your chances of getting cancer.”

Perceived cancer risk and perceived cancer efficacy across Hispanic/Latino background groups

After controlling for covariates, logistic regression analyses demonstrated significant (p ≤ .001) differences across Hispanic/Latino background groups for perceived cancer risk associated with consumption of alcohol, fruits and vegetables, and saturated fat, as well as insufficient exercise, but not for perceived cancer risk associated with smoking or consumption of fiber. Differences were also found for the belief, “It seems like everything causes cancer,” and “There are so many different recommendations about preventing cancer, it’s hard to know which ones to follow,” but not for the belief “There’s not much you can do to lower your chances of getting cancer.” Percentages of participants from each background group that endorsed the perceived risk and perceived efficacy beliefs are presented in Table 2, and significant findings from the logistic regression analyses are presented in Table 3. For the majority of perceived risk and perceived efficacy beliefs assessed, differences were either not found or were observed among only a few groups. However, the likelihood of reporting high perceived cancer risk associated with insufficient exercise was found to significantly differ in nine pairwise comparisons, and the belief that “There are so many different recommendations about preventing cancer, it’s hard to know which ones to follow” significantly differed in five pairwise comparisons. In general, individuals of Mexican background from San Diego (odds ratios (ORs) ranged from 1.93 [95% confidence interval (CI) = 1.43, 2.61] to 2.03 [95% CI = 1.37, 3.02]) and individuals of South American background from Chicago (ORs ranged from 2.87 [95% CI = 1.54, 5.36] to 4.08 [95% CI = 2.14, 7.81]) were significantly more likely to report high perceived cancer risk associated with insufficient exercise as compared to other groups. Additionally, individuals of South American background from Chicago were significantly less likely to disagree with the statement that that “There are so many different recommendations about preventing cancer, it’s hard to know which ones to follow” as compared to other groups (ORs ranged from 0.23 [95% CI = 0.11, 0.51] to 0.29 [95% CI = 0.13, 0.61]).

Table 2.

Cancer perceived risk and efficacy beliefs across Hispanic/Latino background groups (Weighted percent)

Dom – Bronx (n = 496) C Am – Chicago (n = 153) C Am – Miami (n = 323) Cuban – Miami (n = 745) Mex – Chicago (n = 752) Mex – San Diego (n = 1249) PR – Bronx (n = 570) PR – Chicago (n = 253) S Am – Chicago (n = 116) S Am – Miami (n = 149)
High Perceived Risk
 Alcohol 79.3 80.0 82.4 89.2 79.7 76.5 72.7 73.6 85.6 85.0
 Exercise 50.2 65.3 57.3 52.4 63.3 67.0 45.7 49.4 80.2 58.1
 Fiber 61.1 64.7 66.5 60.6 65.7 68.0 51.9 59.2 68.1 65.1
 Fruits/Veg 59.9 71.3 62.4 60.7 63.7 71.2 54.1 61.1 69.3 66.1
 Saturated fat 76.4 82.7 81.6 80.6 85.0 80.8 64.3 74.8 85.1 80.3
 Smoking 98.4 97.8 98.8 96.7 98.5 98.4 96.9 96.4 97.6 99.0
High Perceived Efficacy
 Everything 65.7 51.6 68.2 67.3 56.4 59.2 51.1 47.2 59.2 64.6
 Not Much 57.7 57.9 55.3 54.5 51.3 64.7 64.7 63.9 42.5 56.2
 Confusing 31.0 29.8 29.4 29.8 24.4 32.9 27.4 22.0 12.4 29.1
Open in a new tab

Note. ns for each Hispanic/Latino background group are unweighted. Dom = Dominican; C Am = Central American; Mex = Mexican; PR = Puerto Rican; S Am = South American; Fruits/Veg = Fruits/Vegetables; Everything = “It seems like everything causes cancer”; Not Much = “There’s not much you can do to lower your chances of getting cancer”; Confusing = “There are so many different recommendations about preventing cancer, it’s hard to know which ones to follow”. For risk variables values refer to percentages of participants who agreed that the referenced behavior is risky for cancer. For efficacy variables values refer to the percentages of participants who disagreed with the referenced statement.

Table 3.

Significant findings from logistic regression analyses comparing Hispanic/Latino background groups on cancer perceived risk and efficacy beliefs

Comp Ref Risk or Efficacy Outcome
Exercise Risk Alcohol Risk Fruits/Veg Risk Saturated Fat Risk Everything Efficacy Confusing Efficacy
OR 95% CI OR 95% CI OR 95% CI OR 95% CI OR 95% CI OR 95% CI
Mex – SD Cuban – Miami 1.93 1.43, 2.61 0.47 0.33, 0.67 1.75 1.30, 2.36 --- --- --- --- --- ---
Dom – Bronx 1.99 1.36, 2.92 --- --- --- --- --- --- --- --- --- ---
PR – Bronx 2.03 1.37, 3.02 --- --- 1.80 1.28, 2.52 --- --- --- --- --- ---
S Am – Chicago --- --- --- --- --- --- --- --- --- --- 4.04 1.96, 8.36
Mex – Chicago Cuban – Miami 1.79 1.33, 2.40 0.53 0.37, 0.78 --- --- --- --- --- --- --- ---
PR – Bronx --- --- --- --- --- --- 2.58 1.61, 4.14 --- --- --- ---
Cuban – Miami PR – Bronx --- --- 2.51 1.64, 3.86 --- --- --- --- --- --- --- ---
PR – Chicago --- --- --- --- --- --- --- --- 2.16 1.36, 3.42 --- ---
S Am – Chicago 0.26 .14, .47 --- --- --- --- --- --- --- --- 3.94 1.91, 8.13
S Am – Chicago Dom – Bronx 4.00 2.12, 7.58 --- --- --- --- --- --- --- --- 0.29 0.13, 0.61
C Am – Miami 2.87 1.54, 5.36 --- --- --- --- --- --- --- --- 0.26 0.12, 0.55
PR – Bronx 4.08 2.14, 7.81 --- --- --- --- --- --- --- --- 0.23 0.11, 0.51
PR – Chicago 3.83 1.94, 7.56 --- --- --- --- --- --- --- --- --- ---
Open in a new tab

Note. Only pairwise comparisons that were statistically significantly different are presented in this table (all ps ≤ .001). Comp = Comparison group; Ref = Reference group; C Am = Central American; Dom = Dominican; Mex = Mexican; PR = Puerto Rican; S Am = South American; SD = San Diego; Fruits/Veg = Fruits/Vegetables. For all models the Comparison group was coded as 1 and the Reference group was coded as 0.

Relationships of perceived risk and perceived efficacy to engagement in cancer-related risk behaviors

Two-way interaction analyses demonstrated no significant interactions (p > .001) between perceived risk and perceived efficacy in predicting the engagement in cancer-related risk behaviors after controlling for covariates. In addition, no associated significant main effects of perceived risk or perceived efficacy in predicting cancer-related risk behaviors were found.

Discussion

The first aim of the present study was to describe perceptions of behavior as risky for cancer, and general perceived efficacy for preventing cancer, among Hispanics/Latinos. Perceived cancer risk was lowest with regard to insufficient exercise, providing further support that physical inactivity is under-recognized as a cancer-related risk behavior (Ramirez, Finney Rutten, Vanderpool, Moser, & Hesse, 2013). Rates of perception of smoking, low fiber consumption, and alcohol consumption as risky for cancer were relatively similar to those reported by 2003 HINTS participants (National Cancer Institute, 2003). Perceived risk of saturated fat consumption was higher, and perceived risk of low fruit and vegetable consumption and physical inactivity were lower for participants in the present study. The rates of agreement with the three perceived efficacy items evaluated in the present study were relatively similar to those observed among 2003 HINTS participants. Of note, HINTS is comprised of a nationally representative sample consisting of individuals from a variety of racial and ethnic backgrounds. Thus, the present results suggest that Hispanics/Latinos may have slightly different perceptions of cancer risk than the general U.S. population, though perceived efficacy for preventing cancer was more congruent.

The second aim was to compare Hispanic/Latino background groups on perceptions of which behaviors are risky for cancer and perceived efficacy for preventing cancer. Overall, as compared to other background groups, individuals of Mexican background from Chicago and San Diego, and individuals of South American background from Chicago, were more likely to perceive the evaluated behaviors, with the exception of alcohol consumption, as risky for cancer. Conversely, individuals of Puerto Rican background from the Bronx were less likely to do so. Interestingly, no differences were found across background groups with regard to perceived risk of smoking or low consumption of fiber. Regarding perceived efficacy for preventing cancer, individuals of Puerto Rican background from Chicago were more likely to agree that everything causes cancer as compared to individuals of Cuban background from Miami. Moreover, individuals of South American background from Chicago were significantly more likely to agree that it is hard to know which recommendations should be followed to prevent cancer as compared to other groups. These results augment prior research demonstrating that there are discrepant health beliefs across different Hispanic/Latino background groups (Caban & Walker, 2006), and advance the literature by identifying which cancer-related risk and efficacy beliefs are discrepant and how they vary across geographic regions.

The third study aim was to examine the association of perceived risk and perceived efficacy with cancer-related risk behaviors, both independently and in interaction. In the present study, contrary to the RPA model (Rimal, 2001; Rimal & Real, 2003), perceived risk and perceived efficacy did not interact to predict cancer-related risk behaviors. Of note, prior studies evaluating the RPA model have measured risk and efficacy in similar ways. For example, studies that have assessed risk associated with particular behaviors have evaluated self-efficacy for performing those behaviors (e.g., Rimal & Real, 2003; van Assema et al. 1993), and studies that have assessed more general self-efficacy for preventing a disease such as cancer have assessed more general perceived risk of disease (e.g., Sullivan, Beckjord, Rutten, & Hesse, 2008). The present study extends the literature regarding the RPA model by suggesting that this model may not apply to situations in which perceived risk is assessed more specifically, while perceived efficacy is assessed more broadly.

After controlling for covariates, perceived risk and perceived efficacy were not significantly associated with cancer-related risk behavior. This is partially consistent with prior research, which has supported a relationship between risk perceptions and smoking, but not other behaviors (Sullivan et al., 2010; van Assema et al., 1993). Also, prior studies have found an association between cancer-related risk perceptions and behavior when examining this relationship bivariately, but not after adjusting for covariates, as was done in the present analysis (Hamilton et al., 2012; Kristal et al., 2001). With regard to generalized perceived efficacy for preventing cancer, Cameron (2008) found no relationship with skin cancer prevention behaviors, similar to the present results. However, Niederdeppe and Levy (2007) found that individuals demonstrating high perceived efficacy were less likely to exercise weekly, less likely to consume sufficient fruits and vegetables, and more likely to be smokers. Of note, the sample evaluated by Niederdeppe and Levy was comprised of individuals of diverse racial and ethnic backgrounds, which may contribute to this discrepancy. Additionally, while Niederdeppe and Levy controlled for sociodemographic variables, they did not evaluate perceived risk. Finally, while some of these prior studies did have large sample sizes and utilized random or probability sampling, most did not have a sample size similar to that in the present analysis, and none utilized a community-defined and –directed sampling procedure such as that utilized by the HCHS/SOL to ensure recruitment of a sample representative of a particular minority population.

Study Limitations

Limitations of the present study should be considered. First, the data were cross-sectional, precluding determination of causality. Second, it is possible that environmental factors such as access to healthy food options and family members’ health behaviors, which were not accounted for directly, may have impacted cancer-related risk behaviors. Furthermore, due to sample size restrictions, it was not possible to evaluate whether the relationships of perceived risk and perceived efficacy to cancer-related risk behaviors were consistent across Hispanic/Latino background groups. Finally, results cannot be generalized to the U.S. Hispanic/Latino population at large. However, it is important to note that the community-based probability sample strategies utilized in the HCHS/SOL do enable inferences to the larger Hispanic/Latino populations in the four HCHS/SOL field centers from which the present sample was drawn (LaVange et al., 2010). The HCHS/SOL cohort has good representation of various Hispanic/Latino background groups. The majority of the Hispanic/Latino population in the United States lives in urban areas, and the four field centers are in cities with large Hispanic/Latino populations (ranking of these cities among the metropolitan areas in the US with largest Hispanic population are New York #1, Chicago #5, San Diego #9, and Miami #11 (US Census Bureau, 2010). Florida is the state with the largest population of Cubans (68% of all Cubans), and Miami-Dade County is the county with highest proportion of Hispanics in Florida.

Conclusions and Implications for Health Promotion Practice

In sum, the present findings identified differences across Hispanic/Latino background groups regarding perceptions of which behaviors are risky for cancer, and perceived efficacy for preventing cancers. Ways in which Hispanics/Latinos have different perceptions of cancer risk as compared to the general U.S. population were also clarified. These results are consistent with prior research suggesting that Hispanic/Latino background groups should be viewed distinctly, and expand upon the existent literature by identifying which cancer-related beliefs are discrepant across different background groups, and which background groups may be at high risk for specific cancer-related beliefs. The present findings also suggest that targeting perceived risk and perceived efficacy may not effectively diminish cancer-related risk behaviors among community-dwelling Hispanic/Latino adults, although further research involving prospective designs is needed.

Acknowledgments

The Hispanic Community Health Study/Study of Latinos was carried out as a collaborative study supported by contracts from the National Heart, Lung, and Blood Institute (NHLBI) to the University of North Carolina (N01-HC65233), University of Miami (N01-HC65234), Albert Einstein College of Medicine (N01-HC65235), Northwestern University (N01-HC65236), and San Diego State University (N01-HC65237). The following Institutes/Centers/Offices contribute to the HCHS/SOL through a transfer of funds to the NHLBI: National Center on Minority Health and Health Disparities, the National Institute of Deafness and Other Communications Disorders, the National Institute of Dental and Craniofacial Research, the National Institute of Diabetes and Digestive and Kidney Diseases, the National Institute of Neurological Disorders and Stroke, and the Office of Dietary Supplements. Rina S. Fox was supported through a National Cancer Institute Training Grant (5T32CA193193). The authors thank the staff and participants of HCHS/SOL and the HCHS/SOL Sociocultural Ancillary Study for their important contributions.

Footnotes

Declaration of Conflicting Interests

The Authors declare that there is no conflict of interest.

References

  1. Ajzen I. The theory of planned behavior. Organizational Behavioral and Human Decision Processes. 1991;50:179–211. [Google Scholar]
  2. Ajzen I, Fishbein M. Understanding Attitudes and Predicting Social Behavior. Englewood Cliffs, NJ: Prentice-Hall; 1980. [Google Scholar]
  3. Amercian Cancer Society. Cancer Treatment and Survivorship Facts & Figures 2014–2015. Atlanta: American Cancer Society; 2014. [Google Scholar]
  4. American Cancer Society. Cancer Facts & Figures for Hispanics/Latinos 2015–2017. Atlanta: American Cancer Society; 2015. [Google Scholar]
  5. Bandura A. Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs, NJ: Prentice-Hall; 1986. [Google Scholar]
  6. Bull FC, Maslin TS, Armstrong T. Global Physical Activity Questionnaire (GPAQ): Nine country reliability and validity study. Journal of Physical Activity and Health. 2009;6:790–804. doi: 10.1123/jpah.6.6.790. [DOI] [PubMed] [Google Scholar]
  7. Caban A, Walker EA. A systematic review of research on culturally relevant issues for Hispanics with diabetes. The Diabetes Educator. 2006;32:584–595. doi: 10.1177/0145721706290435. [DOI] [PubMed] [Google Scholar]
  8. Cameron LD. llness risk representations and motivations to engage in protective behavior: The case of skin cancer risk. Psychology and Health. 2008;23:91–112. doi: 10.1080/14768320701342383. [DOI] [PubMed] [Google Scholar]
  9. Cantor D, Covell J, Davis T, Park I, Rizzo L. Health Information National Trends Survey 2005 (HINTS 2005): Final report. Bethesda, MD: National Cancer Institute; 2005. [Google Scholar]
  10. Daviglus ML, Talavera GA, Aviles-Santa ML, Allison M, Cai J, Criqui MH, … Stamler J. Prevalence of major cardiovascular risk factors and cardiovascular diseases among Hispanic/Latino individuals of diverse backgrounds in the United States. JAMA. 2012;308:1775–1784. doi: 10.1001/jama.2012.14517. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Enders CK. Applied Missing Data Analysis. New York: Guilford Press; 2010. [Google Scholar]
  12. Enders CK, Bandalos DL. The relative performance of full information maximum likelihood estimation for missing data in structural equation models. Structural Equation Modeling. 2001;8:430–457. [PubMed] [Google Scholar]
  13. Gallo LC, Penedo FJ, Carnethon M, Isasi CR, Sotres-Alvarez D, Malcarne VL, … Talavera GT. The Hispanic Community Health Study/Study of Latinos Sociocultural Ancillary Study: Sample, design, and procedures. Ethnicity and Disease. 2014;24:77–83. [PMC free article] [PubMed] [Google Scholar]
  14. Gorin SS, Heck JE. Cancer screening among Latino subgroups in the United States. Preventive Medicine. 2005;40:515–526. doi: 10.1016/j.ypmed.2004.09.031. [DOI] [PubMed] [Google Scholar]
  15. Hamilton K, White KM, Young RM, Hawkes AL, Starfelt LC, Leske S. Identifying critical sun-protective beliefs among Australian adults. Health Education Research. 2012;27:834–843. doi: 10.1093/her/cys093. [DOI] [PubMed] [Google Scholar]
  16. Harnack L, Block G, Subar A, Lane S, Brand R. Association of cancer prevention-related nutrition knowledge, beliefs, and attitudes to cancer prevention dietary behavior. Journal of the American Dietetic Association. 1997;97:957–965. doi: 10.1016/S0002-8223(97)00231-9. [DOI] [PubMed] [Google Scholar]
  17. Hawkins NA, Berkowitz Z, Peipins LA. What does the public know about preventing cancer? Results from the Health Information National Trends Survey (HINTS) Health Education & Behavior. 2010;37:490–503. doi: 10.1177/1090198106296770. [DOI] [PubMed] [Google Scholar]
  18. Humes KR, Jones NA, Ramirez RR. Overview of race and Hispanic origin: 2010. 2011 Retrieved from http://www.census.gov/prod/cen2010/briefs/c2010br-02.pdf.
  19. Kristal AR, Hedderson MM, Patterson RE, Neuhauser ML. Predictors of self-initiated, healthful dietary change. Journal of the American Dietetic Association. 2001;101:762–766. doi: 10.1016/s0002-8223(01)00191-2. [DOI] [PubMed] [Google Scholar]
  20. LaVange LM, Kalsbeek WD, Sorlie PD, Avilés-Santa LM, Kaplan RC, Barnhart J, … Ryan J. Sample design and cohort selection in the Hispanic Community Health Study/Study of Latinos. Annals of Epidemiology. 2010;20:642–649. doi: 10.1016/j.annepidem.2010.05.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Lykins EL, Graue LO, Brechting EH, Roach AR, Gochett CG, Andrykowski MA. Beliefs about cancer causation and prevention as a function of personal and family history of cancer: A national, population-based study. Psychooncology. 2008;17:967–974. doi: 10.1002/pon.1306. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Marin G, Sabogal F, Marin BV, Otero-Sabogal R, Perez-Stable EJ. Development of a short acculturation scale for Hispanics. Hispanic Journal of Behavioral Sciences. 1987;9:183–205. [Google Scholar]
  23. National Cancer Institute. Health Information National Trends Survey (HINTS) 2003 Retrieved from http://hints.cancer.gov/default.aspx.
  24. Nelson D, Kreps G, Hesse B, Croyle R, Willis G, Arora N, … Alden S. The Health Information National Trends Survey (HINTS): Development, design, and dissemination. Journal of Health Communication. 2004;9:443–460. doi: 10.1080/10810730490504233. [DOI] [PubMed] [Google Scholar]
  25. Niederdeppe J, Levy AG. Fatalistic beliefs about cancer prevention and three prevention behaviors. Cancer Epidemiology Biomarkers & Prevention. 2007;16:998–1003. doi: 10.1158/1055-9965.EPI-06-0608. [DOI] [PubMed] [Google Scholar]
  26. Passel JS, Cohn D, Lopez MH. Hispanics account for more than half of nation’s growth in past decade. 2011 Retrieved from http://www.pewhispanic.org/2011/03/24/hispanics-account-for-more-than-half-of-nations-growth-in-past-decade/
  27. Patterson RE, Kristal AR, Lynch JC, White E. Diet-cancer related beliefs, knowledge, norms, and their relationship to healthful diets. Journal of Nutrition Education. 1995;27:86–92. [Google Scholar]
  28. Patterson RE, Kristal AR, White E. Do beliefs, knowledge, and perceived norms about diet and cancer predict dietary change? American Journal of Public Health. 1996;86:1394–1400. doi: 10.2105/ajph.86.10.1394. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Prochaska JO, DiClemente CC. Stages and processes of self-change of smoking: Toward an integrative model of change. Journal of Consulting and Clinical Psychology. 1983;51:390–395. doi: 10.1037//0022-006x.51.3.390. [DOI] [PubMed] [Google Scholar]
  30. Ramirez AG, Suarez L, Laufman L, Barroso C, Chalela P. Hispanic women’s breast and cervical cancer knowledge, attitudes, and screening behaviors. American Journal of Health Promotion. 2000;14:292–300. doi: 10.4278/0890-1171-14.5.292. [DOI] [PubMed] [Google Scholar]
  31. Ramirez AS, Finney Rutten LJ, Vanderpool RC, Moser RP, Hesse BW. Correlates and geographic patterns of knowledge that physical activity decreases cancer risk. Journal of Primary Prevention. 2013;34:31–39. doi: 10.1007/s10935-012-0289-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Rimal RN. Perceived risk and self-efficacy as motivators: Understanding individuals’ long-term use of health information. Journal of Communication. 2001;51:633–654. [Google Scholar]
  33. Rimal RN, Real K. Perceived risk and efficacy beliefs as motivators of change. Human Communication Research. 2003;29:370–399. [Google Scholar]
  34. Rosenstock IM. Historical origins of the health belief model. Health Education & Behavior. 1974;2:328–335. [Google Scholar]
  35. Rosenstock IM. Why people use health services. Milbank Quarterly. 2005:83. Online-only. [PubMed] [Google Scholar]
  36. Siega-Riz AM, Sotres-Alvarez D, Ayala GX, Ginsberg M, Himes JH, Liu K, … Rodriguez B. Food-group and nutrient-density intakes by Hispanic and Latino backgrounds in the Hispanic Community Health Study/Study of Latinos. The American Journal of Clinical Nutrition. 2014;99:1487–1498. doi: 10.3945/ajcn.113.082685. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Sorlie PD, Avilés-Santa LM, Wassertheil-Smoller S, Kaplan RC, Daviglus ML, Giachello AL, … Allison M. Design and implementation of the Hispanic Community Health Study/Study of Latinos. Annals of Epidemiology. 2010;20:629–641. doi: 10.1016/j.annepidem.2010.03.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Sullivan HW, Rutten L, Hesse BW, Moser RP, Rothman AJ, McCaul KD. Lay representations of cancer prevention and early detection: Associations with prevention behaviors. Preventing Chronic Disease. 2010;7:A14. [PMC free article] [PubMed] [Google Scholar]
  39. Sullivan HW, Beckjord EB, Rutten LJF, Hesse BW. Nutrition-related cancer prevention cognitions and behavioral intentions: Testing the Risk Perception Attitude Framework. Health Education & Behavior. 2008;35:866–879. doi: 10.1177/1090198108326164. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. van Assema P, Pieterse M, Kok G, Eriksen M, de Vries H. The determinants of four cancer-related risk behaviours. Health Education Research. 1993;8:461–472. doi: 10.1093/her/8.4.461. [DOI] [PubMed] [Google Scholar]
  41. Watters JL, Satia JA. Psychosocial correlates of dietary fat intake in African-American adults: a cross-sectional study. Nutrition Journal. 2009;8:15. doi: 10.1186/1475-2891-8-15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Watters JL, Satia JA, Galanko JA. Associations of psychosocial factors with fruit and vegetable intake among African-Americans. Public Health Nutrition. 2007;10:701–711. doi: 10.1017/S1368980007662284. [DOI] [PubMed] [Google Scholar]
  43. Weinick RM, Jacobs EA, Stone LC, Ortega AN, Burstin H. Hispanic healthcare disparities: Challenging the myth of a monolithic Hispanic population. Medical Care. 2004;42:313–320. doi: 10.1097/01.mlr.0000118705.27241.7c. [DOI] [PubMed] [Google Scholar]
  44. Zsembik BA, Fennell D. Ethnic variation in health and the determinants of health among Latinos. Social Science & Medicine. 2005;61:53–63. doi: 10.1016/j.socscimed.2004.11.040. [DOI] [PubMed] [Google Scholar]

RESOURCES