Kevin D. McCaul

Meta-Analysis of the Relationship Between Risk Perception and Health Behavior: The Example of Vaccination

BACKGROUND Risk perceptions are central to many health behavior theories. However, the relationship between risk perceptions and behavior, muddied by instances of inappropriate assessment and analysis, often looks weak. METHOD A meta-analysis of eligible studies assessing the bivariate association between adult vaccination and perceived likelihood, susceptibility, or severity was conducted. RESULTS Thirty-four studies met inclusion criteria (N = 15,988). Risk likelihood (pooled r = .26), susceptibility (pooled r = .24), and severity (pooled r = .16) significantly predicted vaccination behavior. The risk perception-behavior relationship was larger for studies that were prospective, had higher quality risk measures, or had unskewed risk or behavior measures. CONCLUSIONS The consistent relationships between risk perceptions and behavior, larger than suggested by prior meta-analyses, suggest that risk perceptions are rightly placed as core concepts in theories of health behavior.

WHAT IS THE RELATIONSHIP BETWEEN BREAST CANCER RISK AND MAMMOGRAPHY SCREENING? A META-ANALYTIC REVIEW

This meta-analytic review addresses the issue of how a womans risk of breast cancer relates to the likelihood that she will obtain mammography screenings. Studies that compared women with or without a family history of breast cancer (n = 19) showed that women with a family history were more likely to have been screened. Studies that measured perceived risk (n = 19) showed that feeling vulnerable to breast cancer was positively related to having obtained a screening. Studies that compared women who did or did not have a history of breast problems (n = 10) showed that those with a positive history were more likely to have been screened. Finally, studies that measured worry (n = 6) showed that greater worry was related to higher screening levels. Taken together, these data suggest that increasing perceptions of personal vulnerability may increase screening behavior for breast cancer.

Supportive and Nonsupportive Family Behaviors: Relationships to Adherence and Metabolic Control in Persons with Type I Diabetes

The Diabetes Family Behavior Checklist (DFBC) was administered to 54 adults and 18 adolescents (<19 yr of age) with insulin-dependent diabetes mellitus (IDDM). Subjects and family members completed parallel forms of the DFBC at initial and 6-mo follow-up home interviews. During each of these periods, adherence was assessed via self-report, 1 wk of self-monitoring, and 24-h dietary recalls. The results showed reliable differences between adolescents and adults. More negative interactions with family members were reported by adolescents and their family members, and adolescents were in poorer metabolic control. For adults but not adolescents, negative DFBC scores were prospectively predictive of poorer regimen adherence over the 6-mo interval for measures of glucose testing, insulin injection, and dietary adherence. In addition, higher negative DFBC scores for adults were marginally associated with higher HbA1 levels (P < 0.10). We conclude that the DFBC is a promising measure of family interaction related specifically to the IDDM regimen and that, for adults, higher levels of nonsupportive family behaviors may be related to reduced regimen adherence and poor control.

Risk Perceptions: Assessment and Relationship to Influenza Vaccination

OBJECTIVE Accurate measurement of beliefs about risk probability is essential to determine what role these beliefs have in health behavior. This study investigated the ability of several types of risk perception measures and of other constructs from health behavior theories to predict influenza vaccination. DESIGN Prospective study in which students, faculty, and staff at 3 universities (N = 428) were interviewed in the fall, before influenza vaccine was available, and again early in the next calendar year. MAIN OUTCOME MEASURE Self-reported influenza vaccination. RESULTS Two interview questions that asked about feeling at risk and feeling vulnerable predicted subsequent behavior better (r = .44, p = .001) than 2 questions that asked for agreement or disagreement with statements about risk probability (r = .25, p = .001) or 4 questions that asked respondents to estimate the magnitude of the risk probability (r = .30, p = .001). Of the 4 perceived risk magnitude scales, a 7-point verbal scale was the best predictor of behavior. Anticipated regret was the strongest predictor of vaccination (r = .45, p = .001) of all constructs studied, including risk perceptions, worry, and perceived vaccine effectiveness. CONCLUSION Risk perceptions predicted subsequent vaccination. However, perceived risk phrased in terms of feelings rather than as a purely cognitive probability judgment predicted better. Because neither feeling at risk nor anticipated regret is represented in the most commonly used theories of health behavior, the data suggest that these theories are missing important constructs.

BREAST CANCER WORRY AND SCREENING : SOME PROSPECTIVE DATA

Breast cancer concerns were measured among 353 women, ages 40-75, from North Dakota. One year later, participants were recontacted and asked about their screening behavior during the previous year. Greater concern about breast cancer, even the highest level of concern, was related to a higher likelihood that women performed breast self-examination, had a mammography screening, and had a clinical breast examination. These data do not support the idea that worry inhibits action; instead, they suggest that nonpathological worry motivates self-protective behavior.

Adherence to IDDM Regimens: Relationship to Psychosocial Variables and Metabolic Control

Thirty-four adolescents (ages 12–14 yr) with IDDM completed a questionnaire assessing regimen adherence over the previous week and psychosocial measures potentially related to adherence. Four aspects of the IDDM regimen were studied: insulin injections, dietary patterns, glucose testing, and exercise. Psychosocial variables included (1) Social Learning Theory measures of diabetes-specific family behaviors and barriers to adherence and (2) more general measures of family interaction. Glycosylated hemoglobin levels were predicted accurately (R = 0.68) from a combination of three adherence measures. The psychosocial measures were not directly related to metabolic control, but they were associated with adherence. Degree of adherence to one aspect of the IDDM regimen was not related to adherence to other aspects of the regimen and different psychosocial variables predicted adherence to different regimen components. The diabetes-specific measures were generally more predictive of adherence than were the more global measures. Implications and limitations of this cross-sectional, correlational study were discussed.

Diabetes regimen behaviors. Predicting adherence.

In this study, Social Learning Theory was used to generate psychosocial predictors of regimen adherence among persons with insulin-dependent diabetes mellitus (IDDM). Four categories of predictors were formed: knowledge, expectancies, skills, and environmental suuport. Persons with IDDM (84 adults and 23 adolescents) completed measures within each predictor category on two occasions, separated by 6 months. Multimethod assessment procedures (selfmonitoring, interviews, mechanical devices) were used to measure adherence to four aspects of the IDDM regimen: insulin injections, glucose testing, diet, and exercise. Bivariate analyses demonstrated that measures of expectancies and environmental support were reliably related to several adherence behaviors. In particular, self-efficacy expectations were related to adherence across almost all regimen areas for both adolescents and adults. Skills were also important, though only for adolescents, whereas regimen knowledge was unrelated to adherence for either adults or adolescents. Multivariate analyses showed that expectancies were consistently predictive of adherence and that, in general, psychosocial variables were better predictors of insulin administration and glucose testing than of diet or exercise. The implications of these findings for diabetes education programs are discussed.

Barriers to regimen adherence among persons with insulin-dependent diabetes

The present study was designed to assess the frequency of different barriers to adherence among persons with insulin-dependent diabetes mellitus (IDDM) and to determine the relationship between such barriers and adherence to insulin injection, glucose testing, and dietary and exercise components of the regimen. The behavior analytic model was used to develop a Barriers to Adherence Scale for adults and adolescents with IDDM. Sixty-five outpatient subjects were then interviewed in their home and reassessed 6 months later. Subjects reported the greatest number of barriers to dietary and exercise adherence and the fewest barriers to insulin injections. Females reported more barriers than males, but age was not associated with barriers scores. Self-report measures of adherence revealed consistent relationships between barriers and all areas of regimen adherence assessed. Adherence indices based on self-monitoring and dietary recall data correlated less consistently with barriers scores. We conclude that the behavior analytic model has great potential for use in the development of psychosocial assessment procedures for diabetes.

Attention, distraction, and cold-pressor pain.

This article compares the effects of distracting oneself from, versus attending to, the sensations produced by cold-pressor stimulation. Experiment 1 revealed that distraction is a better coping strategy than attention to sensations when subjects are asked to report pain threshold and tolerance. Experiments 2 and 3 examined the hypothesis that distraction is effective because persons hold a commonsense belief in the benefits of distraction as a coping device. Neither experiment supported the commonsense hypothesis as an explanation for the findings of Experiment 1. In a final experiment, subjects were assigned to either a distraction, attention, or no-instructions condition and were asked to report their distress during a 4-minute cold-pressor trial. Distraction reduced distress early in the trial, but attention to sensations proved to be a superior strategy for the last 2 minutes of the trial. It is proposed that distraction and attention to sensations may be differentially effective depending on the duration of the painful stimulus. Possible mediating processes underlying the two strategies are discussed.

Winning, losing, mood, and testosterone

In two experiments, male college students either won or lost