Lorem, Ipsum

Healthpromotionbehaviors16.pdf

Home >Psychology homework help >Lorem, Ipsum

CHAPTER 7 HEALTH PROMOTION BEHAVIORS OVERVIEW Health promotion behaviors, including physical activity, healthy eating, donating blood, safe driving and screening behaviors, positively influence health and are the focus of this chapter. In Chapter 8 we go on to examine health risk behaviors such as smoking and unhealthy food consumption. A key distinction between these types of behaviors is that we are usually trying to make people approach or increase health promotion behaviors and avoid or decrease health risk behaviors. Nevertheless the ways used to understand and change these two types of health behaviors show some degree of overlap. In this chapter, we take a look at the research that has examined the determinants (predictors) of these behaviors taken largely from the health/social cognition models that we introduced in Chapter 2. In relation to behavior change the assumption is that identified predictors can be targeted and changed by interventions as a way to change behavior. In the second part of the chapter we examine research that has attempted to encourage performance of health promotion behaviors in order to promote health outcomes based on three promising approaches. In particular we look at how research targeting changes in attitudes or self-efficacy can change health promotion behavior; at how simply asking questions about an individual’s plans to perform these behaviors can have an impact; and at how forming more specific plans can help promote performance of these behaviors. In the final section of the chapter, we introduce the Science of Health Behavior Change: In Action feature which is used to evaluate a study testing a behavior change intervention to promote health behavior. This feature illustrates how the scientific, critical approach that we introduced in the first six chapters can be applied in the context of health promotion behaviors.

Prestwich, Andrew, et al. Health Behavior Change : Theories, Methods and Interventions, Taylor & Francis Group, 2017. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/scu/detail.action?docID=5056486. Created from scu on 2022-05-09 03:16:31.

C o p yr

ig h t ©

2 0 1 7 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

144 HEALTH PROMOTION BEHAVIORS

USING SOCIAL/HEALTH COGNITION MODELS TO PREDICT HEALTH PROMOTION BEHAVIORS

The prevalence of health behaviors varies across different social groups defined by sex, age or social class. For example, in the Western world smoking is generally more prevalent among those from economically disadvantaged backgrounds. This finding would suggest socio-demographic factors as a focus of interventions to change health behaviors. However, socio-demographic factors are often impossible to change or require political intervention at national or international levels (e.g., change in income distribution). This is one reason why research has tended to focus on more modifiable factors assumed to mediate or explain the relationship between socio-demographic factors and health-related behaviors. One important set of factors are the thoughts and feelings the individual associates with performing a particular health behavior. These are commonly referred to as health or social cognitions. Take Mark, one of the authors of this book; he says he really enjoys exercising and thinks it is healthy, but does not perceive his friends to be interested in his exercise behavior. Each of these different health cognitions could be key in making Mark exercise. A group of social/health cognition models (SCMs; Conner & Norman, 2015) have been developed to assess the main health cognitions for people in general across a range of behaviors. These SCMs specify the theory-relevant constructs or health cognitions and how they interrelate in determining behavior. We broadly covered and critiqued these models in Chapter 2. Here, we focus more deeply on the most popular models: the Health Belief Model (HBM; e.g., Janz & Becker, 1984), Protection Motivation Theory (PMT; Maddux & Rogers, 1983), Theory of Reasoned Action/Theory of Planned Behavior (TRA/TPB; Ajzen, 1991) and Social Cognitive Theory (SCT; Bandura, 2000). As we will see, each model suggests different health cognitions that should be central to making Mark (or anyone else) exercise.

BURNING ISSUE BOX 7.1

ARE DIFFERENCES IN HEALTH BEHAVIORS ACROSS SOCIO-ECONOMIC STATUS GROUPS REFLECTED IN STUDIES TESTING SOCIAL/HEALTH COGNITION MODELS? Demographic differences (e.g., age, gender, socio-economic status) between individuals have been found to significantly moderate the intention-behavior relationship. Effects for socio-economic status are particularly interesting here because of differences in performance of health behaviors across socio-economic status groups. A variety of studies have found higher levels of engagement with health protective behaviors such as physical activity and healthy eating in higher socio-economic status groups. This could be because of weaker intentions to engage in such behaviors in lower socio-economic status groups, although there is not strong evidence to support this view. A more interesting possibility is that lack of resources available to lower socio-economic status individuals interferes with their ability to translate healthy intentions into healthy behaviors.

C o p yr

ig h t ©

2 0 1 7 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

HEALTH PROMOTION BEHAVIORS 145

HEALTH BELIEF MODEL

The Health Belief Model (HBM) is the oldest and most widely used SCM (see Abraham & Sheeran, 2015). In an early study in this area Hochbaum (1958) reported that perceived susceptibility to tuberculosis and the belief that people with the disease could be asymptomatic (so that screening would be beneficial) distinguished between attendees and non-attendees for chest X-rays. Later, Haefner and Kirscht (1970) extended this research by demonstrating that an intervention designed to increase participants’ perceived susceptibility, perceived severity and anticipated benefits resulted in a greater number of check-up visits to the doctor over an eight-month period compared to a control condition.

A brief description of the HBM is provided in Chapter 2 but, to re-iterate, the HBM suggests that health behavior is determined by two sets of cognitions: perceptions of illness threat and evaluation of behaviors to counteract this threat (see Figure 2.6). Threat perceptions are themselves based on two beliefs: first, the perceived susceptibility of the individual to the illness (‘How likely am I to get ill?’); and second, the perceived severity of the consequences of the illness for the individual (‘How serious would the illness be?’). Similarly, evaluation of behaviors to counteract the threat involves consideration of: first, the potential benefits of performing the behavior; and second, consideration of the barriers or costs to performing the behavior. Together these four beliefs are believed to determine the likelihood of the individual performing a health behavior. So the HBM suggests that individuals are most likely to follow a particular health action if they believe themselves to be susceptible to a particular condition which they also consider to be serious, and believe that the benefits outweigh the costs of the behavior engaged in to counteract the health threat. Many applications of the HBM also include the individual’s overall motivation to protect their health (i.e., health motivation) and any cues that might prompt action (i.e., cues to action) and some later revisions added the concept of self-efficacy.

A main strength of the HBM is the common-sense operationalization it uses in including key beliefs related to decisions about health behaviors. A significant weakness has been the omission of important cognitions such as intentions and self-efficacy that subsequent research has shown to be influential in predicting health behaviors (for a more detailed critique, see Chapter 2). A range of studies have applied the HBM to various different health behaviors. For example, Zhao et al. (2012) used the HBM to look at condom use in female Chinese sex workers. Barriers, self-efficacy and benefits were each significant independent predictors of condom use with the strongest effects being associated with perceived barriers. Moreover, Ar-yuwat, Clark, Hunter and James

Such a moderating effect of socio-economic status on the intention-behavior relationship has been shown for health promotion behaviors such as breastfeeding and physical activity (Conner et al., 2013). In each case the relationship between intentions and behavior was weaker in lower compared to higher socio-economic status groups. This finding could help explain why those from lower socio- economic status groups engage in fewer health promotion behaviors.

C o p yr

ig h t ©

2 0 1 7 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

146 HEALTH PROMOTION BEHAVIORS

(2013) used the HBM to predict physical activity in Thai primary school students. Only barriers (self-efficacy was not assessed) were significantly related to physical activity levels. Across studies of the HBM it is susceptibility and barriers (plus self-efficacy when it is measured) that emerge as the strongest predictors of subsequent behavior (Abraham & Sheeran, 2015). This would suggest that interventions to change behavior should focus on changing these constructs.

PROTECTION MOTIVATION THEORY

Protection Motivation Theory (PMT; Maddux & Rodgers, 1983; see Norman, Boer, Seydel and Mulle, 2015, for a review) is overviewed in Chapter 2 but, in summary, it is a revision and extension of the HBM incorporating various improvements. In PMT, the primary determinant of performing a health behavior is the intention to perform a health behavior (labeled protection motivation here; see Figure 2.5). Intention is determined by six health cognitions. Several are familiar from the HBM. The six health cognitions are: perceptions of vulnerability to the illness (labeled susceptibility in the HBM); perceived severity of the health threat; fear about the health threat; expectancy that carrying out a behavior can remove the threat (response efficacy); belief in one’s capability to successfully execute the recommended courses of action (self-efficacy); and the perceived costs of adopting the behavior (response costs).

PMT has been used to predict a range of health promoting (e.g., physical activity and diet) and health risking (e.g., smoking and alcohol consumption) behaviors. As with the HBM, some studies have highlighted certain PMT constructs to be important determinants of behavior, while other studies suggest different PMT constructs are more important. For example, Plotnikoff and Higginbotham (1998) applied PMT to the prediction of exercise and dietary intentions and behavior among a group of patients who had recently experienced a myocardial infarction or angina. Self-efficacy emerged as the only PMT variable with a significant effect on exercise intentions and intentions were the only significant predictor of exercise behavior. For eating a low fat diet, self- efficacy was the only significant predictor of intentions, while intentions, perceived vulnerability and fear were significant predictors of behavior. However, Abraham, Sheeran, Abrams and Spears (1994) found that self-efficacy and response costs were predictive of condom use intentions among a sample of male and female adolescents. Moreover, sometimes none of the PMT constructs predict health promoting behaviors. For example, Boer and Seydel (1996) used PMT variables to predict attendance at breast cancer screening by mammography. Response efficacy and self-efficacy were predictive of screening intentions but none of the PMT measures predicted attendance at screening at two-year follow-up.

SOCIAL COGNITIVE THEORY

Social Cognitive Theory (SCT; Bandura, 1982; see Luszczynska & Schwarzer, 2015, for a review) was described and evaluated in Chapter 2. As a reminder, in SCT, behavior is determined by three factors: goals, outcome expectancies and self-efficacy (see Figure 2.7). Goals are plans to act and are very similar to intentions to perform

C o p yr

ig h t ©

2 0 1 7 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

HEALTH PROMOTION BEHAVIORS 147

the behavior (see Luszczynska & Schwarzer, 2015). Outcome expectancies are beliefs about the perceived likelihood of different outcomes of performing the behavior and are split into physical, social and self-evaluative depending on the nature of the outcomes considered. Self-efficacy is the belief that a behavior is or is not within an individual’s control and is usually assessed as the degree of confidence the individual has that they could still perform the behavior in the face of various obstacles (e.g., ‘I am confident I can eat healthily even when out with friends’). Bandura (2000) also includes socio-structural factors to his theory. These are factors assumed to facilitate or inhibit the performance of a behavior and affect behavior via changing goals. Socio-structural factors refer to the impediments or opportunities associated with particular living conditions, health systems, political, economic or environmental systems. They are assumed to inform goal setting and be influenced by self-efficacy. The latter relationship arises because self- efficacy influences the degree to which an individual pays attention to opportunities or impediments in their life circumstances. This component of the model incorporates perceptions of the environment as an important influence on health behaviors.

SCT has also been applied to a range of health promoting behaviors, although in many applications not all components of the SCT are examined. For example, Williams and Bond (2002) showed that among diabetics self-efficacy and outcome expectancies were associated with patients’ compliance with blood glucose testing. Similarly, Dilorio, Dudley, Lehr and Soet (2000) showed condom use in sexually active college students was predicted by self-efficacy and outcome expectancies. Studies using objective measures of physical activity, such as motion detectors, have shown that self-efficacy is related to a high level of physical activity among 10- to 16-year-old adolescents (Strauss, Rodzilsky, Burack and Colin, 2001).

THEORY OF PLANNED BEHAVIOR

The Theory of Planned Behavior (TPB; Ajzen, 1991) was developed to explain a range of social behaviors but has been particularly widely applied in relation to health behaviors (see Conner & Sparks, 2015, for a review). It was outlined in Chapter 2 and spells out the cognitions that determine that individual’s decision to perform a particular behavior (see Figure 2.3). Importantly this theory added ‘perceived behavioral control’ to the earlier Theory of Reasoned Action (TRA; Ajzen & Fishbein, 1980). The TPB proposes that the key determinants of behavior are intention to engage in that behavior and perceived behavioral control over that behavior (which influences the relationship between intention and behavior). As in the PMT, intentions in the TPB represent a person’s motivation or conscious plan or decision to exert effort to perform the behavior. Perceived behavioral control (PBC) is a person’s expectancy that performance of the behavior is within their control and confidence that they can perform the behavior and is similar to Bandura’s (1982) concept of self-efficacy (as in the PMT).

In the TPB, intention is assumed to be determined by three factors: attitudes, subjective norms and PBC. Attitudes are the overall evaluations of the behavior by the individual as positive or negative. Subjective norms are a person’s beliefs about whether significant others think they should engage in the behavior. PBC is assumed to influence both intentions and behavior because we rarely intend to do things we know we cannot

C o p yr

ig h t ©

2 0 1 7 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

148 HEALTH PROMOTION BEHAVIORS

do and because believing that we can succeed enhances effort and persistence and so makes successful performance more likely. In turn, attitudes are based on behavioral beliefs (or outcome expectancies), that is, beliefs about the perceived outcomes of a behavior. In particular, they are a function of the likelihood of the outcome occurring as a result of performing the behavior (e.g., ‘How likely is this outcome?’) and the evaluation of that outcome (e.g., ‘How good or bad will this outcome be for me?’). It is assumed that an individual will have a limited number of consequences in mind when considering performing a behavior or not (i.e., only a few outcomes will be salient). This expectancy-value framework is based on Fishbein’s (1967) earlier summative model of attitudes. Subjective norm is based on beliefs about salient others’ approval or disapproval of whether one should engage in a behavior (e.g., ‘Would my best friend want me to do this?’) weighted by the motivation to comply with each salient other on this issue (e.g., ‘Do I want to do what my best friend wants me to do?’). Again it is assumed that an individual will only have a limited number of individuals or groups (often referred to as referents) in mind when considering performing a behavior. PBC is based on control beliefs concerning whether an individual has access to the necessary resources and opportunities to perform the behavior successfully (e.g., ‘How often does this facilitator/inhibitor occur?’), weighted by the perceived power, or importance, of each factor to facilitate or inhibit the action (e.g., ‘How much does this facilitator/ inhibitor make it easier or more difficult to perform this behavior?’). These factors include both internal control factors (information, personal deficiencies, skills, abilities, emotions) and external control factors (opportunities, dependence on others, barriers). As for the other types of beliefs it is assumed that an individual will only consider a limited number of control factors when considering performing a behavior (see Chapter 2 for a more concise description).

In a review of the TPB as applied to health behaviors McEachan, Conner, Taylor and Lawton (2011) reported the results across 237 prospective tests of the TPB (i.e., when the components of the TPB are measured at one time point and behavior is measured at a later time point preserving the presumed causal ordering of variables). Interestingly the type of behavior moderated the effectiveness of the model. In particular, the TPB did well at predicting physical activity and dietary behaviors (23.9% and 21.2% variance explained respectively; see Chapter 2, Critical Skills Toolkit 2.1, to see an account of the concept of explaining variance) but less well at predicting risk, detection, safer sex and drug abstinence behaviors (between 13.8% to 15.3% variance explained). Of greater interest, this meta-analysis showed that the cognitions most strongly associated with intentions also varied as a function of behavior (Table 7.1). In particular, subjective norms were strong predictors of intentions to engage in sex, dietary behaviors and risk behaviors, but weaker predictors of detection behaviors, abstinence behaviors and physical activity. This would suggest the value of targeting subjective norms when trying to change sex, dietary or risk behaviors and also that such a focus would be less useful when trying to change detection, abstinence or physical activity behaviors. There was also some variation in the cognitions most strongly associated with behavior although this was less pronounced (Table 7.2). For example, intentions were strong predictors of physical activity but weaker predictors of safe sex. It is worth thinking about the value of targeting different components of the TPB when trying to change different behaviors.

C o p yr

ig h t ©

2 0 1 7 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

HEALTH PROMOTION BEHAVIORS 149

TABLE 7.1 Relationship between intentions and predictors in TPB (from McEachan et al., 2011)

Predictor Type of health behavior N k r

Attitude A. Risk 14673 29 0.46

B. Detection 8370 17 0.45

C. Physical activity 23905 101 0.51

D. Dietary 9823 30 0.52

E. Safer sex 4958 15 0.51

F. Abstinence 6351 13 0.47

Subjective norms A. Risk 14673 29 0.40

B. Detection 8370 17 0.33

C. Physical activity 23499 100 0.32

D. Dietary 9823 30 0.35

E. Safer sex 4958 15 0.45

F. Abstinence 6351 13 0.33

PBC A. Risk 14673 29 0.43

B. Detection 8370 17 0.45

C. Physical activity 23996 102 0.47

D. Dietary 9823 30 0.44

E. Safer sex 4958 15 0.44

F. Abstinence 6351 13 0.43

Note: N is number of participants; k is number of studies and r is the average correlation. These effect sizes are in the medium to large range (Cohen, 1992, suggests that r = 0.1 equates to a small effect size, r = 0.3 to a medium effect size and r = 0.5 to a large effect size).

TABLE 7.2 Relationship between behavior and predictors in TPB (from McEachan et al., 2011)

Predictor Type of health behavior N k r

Intentions A. Risk 13710 29 0.37

B. Detection 8370 17 0.37

C. Physical activity 23376 103 0.45

D. Dietary 9047 30 0.38

E. Safe sex 2605 15 0.34

F. Abstinence 4406 13 0.35

PBC A. Risk 13713 29 0.22

B. Detection 8370 17 0.20

C. Physical activity 23385 103 0.31

D. Dietary 9047 30 0.30

C o p yr

ig h t ©

2 0 1 7 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

150 HEALTH PROMOTION BEHAVIORS

INTEGRATING CONSTRUCTS FROM DIFFERENT THEORIES

As we noted in Chapter 2 there are overlaps between the constructs in different theories. This opens up the possibility of producing an integrated theory such as those attempted by Fishbein et al. (2001) and Michie et al. (2005) overviewed in Chapter 2. Fishbein and Ajzen (2010) have also produced an integrated model that they call the Reasoned Action Approach (RAA; see Conner & Sparks, 2015). This is mainly based on the TRA and the TPB but is intended to integrate various influences on the performance of a behavior.

In the TPB, intention is the main predictor of behavior and intentions are determined by three variables as noted earlier in this chapter: attitudes, subjective norms and PBC. In the RAA, attitude, subjective norms and PBC each break down into two components (see Figure 7.1). Attitudes are replaced with affective (e.g., the extent the behavior is rated as pleasant-unpleasant; interesting-boring) and instrumental attitudes (e.g., the extent the behavior is rated as valuable-worthless), subjective norms are replaced with injunctive norms (concerning the social approval of others) and descriptive norms (perceptions of what others do), while PBC is replaced with (perceived) self-efficacy (the perceived ease/difficulty of performing the behavior and people’s confidence in performing the behavior should they wish) and perceived control (people’s belief that they have control over the behavior and its performance is up to them). Each of the six components is treated as an independent predictor of intentions (Conner & Sparks, 2015).

Predictor Type of health behavior N k r

E. Safe sex 3674 19 0.21

F. Abstinence 4406 13 0.26

Attitude A. Risk 13713 29 0.27

B. Detection 8370 17 0.22

C. Physical activity 23141 101 0.30

D. Dietary 9046 30 0.29

E. Safe sex 2234 14 0.23

F. Abstinence 4406 13 0.26

Subjective norms A. Risk 13189 29 0.26

B. Detection 8370 17 0.19

C. Physical activity 22849 100 0.18

D. Dietary 9049 30 0.15

E. Safe sex 2235 14 0.21

F. Abstinence 4406 13 0.21

Note: N is number of participants; k is number of studies and r is the average correlation (Cohen, 1992, suggests that r = 0.1 equates to a small effect size, r = 0.3 to a medium effect size and r = 0.5 to a large effect size).

TABLE 7.2 continued

C o p yr

ig h t ©

2 0 1 7 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

HEALTH PROMOTION BEHAVIORS 151

Recent research has supported the RAA as a powerful predictor of health behaviors. For example, a meta-analysis by McEachan et al. (2016) reported the RAA to explain 58.7% of the variance in intentions and 30.9% of the variance in behavior across a range of health behaviors. Interestingly, in addition to intentions and self-efficacy, affective attitudes and descriptive norms emerged as direct predictors of behavior (independent of intentions) suggesting directions for future research. Moreover, studies have also found affective measures of attitudes to be more closely linked to intentions and behavior than instrumental attitudes (Lawton, Conner & McEachan, 2009).

FIGURE 7.1 The Reasoned Action Approach

BURNING ISSUE BOX 7.2

HOW CAN BEING CONSCIENTIOUS IMPROVE HEALTH? It pays to be conscientious! Conscientiousness refers to the ability to control one’s behavior and to complete tasks. Highly conscientious individuals are more organized, careful, dependable, self-disciplined and achievement-oriented

Affective attitudes e.g., Exercising 3 times this week will be fun.

Instrumental attitudes e.g., Exercising 3 times this

week will be healthy

Descriptive norms e.g., People important to me

exercise 3 times a week

Injunctive norms e.g., People important to me

approve of me exercising 3 times this week

Perceived control e.g., I have control over my excercising 3 times

this week

Self-efficacy e.g., I am confident I can exercise 3 times this week

Intention e.g., I intend to excercise

3 times this week

Behavior e.g., I exercised

3 times this last week

C o p yr

ig h t ©

2 0 1 7 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

152 HEALTH PROMOTION BEHAVIORS

CHANGING HEALTH PROMOTION BEHAVIORS

The work on social/health cognition models reviewed in the previous section provides the basis for a better understanding of the determinants of health promotion behaviors. A major assumption underlying behavior change, according to these SCMs, is that changing these determinants (predictors) is a key route to change behavior. For example, promoting intentions and self-efficacy to exercise could help promote exercise behavior and so contribute to improving individuals’ health outcomes. In this section we review research on changing health behaviors through changing attitudes or self-efficacy.

CHANGING ATTITUDES

Attitude change in response to a persuasive message is something we are all familiar with. But what factors influence the amount of attitude change? In general, research

than those low in conscientiousness (McCrae & Costa, 1987). Individuals high

Personality Item Pool: http://ipip.ori.org/).

A growing body of research shows the personality trait conscientiousness to have impacts on health behaviors, health outcomes and even longevity. For example, Friedman et al. (1993) showed that those high in conscientiousness at age 11 were likely to live longer (by about two years) compared to those low in conscientiousness.

An important mechanism by which conscientiousness may influence health is through health behaviors. Friedman et al. (1995) showed that the impact of conscientiousness on longevity was partly accounted for by its effect on reducing smoking and alcohol use. A review of work on the relationship between conscientiousness and behavior (Bogg & Roberts, 2004) showed conscientiousness to be positively related to a range of protective health behaviors (e.g., exercise) considered in this chapter but negatively related to a range of risky health behaviors (e.g., smoking) considered in the next chapter. Other studies have shown this effect of conscientiousness on health behaviors such as physical activity to be explained by impacts on intentions. In some studies intentions have been shown to mediate the impact of conscientiousness on behavior. For example, Conner and Abraham (2001) showed that those with higher levels of conscientiousness also had stronger intentions to exercise and this helped explain the effects of conscientiousness on exercise behavior. Other studies have shown moderation effects. For example, Conner, Rodgers and Murray (2007) showed that intentions were better predictors of exercise behavior among those with high compared to low levels of conscientiousness suggesting highly conscientious individuals are more likely to fulfill their intentions.

C o p yr

ig h t ©

2 0 1 7 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

http://www.ipip.ori.org/

HEALTH PROMOTION BEHAVIORS 153

focusing on changing health behaviors through changing attitudes has tended to focus on developing strong messages to change behavior. Petty and Cacioppo (1986) define strong messages as those that produce mainly favorable thoughts about the message. So, if after reading a message about the benefits of eating five portions of fruit and vegetables a day you have mainly positive thoughts, then your attitude towards eating five portions of fruit and vegetables a day is likely to become more positive. If your reactions are negative or quite mixed little or no attitude change will occur.

Changing attitudes via systematic processing and heuristics

Models of attitude change suggest there are two distinct routes to attitude change. In one route, the information in the persuasive message is systematically and carefully considered and attitude change is determined by the extent to which the message produces mainly favorable thoughts about the message. This route to persuasion is called the central or systematic route. It is what we traditionally think of as persuasion and requires quite a bit of mental effort. The second route to persuasion does not require careful scrutiny of the message or detailed thought and is labeled the peripheral or heuristic route. Here, persuasion depends on the presence of peripheral cues that prompt the use of heuristics. For example, one heuristic is that messages from an expert are more likely to be believed and lead to positive reactions to the message and subsequent attitude change independent of the message content. So if a message is known to come from an expert (e.g., the Chief Medical Officer) it will generally produce more attitude change than the same message from a non-expert. Figure 7.2 sets out the two routes, the factors influencing which route dominates and the consequences of each route for attitude change.

Petty and Cacioppo (1986) argue that because we receive so many messages each day we do not have the motivation or ability to carefully process each one. Petty and Cacioppo refer to the amount of systematic processing devoted to a message as ‘cognitive elaboration’ and, consequently, their model is known as the Elaboration Likelihood Model (ELM) (see Figure 7.2). High elaboration is associated with central route (effortful) processing of messages while low elaboration is associated with peripheral route (effortless) processing. The ELM suggests that both central processing and peripheral processing occur simultaneously for all messages but that usually one or other will dominate. Likelihood of elaboration is determined by both motivation and ability to think about persuasive messages.

When we are highly motivated because the message is about an issue of interest to us we are more likely to elaborate and so engage in central route processing (Petty & Cacioppo, 1986). Ability to think about a message is determined by factors like not having time pressure or distraction. Central route processing involves greater cognitive elaboration and the strength of the arguments in the message is critical to the amount of persuasion that occurs (via making us think positive thoughts about the message). This is consistent with traditional views of how persuasion works: strong arguments will persuade us to change our views; weak arguments will be dismissed and have little impact on attitude change.

When motivation or ability is low (e.g., when messages are presented quickly amid distractions as is the case in many television or radio advertisements), then elaboration

C o p yr

ig h t ©

2 0 1 7 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

154 HEALTH PROMOTION BEHAVIORS

will be less likely and persuasion can occur only through the peripheral route. This route involves little systematic processing (low cognitive elaboration) and other characteristics of the message are more likely to determine whether or not it is persuasive. For example, people use simple rules or decision-making heuristics to evaluate messages (Chaiken, 1980). These include ‘expertise = accuracy’, that is, they are an expert so what they say must be right, or ‘consensus = correctness’, that is, if so many people agree they must be right and ‘length = strength’, that is, there are lots of arguments so it must be true.

Individual differences mean that some people are more or less likely than others to engage in systematic processing. For example, Chaiken (1980) identified people who agreed or disagreed with the ‘length = strength’ heuristic (using agreement with questionnaire items such as, ‘the more reasons a person has for some point of view the more likely they are correct’). These people were then presented with a message containing six arguments in favor of cross-course, end-of-year examinations for students. However, the message was described to participants as either containing ten or two arguments (although it always contained the same six arguments). The results showed that those who endorsed the ‘length = strength’ heuristic were more likely to be persuaded when the message was described as having ten arguments than were those who did not endorse the heuristic.

In general, research using the ELM has focused on both developing strong messages to change behavior and using peripheral cues to change behavior. Attitude change resulting

FIGURE 7.2 The Elaboration Likelihood Model

ABLETO YES ELABORATE?

YES

MOTIVATED TO

ELABORATE?

PERSUASIVE COMMUNICATION

NATUREOF COGNITIVE

PROCESSING

NO PERIPHERAL CUE

PRESENT?

FAVORABLE THOUGHTS

PREDOMINATE

UNFAVORABLE THOUGHTS

PREDOMINATE

NEITHER OR NEUTRAL

PREDOMINATE

YES

RETAIN INITIAL ATIITUDE

COGNITIVE STRUCTURE CHANGE?

PERIPHERAL ATTITUDE SHIFT

(TEMPORARY, SUSCEPTABLE, UNPREDICTIVE)

YES

CENTRAL ATTITUDE CHANGE

(ENDURING, RESISTANT, PREDICTIVE)

C o p yr

ig h t ©

2 0 1 7 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

HEALTH PROMOTION BEHAVIORS 155

from systematic (central route) processing is generally more likely to be stable and more likely to influence subsequent behavior. However, as a persuader if you do not have strong arguments then you are better discouraging systematic processing and relying instead on peripheral cues like numerous arguments, consensus and perceived expertise. So, if your health product or service has a number of features that are likely to be valued by consumers that are different from similar products then developing a strong message about those features may be the best way to win new customers. This sort of persuasive message is common in paper adverts for high cost items like cars and computers where customers have the motivation and opportunity to carefully consider arguments. In contrast, if your product is very similar to other products then using peripheral cues may be a better approach. This sort of persuasive message is common in television adverts for low cost items like washing powder or food products where customers do not have the motivation or opportunity to carefully consider the product. Peripheral cues like associating the product with a happy tune or getting celebrity endorsements are more common here.

The impact of changing affective attitudes

One example of an attempt to change a health promoting behavior through use of persuasive messages is the study by Conner, Rhodes, Morris, McEachan and Lawton (2011b) focusing on exercise behavior. A sample of university students read a definition of exercise and then self-reported their levels of physical activity. Students were then randomized to one of three conditions: control, cognitive message, affective message. In the control condition participants received no further information. In the cognitive message condition participants read messages focusing on the many benefits of exercising for health and well-being. To emphasize that the evidence supporting these claims were strong, citations to scientific articles supporting these findings were provided (see Figure 7.3). In the affective condition participants read messages focusing on the many affective benefits of exercising including making you feel good and being enjoyable. Again all claims were supported by citations to scientific articles supporting expert sources behind these messages (see Figure 7.3). Participants in all three conditions then reported their cognitive and affective attitudes towards exercising along with a number of other measures. Three weeks later respondents reported their levels of exercise in the intervening three-week period. Figure 7.4 shows the pattern of results found. In the control condition participants’ levels of exercise showed a modest decrease from baseline to follow-up. In the cognitive message condition participants’ levels of exercise showed a modest increase from baseline to follow-up. In contrast the participants in the affective message condition showed a significant increase in exercise from baseline to follow-up and were reporting several more exercise sessions per week at follow-up compared to the other two conditions. Importantly, additional analyses showed that changes in behavior were explained or mediated by changes in affective attitudes. The authors argue that the data show the value of targeting affective attitudes as one means to achieve behavior change. It may also be the case that the affective messages were more novel to respondents compared to the cognitive messages that should have been more familiar to respondents through general health messages about the benefits of exercising. This greater novelty might have led to more systematic (i.e., careful) processing of the affective messages resulting in more positive affective attitudes and greater impacts on behavior.

C o p yr

ig h t ©

2 0 1 7 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

FIGURE 7.3A Affective physical activity messages used in Conner et al. (2011b) to promote physical activity

AFFECTIVE CONDITION:

It is well-known that regular physical activity can have a tremendous effect on your immediate well-being. Experts say that adults should engage in approximately 30-60 minutes of accumulated physical activity per day (where each separate bout is a minimum of ten minutes)'. Another position stand is Activity for adults is defined as a daily energy expenditure of 1.5 kilocalories/kilogram of body weight/day or more; roughly equivalent to brisk walking one half hour every day or more)2. Read on for some of the ways physical activity can improve your daily life:

Regular physical activity has been shown to reduce anxiety, depression, and stress which can improve how you feel, your mood, and increase your sense of wellbeing. Research has shown that regular activity improves general reports of quality of life. Physical activity can be associated with increased energy levels and reduced fatigue, giving you more energy and vitality to enjoy your day. Physical activity is an outlet for socializing with friends and creating new social connections for many people through sports, fitness clubs, as weil as with exercise partners and groups. Many types of physical activity provide a fun, enjoyable activity to do in your leisure time. Physical activity often improves the way one feels about their body/appearance, through more positive body image and self-esteem.

Some Specifics From the Research:

85% of studies looking at the acute affects of physical activity on mood showed some degree of improved mood following exercise. 3

Some studies suggest that physical activity can raise endorphin levels, thus decreasing feelings of depression, and elevating mood. 4

After only 20 minutes of moderate to vigorous physical activity, anxiety symptoms have been shown to decrease.5

60% of studies report a positive association between physical activity and self-esteem. 6

1. Public Health Agency of Canada. Physical Activity Unit, retrieved October 31, 2006 from: http://www.phac-aspc.gc.ca/pau-uap/paguide/why.html

2. World Health Organization. Physical Activity, retrieved October 31,2006 from : http://www.who.int/dietphysicalactivity/publications/facts/pa/en/index.html

3. Yeung, R. (1996). The acute affects of exercise on mood state. Journal of Psychosomatic Research, 40,123-141.

4. O'Neal, H., Dunn, A, & Martinsen, E. (2000). Depression and exercise. Journal of Sport Psychology, 31, 110-135.

5. O'Connor, P., Raglin, J., & Martinsen, E. (2000). Physical activity, anxiety, and anxiety disorders. Journal of Sport Psychology, 31, 136-155.

C o p yr

ig h t ©

2 0 1 7 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

http://www.phac-aspc.gc.ca/pau-uap/paguide/why.html

http://www.who.int/dietphysicalactivity/publications/facts/pa/en/index.html

FIGURE 7.3B Instrumental physical activity messages used in Conner et al. (2011b) to promote physical activity

INSTRUMENTAL CONDITION:

It is well-known that regular physical activity can have a tremendous effect on your health. Experts say that adults should engage in approximately 30-60 minutes of accumulated physical activity per day (where each separate bout is a minimum of ten minutes)1. Another position stand is for adults is defined as a daily energy expenditure of 1.5 kilocalories/kilogram of body weight/day or more; roughly equivalent to brisk walking one half hour every day or more)2.4. Research has shown that regular moderate-vigorous physical activity is associated with the following health benefits: 1.2.3.4.5

Weight control and decreased risk of obesity. Reduces the risk of dying prematurely. Reduces the risk of heart disease. Reduces the risk of developing diabetes. Reduces the risk of developing high blood pressure. Helps reduce blood pressure in people who already have high blood pressure. Reduces the risk of developing colon and breast cancer. Helps build and maintain healthy bones, muscles, and joints. There is a linear relationship between increased activity and health benefits (t PA = t benefits).

Some Statistics!:

Globally, there are more than 1 billion overweight adults, at least 300 million of them obese. An estimated 16.7 million - or 29.2% of total global deaths - result from the various forms of cardiovascular disease (CVD), many of which are preventable by action on the major primary risk factors: unhealthy diet, physical inactivity, and smoking. Inactivity greatly contributes to medical costs - by an estimated $75 billion in the USA in 2000 alone. At least 60% of the global population fails to achieve the minimum recommendation of 30 minutes moderate intensity physical activity daily. Physical inactivity is estimated to cause 2 million deaths worldwide annually. Globally, it is estimated to cause about 10-16% of cases each of breast cancer, colon cancers, and diabetes, and about 22% of ischaemic heart disease. The risk of getting a cardiovascular disease increases by 1.5 times in people who do not follow minimum physical activity recommendations. In Canada, physical inactivity accounts for about 6% of total health care costs.

1. Public Health Agency of Canada. Physical Activity Unit, retrieved October 31, 2006 from: http://www.phac-aspc.gc.ca/pau-uap/paguide/why.html

2. CFLRI. Surveys Statistics Summaries, retrieved October 31, 2006 from: http://dlri.ca/eng/statistics/index.php

3. Centers for Disease Control and Prevention. Physical Activity and Health, retrieved October 31, 2006 from: http://www.cdc.gov/nccdphp/sgr/contents.htm

4. World Health Organization. Physical Activity, retrieved October 31, 2006 from: http://www.who.int/dietphysicalactivity/publications/facts/pa/en/index.html

5. Warburton, D., Nicol, c., & Bredin, S. (2006). Health benefits of physical activity: the evidence. Canadian Medical Association Joumal, 174, 801-809.

C o p yr

ig h t ©

2 0 1 7 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

http://www.phac-aspc.gc.ca/pau-uap/paguide/why.html

http://www.dlri.ca/eng/statistics/index.php

http://www.cdc.gov/nccdphp/sgr/contents.htm

http://www.who.int/dietphysicalactivity/publications/facts/pa/en/index.html

158 HEALTH PROMOTION BEHAVIORS

CHANGING SELF-EFFICACY

Bandura (1997) has shown that self-efficacy over performing a behavior can be enhanced in four distinct ways: through mastery experiences, vicarious experience, verbal persuasion or changing perception of physiological and affective states (see Chapter 2). Mastery experiences (i.e., experience of successfully performing the behavior) give people confidence that they can tackle new tasks because they know they have previously succeeded with similar challenges. This finding would suggest the value of graded tasks which involves identifying manageable tasks and only increasing difficulty as confidence and skill grow. Self-efficacy can also be enhanced through observing others’ success (i.e., vicarious experience), especially when we see others successfully performing the behavior as being like us. For example, Bandura (1997) showed that observing a model judged to have less skill than ourselves fail had little or no impact on self-efficacy, while observing a model judged to have similar skills to us fail resulted in reduced self-efficacy. When direct experience or modeling are not possible, self-efficacy can be enhanced through verbal persuasion. People can be persuaded by arguments demonstrating that others (like them) are successful in meeting challenges similar to their own as well as persuasion highlighting individuals’ own skills, and past success. Finally, changing our own perception of physiological reactions and our interpretations of these reactions can be used to change self-efficacy. Mood, stress and anxiety during performance of a behavior can bolster or undermine self-efficacy. For example, although arousal is normal during demanding performances, it can be interpreted as a sign of panic or incompetence. Such interpretations are likely to disrupt and undermine performance. By contrast, acknowledging arousal as a natural response to performance

FIGURE 7.4 Frequency of moderate/vigorous exercise of at least 30 minutes duration by condition (from Conner et al., 2011b)

6 Control

5 Affect

Cognitive

2 Time 1 Time 2

C o p yr

ig h t ©

2 0 1 7 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

HEALTH PROMOTION BEHAVIORS 159

demands may add to excitement and commitment. Thus interventions designed to reduce negative moods and anxiety and to re-interpret destructive interpretations of arousal can be used to enhance self-efficacy and facilitate skilled performance.

Studies using these techniques either pick on particular techniques or attempt to use all the techniques. For example, Baranowski et al. (2003) targeted vicarious and mastery experiences in relation to children asking for healthy foods inside and outside the home by means of computerized interventions. Using a multimedia game this approach increased mastery in asking for healthy foods at home and when eating out and resulted in increased fruit and vegetable consumption. Based on using all these different means of promoting self-efficacy, Lawrence and colleagues (1997) developed an intervention that included HIV/AIDS education, teaching and rehearsing skills targeting social competence (negotiations with a partner, refusal), mastery of self-protective skills (condom application and increasing sterility of intravenous drug application), technical competence, generating a supportive climate amongst participants and normalizing self-protective behaviors. The intervention targeted a high-risk population of women in prison. After the intervention the women reported higher self-efficacy and frequency of communication with their partners about condom use. They also exhibited more knowledge about HIV/AIDS and improved their ability to apply condoms.

TARGETING INTENTIONS TO CHANGE HEALTH PROMOTION BEHAVIORS

In this section we focus on research that draws on simple techniques to help promote behavior change in those individuals who are generally positively disposed (i.e., motivated) towards performing the behavior but do not seem to get round to doing it. This contrasts with the techniques considered in the previous section that are more useful in getting individuals more motivated to perform a behavior. We consider two such techniques: the question-behavior effect and implementation intentions. A range of other behavior change techniques (motivational interviewing; self-monitoring; incentives), as well as implementation intentions, are overviewed and evaluated in Chapter 3.

Question-behavior effect

Research has indicated that merely asking questions about a behavior may be sufficient to produce changes in that or related behaviors (for reviews see Wilding et al., 2016; Wood et al., 2016). This has come to be known as the question-behavior effect (QBE). Use of the QBE in relation to changing health behavior is illustrated by Godin, Sheeran, Conner and Germain (2008). This study showed that receiving a questionnaire containing questions about intentions to donate blood (along with a range of other questions from the TPB) resulted in increased blood donation at 6 months (donation rates of 54% vs. 49% respectively) and 12 months (70% vs. 65% respectively) compared to a group not receiving a questionnaire. The QBE has subsequently been tested across a range of health behaviors including physical activity (e.g., Sandberg & Conner, 2011), screening attendance (Sandberg & Conner, 2009) and influenza vaccination (Conner, Godin, Norman & Sheeran, 2011a, Study 2).

C o p yr

ig h t ©

2 0 1 7 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

160 HEALTH PROMOTION BEHAVIORS

The most common explanation of the QBE is that asking behavioral intention questions heightens the accessibility of the person’s attitude towards that behavior (attitude accessibility) which in turn increases the likelihood that attitude- consistent behavior will be performed. For example, Morwitz and Fitzsimons (2004) showed that completing purchase intention questions increased the activation level of pre-existing brand attitudes. When the brand attitude was both highly accessible and positively valenced participants were more likely to choose that brand, whereas when the activated attitude was both highly accessible and negatively valenced participants were less likely to choose that brand. Wood, Conner, Sandberg, Godin and Sheeran (2014) showed that changes in the accessibility of attitudes mediated, or explained, the effects of asking intention questions on behavior. An interesting consequence of this suggested mechanism is that the QBE can decrease performance of the behavior among those with negative reactions to the behavior. For example, Conner et al. (2011a) showed that screening attendance and influenza vaccination rates among those with negative attitudes and intentions to these behaviors were actually lower for those who completed a questionnaire about these behaviors compared to those who did not receive a questionnaire. Putting these findings together (i.e., QBE increases behavior when the underlying cognitions are positive and the QBE decreases behavior when the underlying cognitions are negative), Ayres et al. (2013) showed that measuring intentions compared to not measuring intentions only resulted in an increase in the behavior (requesting a personalized health plan) when motivation to protect their health was also high (based on receiving feedback or not on their risk factors).

The findings from these and other QBE studies suggest that a relatively simple and cost effective way to promote various health protection behaviors and health detection behaviors is to get individuals to complete intention questions focused on that behavior. However, such effects are only likely to be effective if the underlying cognitions are positive; this makes sense in that it seems unlikely that by simply asking questions we can make someone do something they do not want to do.

Implementation intentions

Another way in which we can increase the performance of health promotion behaviors is through the use of implementation intentions or simple if-then plans. Prestwich, Sheeran, Webb and Gollwitzer (2015, p. 324) note that:

to form an implementation intention, the person must first identify a response that will lead to goal attainment and, second, anticipate a suitable opportunity to initiate that response. For example, in order to enact the goal intention to exercise, the

Mark uses implementation intentions to try and help him exercise at least once per week. Mark’s implementation intention is that if it is 5pm on a Wednesday he will get ready to do some exercise. Usually this involves putting his sports gear in a bag to play squash, going to the gym or the climbing wall, although if he is at home it might mean putting his gear on to go out on his mountain bike. Gollwitzer (1993) argues that by forming implementation intentions individuals pass control of intention enactment to

C o p yr

ig h t ©

2 0 1 7 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

HEALTH PROMOTION BEHAVIORS 161

the environment. The specified environmental cue prompts the action so that the person does not have to remember the goal intention or decide when to act.

An increasing number of studies have shown the power of implementation intentions to promote both health detection and health promotion behaviors. In relation to the former, Orbell, Hodgkins and Sheeran (1997) was one of the first studies to demonstrate an effect of implementation intentions for a health detection behavior. This study looked at breast self-examination in young women to ensure early detection of abnormalities that might indicate early signs of breast cancer. The young women in the study were randomly allocated to one of two conditions, one where no implementation intention was formed and one where an implementation intention was formed (e.g., ‘I will perform breast self-examination when I take my bath on a Friday night’). At follow-up, self- reported rates of breast self-examination were dramatically different: 16% in the control condition and 64% in the implementation intention condition. Studies have shown similar effects in relation to health protection behaviors. For example, Prestwich, Lawton and Conner (2003) showed implementation intentions could be used to promote exercise and increased both self-reported exercise and objectively assessed fitness. Importantly this study showed that forming implementation intentions was most effective when combined with a motivational manipulation. This makes sense as we might expect an implementation intention to be mainly useful in changing behavior among those who want to change this behavior. Implementation intentions have been shown to increase the performance of a range of health behaviors with, on average, a medium effect size (see Gollwitzer and Sheeran, 2006, for a meta-analysis; for a more detailed evaluation of their impact on behavior, see Chapter 2).

Prestwich et al. (2015) provide an in-depth review of both basic and applied research with implementation intentions along with a taxonomy of implementation intentions to change behavior (see Figure 7.5). It is interesting to note that the application of implementation intentions to changing health protection behaviors usually requires the individual to identify appropriate opportunities to perform the behavior. In contrast, the application of implementation intentions to changing health risk behaviors usually requires the individual to identify appropriate alternative ways to act or other strategies when faced with the temptation to perform the health risk behavior (see Figure 7.5).

An interesting recent development has been the idea of collaborative implementation intentions. This is where a pair of individuals form an implementation intention to perform the behavior together. Prestwich et al. (2012) showed such collaborative implementation intentions to be more effective than individual implementation intentions, partner support or a control condition in promoting physical activity over 1-, 3- and 6-month periods. Studies have also shown such collaborative implementation intentions to be effective for promoting breast self-examination (Prestwich et al., 2005), though effects on healthy eating are less clear cut (Prestwich et al., 2014).

C o p yr

ig h t ©

2 0 1 7 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

FIGURE 7.5 A framework for operationalizing implementation intentions in relation to particular volitional problems (redrawn from Prestwich et al., 2015). Note panel B applies to using implementation intentions to increase health promotion behaviors (Chapter 7) and panel C applies to using implementation intentions to reduce health risk behaviors (Chapter 8)

PANELA Goal or Behavior

Identify goal-directed response

Obtain wanted response Control unwanted response

PANEL B

Problems of initiating response

if-then plan to instigate action

PANEL C

Problems of overcoming habitual responses

Obtain wanted response

Problems of maintaining response

Effort Levelof performance

Problems of getting derailed by contextual threats

(alternatively: for direct control of contextual threat,

see Panel C)

if-then plan to mobilize effort

if-then plan relating to performance target or orientation

Control unwanted response

Problems of overcoming contextual threats

if-then plan to if-then plan to if-then plan to substitute if-then plan to ignore the moderate the entirely suppress the an antagonistic wanted triggering stimuli for

unwanted response unwanted response response (see Panel B) unwanted responses

C o p yr

ig h t ©

2 0 1 7 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

HEALTH PROMOTION BEHAVIORS 163

BURNING ISSUE BOX 7.3

QUESTIONS AND ANSWERS ON CHANGING HEALTH PROMOTION BEHAVIORS

Question 1: Are particular health promotion behaviors easier to predict than others?

The most common way to measure how easy it is to predict behaviors is in terms of how much variance in behavior key predictors account for. So, for example, McEachan et al. (2011) reviewed the use of the Theory of Planned Behavior (TPB) to predict various health behaviors. They found that the TPB explained the most variance in physical activity and the least variance in safer sex behaviors. That would suggest that physical activity is easier to predict than safer sex! However, we must remember that studies will vary in other ways than just the behavior they measure. For example, many physical activity studies now report objective measures of behavior while safer sex studies are usually reliant on self-reported behavior.

Question 2: Could particular theories work better for certain health behaviors?

The social/health cognition models we have examined were developed to work for a broad range of health behaviors rather than being specific to particular behaviors. As such it could be that particular theories work better for some types of health behavior and other theories work better for other types of health behaviors. However, directly comparing theories in this way can be difficult because different numbers and types of predictors are involved. It is easier to compare particular components of theories across different behaviors. For example, McEachan et al. (2011) showed that intentions were stronger predictors of physical activity than abstinence behaviors like reducing drinking.

Question 3: How can intentions be made more stable?

To increase the performance of health promotion behaviors getting individuals to have positive and stable intentions to perform them may lead to long- term performance. For example, Conner, Norman and Bell (2002) showed that stable intentions to eat healthily were associated with healthy eating six years later. However, to date we know relatively little about what promotes stable intentions. Conner et al. (2016) suggest that if we can get individuals to prioritize their intentions to perform health promotion behaviors this may lead to more stable intentions. They also showed that simple messages suggesting the importance of such prioritization could be effective. Nevertheless future studies identifying effective interventions to promote stable intentions would be a useful development in this area.

C o p yr

ig h t ©

2 0 1 7 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

164 HEALTH PROMOTION BEHAVIORS

SCIENCE OF HEALTH BEHAVIOR CHANGE: IN ACTION In the first section of the book, you learned about how to take a scientific approach to health behavior change (broadly overviewed in Chapter 1). In particular, you learned about different theories (Chapter 2) and behavior change techniques (Chapter 3) and how to apply them to develop theory-based interventions (Chapter 4). In addition, you learned about how to design studies adopting different methodological approaches (Chapter 5) and to statistically test whether a particular intervention was effective in changing behavior (Chapter 6). Along the way, you encountered examples of different critical thinking skills to help you to evaluate the quality of health behavior change studies. Below, we present the first of our Science of Health Behavior Change: In Action boxes. These boxes take a published individual study that attempted to change a health behavior and critically evaluate it against what we have learnt. Chapters 8, 9 and 10 similarly provide In Action boxes in relation to the types of health behaviors considered in those chapters.

In Action Box 7.1 considers the study of Milne, Orbell and Sheeran (2002) who tested the impact of implementation intentions on exercise participation. In particular, they aimed to assess the effects of an implementation intention plus a motivational intervention compared to a motivational intervention alone or a control condition. Theory would suggest that implementation intentions are particularly effective when individuals are motivated to perform the behavior. Therefore it was expected that the combined implementation intention plus motivational intervention would be particularly effective in increasing exercise participation.

Question 4: What’s more important in predicting behavior: an individual’s personality or their cognitions?

Personality traits like conscientiousness show only small- to medium-sized effects on health behaviors while cognitions like intentions show medium- to large-sized effects. So cognitions are the more important predictors (see Conner & Abraham, 2001).

Question 5: Given the effects of the QBE on behavior change tend to be small, is it an important behavior change technique?

In selecting a behavior change technique to apply the associated effect size is a key consideration. However, cost of implementing the intervention is often another consideration. While the QBE is associated with a small effect size it would be a cheap intervention to implement. It may well therefore be a useful intervention in certain circumstances. For example, various screening programs that require individuals to be sent invitations could very cheaply add a questionnaire about the screening behavior to produce a valuable increase in participation rates.

C o p yr

ig h t ©

2 0 1 7 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

HEALTH PROMOTION BEHAVIORS 165

SCIENCE OF HEALTH BEHAVIOR CHANGE

In Action Box 7.1

Study: Milne et al. (2002). Combining motivational and volitional interventions to promote exercise participation: Protection motivation theory and implementation intentions

Aim: To increase exercise in undergraduate students.

Method: Participants were randomized to one of three conditions: a combined motivational and volitional intervention (based on Protection Motivation Theory (PMT) and implementation intentions), a motivational intervention only (the same PMT-based intervention) or a control.

Results: The PMT-manipulation changed PMT constructs related to threat and appraisal but it did not influence exercise behavior. Those in the combined group increased their exercise significantly more than those in the motivational intervention group or control group.

Authors’ conclusion: Both motivation and volition are needed for goal attainment.

THEORY AND TECHNIQUES OF HEALTH BEHAVIOR CHANGE

BCTs INTERVENTION: Based on Michie et al.’s (2013) taxonomy:

Combined intervention: 1.1 Goal setting (behavior); 1.4 Action planning; 5.1 Information about health consequences; 5.2 Salience of consequences; 15.1 Persuasion of ability; 15.2 Mental rehearsal of successful performance.

Motivational intervention only: 5.1 Information about health consequences; 5.2 Salience of consequences; 15.1 Persuasion of ability; 15.2 Mental rehearsal of successful performance.

CONTROL: None

Critical Skills Toolkit

3.1 Does the design enable the identification of which BCTs are effective?

No – adding an implementation intention-only condition would ensure a 2 (implementation intention: yes/no) x 2 (PMT: yes/no) full-factorial design. Without this condition, it is unclear whether both implementation intentions and the motivational intervention are needed or whether implementation intentions only are sufficient for behavior change.

4.1 Is the intervention based on one theory or a combination of theories (if any at all)?

Combined intervention: Protection Motivation Theory (PMT) with implementation intentions (a BCT rather than a theory) added to it. While adding implementation intentions to PMT risks undermining the established theory (by modifying PMT), the modification is in keeping with the Model of Action Phases which specifies behavior is determined through a combination of motivational and volitional elements.

PMT-only intervention: based only on PMT.

C o p yr

ig h t ©

2 0 1 7 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

166 HEALTH PROMOTION BEHAVIORS

Are all constructs specified within a theory targeted by the intervention?

The main determinants of protection motivation: threat appraisal (via perceived severity and perceived vulnerability) and coping appraisal (via self-efficacy, response efficacy and response costs) were targeted by specific BCTs in the PMT conditions. The only element of the theory that was not explicitly targeted by the PMT-interventions was rewards (intrinsic and extrinsic) that increase the likelihood of a maladaptive response.

Are all behavior change techniques explicitly targeting at least one theory-relevant construct?

Yes – 1.1 Goal setting (behavior) and 1.4 Action planning targeted implementation intentions; 5.1 Information about health consequences targeted perceived severity, perceived vulnerability and response efficacy; 5.2 Salience of consequences targeted perceived vulnerability; 15.1 Persuasion of ability and 15.2 Mental rehearsal of successful performance targeted self-efficacy.

Do the authors test why the intervention was effective or ineffective (consistent with the underlying theory)?

No – mediation analyses were not conducted. Although the PMT variables changed as a result of the PMT manipulation, it is not clear whether changes in the PMT variables mediated the effect of the PMT-based messages on intentions.

4.2 Does the study tailor the intervention based on the underlying theory?

No – all participants within a study condition received the same materials. However, given the brevity and likely low cost of the intervention, the lack of tailoring is unlikely to represent a major issue.

2.1, 2.2 What are the strengths/ limitations of the underlying theory?

See Burning Issue Box 2.3 for strengths and limitations of Protection Motivation Theory.

THE METHODOLOGY OF HEALTH BEHAVIOR CHANGE

5.1, 5.3, 5.4, 5.6

Methodological approach

The study adopted an experimental design (see Critical Skills Toolkit 5.4).

6.1 For experimental designs: is the study between- subjects, within- subjects or mixed?

Mixed design because participants were allocated to one of three conditions (between-subjects) and completed measures at multiple time-points (within-subjects). For advantages and disadvantages of this design, see Critical Skills Toolkit 6.1.

5.2, 5.5 Are the measures reliable and valid?

The main outcome variable – exercise – was measured by a single item with no known reliability or validity. The measures of PMT variables were, in some cases, not internally reliable (though these were appropriately analyzed as single items) and these measures had not been previously validated.

C o p yr

ig h t ©

2 0 1 7 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .

HEALTH PROMOTION BEHAVIORS 167

5.5 May other variables have been manipulated other than the independent variable?

Low risk of bias. The manipulations appeared appropriate. Thus, the internal validity of the experiment was not threatened by the risk of confounds.

Non-random allocation of participants to condition

Unclear risk. Participants were randomized to condition but it is not clear how this randomization took place (hence, an inappropriate method of randomization could have been used). Although there were no reported differences across the groups at baseline (suggesting randomization was successful), differences between the groups (arising from potentially inappropriate randomization methods) could be present on unmeasured variables.

Blinding and allocation concealment

Unclear risk. The authors note that the ‘participants were anonymous to the experimenter’ but this does not necessarily mean that the experimenter was blinded to condition. There was no other evidence of blinding or allocation concealment and this may be more problematic given exercise was measured by self-report rather than objectively.

There was also risk of demand effects given the text used in the implementation intention condition may have led participants to believe that their behavior should change: ‘It has been found that if you form a definite plan of exactly when and where you will carry out an intended behavior you are more likely to actually do so’ (p. 170).

ANALYZING HEALTH BEHAVIOR CHANGE DATA

6.2 Was the sample size calculated a-priori?

The sample size was not calculated a-priori. However, given the effects of the combined intervention was large, this is unlikely to be a major issue.

Fig. 6.1 Was the hypothesis tested with an appropriate statistical test?

Yes.

5.5, 6.4

Incomplete outcome data

It is unclear whether attrition rates differed across the three study conditions. It is a possibility, therefore, that attrition may have been higher in the experimental groups which would suggest an issue with acceptability. Moreover, the analyses were not conducted on an intention-to-treat basis. However, the attrition rates were reasonably low (of the 273 participants who completed the questionnaire at baseline, 250 participants completed the questionnaires at both follow-ups). Moreover, given the effects of the combined intervention were large, it is unlikely that the results would suggest different conclusions if they were analyzed on an intention-to-treat basis. The participants who dropped out were similar on the measured variables compared to those who completed the study, suggesting that the results are generalizable to the types of participants recruited (i.e., other undergraduate students).

C o p yr

ig h t ©

2 0 1 7 . T

a yl

o r

& F

ra n ci

s G

ro u p . A

ll ri g h ts

r e se

rv e d .