Health Psychology Textbook PDF

7 Changing Health Behaviours Learning Objectives To understand: 1. The Need to Change Behaviour 2. Lea...

7 Changing Health Behaviours Learning Objectives To understand: 1. The Need to Change Behaviour 2. Learning and Cognitive Theory 3. Social Cognition Theory 4. Stage Models 5. The Role of Affect 6. Integrated Approaches 7. Thinking Critically About Changing Health Behaviours © Shutterstock / Creative Cat Studio 170 PART TWO Staying Well: Health Beliefs, Behaviour and Behaviour Change CHAPTER OVERVIEW Chapter 2 explored factors that help us to understand and predict health-related behaviours. Chapters 3–6 then focused on individual health behaviours. This chapter explores a number of different approaches that have been developed to change health-related behaviours. Although there are a multitude of strategies that can be used to change individual behaviour, this chapter describes those that have been informed by four main theoretical perspectives: (1) learning and cognitive theory leading to behavioural strategies, cognitive behavioural therapy and relapse prevention; (2) social cognition theory and the use of social cognition models to frame interventions, planning, implementation intentions and information; (3) stages of change theory (SOC) and the development of stage matched interventions and motivational interviewing; (4) theories of affect and the use of fear appeals, visualization and self-affirmation. The chapter finally explores how these different approaches have been integrated through the creation of a science of behaviour change interventions, the use of modern technologies, the mass media and a focus on sustained behaviour change. CASE STUDY Sapphira is a dietician and works with patients with diabetes. Many of her patients are overweight and some take medication. Her role is to encourage them to eat a healthy diet, do more exercise, stop smoking and take their medication. This is not an easy task as many of their unhealthy habits have been entrenched for many years. In addition, she only sees them for 50 minutes every month. She does, however, always see the same patients and over time manages to build a relationship with them. During these consultations she finds that trying to understand their behaviour from their own per- spective, listening rather than just giving advice, setting doable short-term goals which involve small changes that fit into their lives, rewarding them whenever they try to make a difference and keeping positive, work the best. She has also learned from experience that being disappointed or irritated when patients don’t follow her advice just means that they don’t come back. After many years in her job, Sapphira believes that the most important part of her role is to build a good relationship with her clients so that they keep coming back. Through the Eyes of Health Psychology... Changing behaviour is difficult as many behaviours are habitual and have become embedded over many years. Sapphira’s story illustrates the difficult task faced by those trying to help others change their behaviour, which is far more complex than simply telling patients what to do. This chapter will describe the key theoretical approaches to behaviour change that highlight some of the behaviour change strategies illustrated above, such as reinforcement (rewarding patients and keeping positive); planning (setting doable goals); and tailored approaches (small changes that fit into their lives). It will then describe how these approaches have been integrated for use by different professionals and across the different health behaviours. 1 THE NEED TO CHANGE BEHAVIOUR Before the twentieth century, the most common causes of death were childbirth or acute illnesses, such as tuberculosis or flu, caused by bacteria or viruses. Over the past 100 years this has changed and now most people in the developed world die as a result of chronic conditions such as heart disease, cancer, diabetes or obesity. Furthermore, the majority of those seeking medical help may not only have these more common chronic conditions but others including chronic back and joint pain, asthma,  CHAPTER 7 Changing Health Behaviours 171 multiple sclerosis (MS), fibromyalgia, inflammatory bowel disorder (IBD), HIV/AIDS, high blood pressure, fatigue and headaches. These chronic conditions illustrate a key role for behaviour and why changing health behaviour is central to health care management: To prevent illness: Behaviour change is key to preventing chronic conditions. For example, stopping smoking can prevent lung cancer, eating a healthier diet can prevent bowel cancer and doing more exercise can prevent heart disease. This is sometimes called primary prevention. To manage illness: Once diagnosed with a chronic condition, behaviour change is also key to illness management. For example, dietary change plays a core role in the management of obesity and diabetes, increased exercise is central to the management of patients post heart attack and encouraging medication adherence is important for many illnesses such as asthma and those with high blood pressure. This is sometimes known as secondary prevention. To reduce physical symptoms: Chronic conditions can result in a multitude of physical symptoms such as fatigue, pain, nausea and bowel problems. Behaviour change can help to reduce these symp- toms. For example, dietary change may reduce nausea or bowel problems and exercise may reduce fatigue and pain. To improve well-being: Having cancer, being diagnosed with heart disease or living with MS, fibro- myalgia, asthma or IBD can be miserable and reduce a person’s sense of well-being and quality of life. Behaviour change can also help to improve well-being. In particular, being more active has positive effects on mood (see Chapter 5) and can lead to more social support if people join group activities. Further, eating a healthier diet may or may not improve physical health outcomes for cancer but it can help people feel more in control of their lives and that they can make a difference in a small way. Behaviour change is therefore central to many aspects of health. Research has explored ways to help people change their behaviour. Behaviour change can be seen as being either conscious and effortful, whereby the individual makes choices about how and whether to change their behaviour (e.g. joining a gym) or effortless, whereby behaviour change occurs with no conscious processing (e.g. buying wholewheat bread as that is all that is on offer). Most psychological approaches encour- age effortful changes in behaviour through interventions targeted at the individual. In contrast, public health inter- ventions focus more on structural and environmental changes that bring about shifts in behaviour without the individual necessarily knowing they are involved in an intervention or even that they have changed their behav- iour. Community events can encourage behaviour change by providing the right environment for a behaviour and then encouraging an individual the make the choice to take part. A good example of this is parkrun which takes place every Saturday morning at 9 o’clock. It happens in many different countries across the world and people of all abili- ties run a timed 5 km together. Similarly, local fun runs can get people running who have never run before. This chapter will explore both effortful and effortless behaviour change strategies in the context of both indi- vidual and public health approaches to behaviour change. It will also explore how a range of strategies have evolved from a number of key psychological theories (see Chapter 3 Community events can help to change for details on smoking- and alcohol-specific interventions). behaviour 172 PART TWO Staying Well: Health Beliefs, Behaviour and Behaviour Change Most psychology-based interventions fall within the framework of four main theoretical perspectives: learning and cognitive theory; social cognition theory; stage models; the role of affect. These will now be explored together with attempts to integrate these approaches and generate integrated models of behaviour change. 2 LEARNING AND COGNITIVE THEORY Learning theory forms the basis of much psychological work with its emphasis on associative learn- ing, reinforcement and modelling. From this perspective we eat chocolate when we are feeling fed up because we associate chocolate with feeling special from when we were children (associative learn- ing), because our parents commented how lucky we were when they gave it to us (reinforcement) and because we saw them eat it (modelling). Cognitive theory then adds to this approach by exploring how people think as well as how they behave. These theoretical perspectives have informed a number of therapeutic approaches which have been used to promote behaviour change. This chapter first describes those approaches informed by learning theory (reinforcement, incentives, modelling, associative learning, exposure). It then describes those informed by both learning and cognitive theories (cognitive behavioural therapy (CBT) and relapse prevention). These different approaches are shown in Figure 7.1. Cognitive Reinforcement Modelling behaviour therapy (CBT) Learning theory ADD Cognitions Associative Exposure Relapse learning preventions Incentives Figure 7.1 How learning theory and cognitive theory inform behaviour change LEARNING THEORY APPROACHES Reinforcement One way to change behaviour is to positively reinforce the desired behaviour and ignore or punish the less desired behaviour. For example, a child is more likely to eat fruit and vegetables if their parent smiles while they are eating them and an adult is less likely to return for a screening test if they found the last one embarrassing and painful. This process has been assessed by a number of different experi- mental studies and interventions. For example, Barthomeuf et al. (2007) explored whether the emotion expressed on people’s faces could influence food preferences. Men and women were exposed to a series of pictures of liked and disliked foods that were either on their own or accompanied by people eating them and expressing one of three emotions: disgust, pleasure or neutrality. The results showed that the expression of the eater influenced ratings of preference. Therefore, pairing a food with emo- tion changes the preference for that food. Similarly, Gwozdz et al. (2020) conducted a field experiment in ten primary schools in five European countries and concluded that smiley stamps promoted fruit and  CHAPTER 7 Changing Health Behaviours 173 vegetable eating among children in some but not all countries. Likewise, Harne-Britner et al. (2011) conducted a trial to improve healthcare workers’ hand hygiene using positive reinforcement with a sticker reward system. The results showed a 15.5 per cent increase in hand hygiene compliance during the first month. They also found, however, that this effect was not sustained in the longer term. Further, in 2019, Fazzino et al.’s systematic review of 10 studies concluded that reinforcement-based interventions were effective in promoting positive outcomes in substance use. The role of reinforce- ment is also implicit with several medical approaches to managing behaviour. For example, the drug Antabuse induces sickness if the individual also consumes alcohol and the drug orlistat causes anal leakage if taken with fatty foods. Both these consequences act as a deterrent for future behaviours (Ogden and Sidhu 2006; Hollywood and Ogden 2016). Furthermore, research indicates that people often change their behaviour in the longer term when the old unhealthy behaviour is no longer functional. For example, people stop smoking when it no longer offers them a way to spend time with friends or change their diet when they find different foods more enjoyable (Ogden and Hills 2008). Incentives Research has also explored the impact of financial incentives as a means to change behaviour. Incentivizing behaviour is rooted in a notion of reinforcement and has long been the standard tool of the retail industry which aims to encourage consumers to purchase a particular brand rather than a competitor’s brand. It is also linked to the notion of a token economy, In the context of health-related behaviours, incentivizing can take the form of centralized changes to the cost of products such as cigarettes, fatty foods and fizzy drinks or directly paying people to lose weight, stop smoking or be more physically active. Over the past few years research has addressed the effectiveness of these simple (and fairly crude) approaches and indicates that, in general, changes in cost and direct financial rewards can effectively change behaviour. For example, increased taxes on both alcohol and cigarettes over the past few decades have been linked with a reduction in drinking and alcohol (Sutherland et al. 2008) and current ongoing direct payment schemes include giving pregnant women in the UK £20 food vouchers for one-week smoking cessation, £40 after four weeks and £40 after one year (North East Essex NHS Trust 2009); paying men and women $45 in Tanzania to have regular tests for sexually trans- mitted diseases (World Bank 2008) and giving points to children in Scotland for eating healthy school meals which can be exchanged for donations to a Save the Children project abroad (East Ayrshire Council 2011). Marteau et al. (2009) reviewed the evidence for incentivizing behaviour change and concluded the following: The greater the incentive, the greater the likelihood of behaviour change. Incentives are better at producing short-term rather than longer-term changes. The impact of the incentive depends upon the financial state of the individual. Incentives are more effective if the money is paid as close as possible to that target behaviour. Incentives work better for discrete and infrequent behaviours such as having vaccinations rather than repeated habitual behaviours such as diet or smoking. Marteau et al. (2009) also concluded that there may be three unintended consequences of incentiv- izing behaviour. These are: Incentives may undermine an individual’s intrinsic motivation for carrying out a behaviour (e.g. ‘I ate healthily but now I don’t really like healthy foods’). Incentives are a form of bribery which undermine an individual’s informed consent and autonomy. Incentives may change the doctor–patient relationship if the patient is paid by the doctor to behave in certain ways. Incentives therefore seem to change behaviour through a crude version of reinforcement. However, they may also have unintended consequences which may undermine changes in behaviour in the longer term. 174 PART TWO Staying Well: Health Beliefs, Behaviour and Behaviour Change Modelling Modelling healthy behaviour can also change behaviour: a child is more likely to smoke if their parents smoke and less likely to take up exercise if they see their parents sitting on the sofa watching TV (see Chapters 3 and 4). Research shows that adolescents are more likely to eat breakfast if their parents do (Pearson et al. 2009) and that adolescents are more likely to eat for emotional reasons such as bore- dom or comfort if their parents do this as well (Snoek et al. 2007). An intervention study explicitly used modelling to change children’s eating behaviour (Lowe et al. 1998). This series of studies used video material of ‘food dudes’ – older children enthusiastically consuming refused food – which was shown to children with a history of food refusal. The results showed that exposure to the ‘food dudes’ significantly changed the children’s food preferences and specifically increased their consumption of fruit and veg- etables as the participants modelled their behaviour on that of the ‘food dudes’ in the video. The Food Dudes healthy eating programme has now been adopted in parts of England and the Republic of Ireland and is based upon the three ‘Rs’: role modelling, repeated tasting, and rewards. Marcano-Olivier et al. (2021) evaluated the intervention and reported an increase in children’s intake of fruit, vegetables, vita- min C and E and a decrease in their total energy consumption of fat, saturated fat and sodium intake in the intervention school but not in the control school. Similarly, Sanderson and Yopyk (2007) showed young people a video consisting of other young people with positive attitudes towards condoms who modelled strategies for using them appropriately. The results showed increased intentions to have safe sex, higher self-efficacy to refuse unsafe sex and higher condom use four months later. Further, using a qualitative method, Deliens et al. (2015) reported that university students identified the behaviour of their social networks as the key determinants of their own physical activity and sedentary behaviours. Associative Learning Associative learning involves pairing two variables together so that one variable acquires the value or meaning of the other. For example, in the classic early studies Pavlov’s dogs heard a bell ring whenever they were given food and after a while they started to salivate when they heard the bell (even without the food). Similarly, Van den Akker et al. (2017) paired time and chocolate and showed that the repeated consumption of chocolate at a specific time of day over 15 days increased the desire to eat chocolate at this time. One form of associative learning is evaluative conditioning whereby an attitude object is paired repeatedly with an object which is either viewed positively or nega- tively as a means to make the attitude object either more positive or negative. This method is frequently used in marketing as a means to make relatively neutral objects (e.g. perfume, cigarettes, pet food, air freshener) seem more positive by pairing them with something that is inherently attractive (e.g. attrac- tive people, green fields, romantic music, etc.). Gibson (2008) tested this process experimentally and reported that evaluative conditioning could make participants predictably choose between Coca-Cola or Pepsi depending on which one had been paired with positive meaning. In terms of health, Hollands et al. (2011) used an evaluative conditioning procedure to increase the negative value attached to unhealthy snacks such as crisps and chocolate. Participants were shown images of unhealthy snacks interspersed with aversive unhealthy images of the body for the experimental condition (e.g. artery disease, obesity, heart surgery), or a blank screen for the control condition. The results showed that the intervention resulted in more positive implicit attitudes compared to the control condition. In addition, those in the experimental condition also chose fruit rather than high calorie snacks in a behavioural task. But whether these changes are sustained in the longer term remains unknown. Exposure One of the best predictors of future behaviour is past behaviour (see Chapter 2), as having already per- formed a behaviour makes that behaviour seem familiar and can increase an individual’s confidence that they can carry out the behaviour again. Therefore, one of the simplest ways to change behaviour  CHAPTER 7 Changing Health Behaviours 175 is through exposure to the behaviour, practice or skills training. In terms of eating habits, research shows that we eat what we are familiar with and have been exposed to. For example, Wardle et al. (2003) carried out a study whereby children aged 2–6 identified a vegetable they least liked and then were exposed to this vegetable for 14 days (compared to children who were either given information or were in the control group). The results showed that daily exposure resulted in the children eating more of the vegetable in a taste test and reporting greater prefer- ence for the vegetable than those in the other two groups. Similarly, research indicates that children can identify and are willing to taste vegetables if their parents purchase them (Busick et al. 2008). Simple exposure can therefore change intake and preference. In a similar vein, actually performing a behaviour once can increase the chances that this behaviour will occur again in the future. For example, as part of sexual education interventions, research shows that basic skills training in negotiating how to ask for a Skills training for condom use can condom, putting a condom on or buying a condom, which promote health behaviours through exposure to those behaviours in a safe involves rehearsing these behaviours in a safe environ- context (This is a banana!) ment, improves the likelihood of these behaviours in the SOURCE: © Shutterstock / pics five future (e.g. Weisse et al. 1995; see Chapter 6). Furthermore, not only does past behaviour predict future behaviour but it also predicts and changes cognitions that then predict behaviour (Gerrard et al. 1996). For example, if I think ‘condoms are difficult to put on’ and my behaviour is ‘I don’t use condoms’ and then I put one on a banana during a skills training session, my cognition will shift to ‘actually I can use condoms’ and my behaviour will change as well to ‘I now use condoms’. ADDING COGNITIVE THEORY Reinforcement, incentives, modelling, associative learning and exposure are behaviour change approaches derived from learning theory. Adding cognitive theory leads to cognitive behaviour therapy (CBT) and relapse prevention. Cognitive Behavioural Therapy (CBT) Interventions to change behaviour tend to combine cognitive and behaviour strategies into cognitive behavioural therapy (CBT). Freeman (1995) describes how CBT emphasizes the following: The link between thoughts and feelings. Therapy as a collaboration between patient and therapist. The patient as scientist and the role of experimentation. The importance of self-monitoring. The importance of regular measurement. The idea of an agenda for each session set by both patient and therapist. The idea that treatment is about learning a set of skills. The idea that the therapist is not the expert who will teach the patient how to get better. The importance of regular feedback by both patient and therapist. 176 PART TWO Staying Well: Health Beliefs, Behaviour and Behaviour Change CBT can vary according to client group and the problem being addressed, but involves a more structured form of intervention than most therapies and often includes the following cognitive and behavioural strategies: 1 Keeping a diary: many behaviours and thoughts occur without people being fully aware of them. For CBT, clients are asked to keep a diary of significant events and associated feelings, thoughts and behaviours. This process of self-monitoring enables clients to understand the patterns in their lives and the ways in which they are responding to whatever is happening to them. For someone trying to change their diet, a diary could reveal that they eat while watching the TV or turn to food at work when feeling under stress. 2 Gradually trying out new behaviours: many behaviours are habitual and over time we learn to practise those behaviours which make us feel good and avoid those that make us feel uncomfort- able. For CBT, clients are asked either on their own or with the therapist to try out new behaviours or face activities that have been avoided. This enables people to build confidence and familiarity with new behaviours and try to unlearn old behaviours. 3 Cue exposure: many people find that unhealthy behaviours can be triggered by certain situations (e.g. the desire to smoke when drinking alcohol). For CBT, clients are sometimes exposed to such situations when with the therapist in order to help them learn new coping responses and extinguish the old unhealthy reactions to these situations. For example, people addicted to drugs may be gradu- ally exposed to the paraphernalia of drugs (e.g. silver foil, needles, cigarette papers, etc.) as a means to change their response to them. 4 Relaxation techniques: clients may use music, repeated clenching and relaxing of muscles, record- ings of soothing voices or recordings of subliminal messages as a means to aid relaxation. This can help them to reduce their anxiety and negative thoughts about aspects of their lives. 5 Distraction techniques: distraction can be a powerful method for managing anxiety or preventing unhealthy responses to certain situations. In CBT, clients can be helped to find distraction strategies that work for them. For example, if a person feels the need to smoke when with certain friends, they can be taught how to focus on other aspects of their lives at these times or encouraged to use a telephone help line. 6 Cognitive restructuring: central to CBT is the notion that behaviour is maintained through a series of distorted cognitions and a vicious cycle between thoughts and behaviours which is perpetuated by irrational self-talk. Such distorted cognitions are: Selective abstraction, which involves focusing on selected evidence (e.g. ‘drinking alcohol is the only way I can unwind after work’). Dichotomous reasoning, which involves thinking in terms of extremes (e.g. ‘If I am not in complete control, I will lose all control’). Overgeneralization, which involves making conclusions from single events and then generaliz- ing to all others (e.g. ‘I failed last night so I will fail today as well’). Magnification, which involves exaggeration (e.g. ‘Stopping smoking will push me over the brink’). Superstitious thinking, which involves making connections between unconnected things (e.g. ‘If I do exercise, I will have another heart attack’). Personalization, which involves making sense of events in a self-centred fashion (e.g. ‘They were laughing, they must be laughing at me’). CBT then uses a number of cognitive strategies to challenge and change these distorted cognitions and replace them with more helpful ones. The main approach involves Socratic questions, with the therapist challenging the client’s cognitions by asking for evidence and attempting to help the client to develop a different perspective. Questions could include: ‘What evidence do you have to support your thoughts?’; ‘How would someone else view this situation?’; ‘When you say “everyone”, who do you mean?’;  CHAPTER 7 Changing Health Behaviours 177 ‘When you say “all the time”, can you think of times when this is not the case?’ To aid this process the therapist can use role play and role reversal so that the client can watch and hear someone else using their cognitions and learn to see how unhelpful and irrational they are. There is much evidence to support the use of CBT for behaviour change. For example, Rüther et al. (2018) used CBT to promote smoking cessation using a RCT with 155 smokers aged 18–70. They found that those in the intervention group reduced smoking significantly more compared to controls, but this effect was not maintained over time. Likewise, Çelik and Sevi’s (2020) systematic review of 20 stud- ies concluded that CBT-based treatments promoted smoking cessation, especially when combined with medication and nicotine replacement therapy. Further, Vinci’s (2020) review concluded that CBT was effective for smoking cessation for special populations (e.g., low SES; pregnant smokers) when delivered via mhealth/ehealth. Similarly, Riper et al. (2014) conducted a meta-analysis of 12 studies com- prising 1,721 patients investigating the effectiveness of combined CBT and motivational interviewing (MI) for alcohol use disorder. They found that the combination of CBT and MI has a small but clinically significant impact on decreased alcohol use and depression symptoms compared with treatment as usual. This is also supported by Hadjistavropoulos et al.’s (2020) systematic review of 14 studies which compared internet-delivered CBT (ICBT) and therapist-guided therapy (TCBT) for alcohol misuse. The results indicated that while small effects were found for ICBT, small to large effects were found to TCBT. CBT and Chronic Illness Antoni and colleagues (Antoni et al. 2001, 2002) have developed structured guidelines for using CBT with patients with a range of chronic illnesses including those with HIV and cancer as a means to change cognitions and promote behaviour change. For example, beliefs that HIV or cancer are terminal illnesses and that nothing can be done will change a person’s help seeking, and feelings of hopelessness and helplessness will change their mood and quality of life. They outline a detailed system for changing irrational thoughts using rational thought replacement, which they call the ABCDE system. This is as follows: Awareness: because much of our self-talk is automatic, the first step is to become aware of the cognitions we hold and the ways in which these impact upon emotional and physical responses. This awareness process can involve diary-keeping, reflection and talking to a therapist. Beliefs: clients are then asked to rate their beliefs about each of the self-talk processes they hold to identify how strong their cognitions are. They should ask themselves ‘How much do I believe that each of these cognitions is true?’. Challenge: clients challenge their thoughts through questions which ask for evidence or encourage the client to think through what other people would think or do in the same situation. Delete: Antoni and colleagues then argue that clients need to delete these self-statements and replace them with constructive cognitions. This can involve thinking through alternative explanations and different ways of making sense of what happens to them. Evaluate: the final stage is for the client to evaluate how they feel after the cognitions have been deleted and whether they feel the process has been successful. CBT has been most commonly used within the mental health domain to treat problems such as panic disorders, obsessive compulsive disorder (OCD) and eating disorders. It is also used in health psychol- ogy, particularly for a number of chronic conditions, to change behaviours such as physical activity, diet, safe sex practices, smoking and alcohol intake (Antoni et al. 2001, 2002, 2006). For example, Vanderlinden et al. (2012) evaluated the effectiveness of CBT delivered for 1 day a week for an aver- age of 7 months to patients with obesity and binge eating disorder. The results showed improvement in eating behaviours, weight and psychological parameters that lasted up to 3 and a half years post treat- ment. Similarly, Bennebroek Evertsz et al. (2017) investigated the effectiveness of CBT for patients 178 PART TWO Staying Well: Health Beliefs, Behaviour and Behaviour Change with inflammatory bowel disease and showed improvements in quality of life, anxiety and depression. Further, Aricó et al. (2016) conducted a review of 16 studies on the effectiveness of CBT for insom- nia in breast cancer survivors and concluded that CBT was effective at improving insomnia, but also improved mood, general and physical fatigue, and global and cognitive dimensions of quality of life. Relapse Prevention CBT describes a number of cognitive and behavioural strategies to help people change their behaviour. Marlatt and Gordon (1985) developed a relapse prevention model to explore the processes that occur when a change in behaviour fails to last and people relapse. This model was developed in the context of addictions to substances such as nicotine, alcohol and drugs but has implications for understanding all other forms of behaviour change that may or may not be sustained. The relapse prevention model was based on the following concept of addictive behaviours: Addictive behaviours are learned and therefore can be unlearned; they are reversible. Addictions are not ‘all or nothing’ but exist on a continuum. Lapses from abstinence are likely and acceptable. Believing that ‘one drink = a drunk’ is a self-fulfilling prophecy. Marlatt and Gordon distinguished between a lapse, which entails a minor slip (e.g. a cigarette, a cou- ple of drinks), and a relapse, which entails a return to former behaviour (e.g. smoking 20 cigarettes, getting drunk). They examined the processes involved in the progression from abstinence to relapse and in particular assessed the mechanisms that may explain the transition from lapse to relapse (see Figure 7.2). These processes are described below. Coping Increased No response self-efficacy relapse High-risk situation No coping Decreased Lapse Abstinence Relapse response self-efficacy Initial violation effect Positive use of Cognitive outcome substance dissonance expectancies Internal attributions Figure 7.2 The relapse process SOURCE: Adapted from Marlatt and Gordon (1985) Baseline State Abstinence. If an individual sets total abstinence as the goal, then this stage represents the target behaviour and indicates a state of behavioural control. Pre-Lapse State High-risk situation. A high-risk situation is any situation that may motivate the individual to carry out the behaviour. Such situations may be either external cues, such as someone else smoking or the availability of alcohol, or internal cues, such as anxiety. Research indicates that the most commonly reported high-risk situations are negative emotions, interpersonal conflict and social pressure  CHAPTER 7 Changing Health Behaviours 179 (Marlatt and Gordon 1985). This is in line with social learning theories, which predict that internal cues are more problematic than external cues. Coping behaviour. Once exposed to a high-risk situation the individual engages the coping strategies. Such strategies may be behavioural, such as avoiding the situation or using a substitute behaviour (e.g. eating), or cognitive, such as remembering why they are attempting to abstain. Positive outcome expectancies. According to previous experience the individual will either have positive outcome expectancies if the behaviour is carried out (e.g. ‘smoking will make me feel less anxious’) or negative outcome expectancies (e.g. ‘getting drunk will make me feel sick’). No Lapse or Lapse? Marlatt and Gordon (1985) argue that when exposed to a high-risk situation, if an individual can engage good coping mechanisms and also develop negative outcome expectancies, the chances of a lapse will be reduced and the individual’s self-efficacy will be increased. However, if the individual engages poor coping strategies and has positive outcome expectancies, the chances of a lapse will be high and the individual’s self-efficacy will be reduced. No lapse: good coping strategies and negative outcome expectancies will raise self-efficacy, causing the period of abstinence to be maintained. Lapse: poor or no coping strategies and positive outcome expectancies will lower self-efficacy, causing an initial use of the substance (the cigarette, a drink). This lapse will either remain an isolated event and the individual will return to abstinence, or will become a full-blown relapse. Marlatt and Gordon describe this transition as the abstinence violation effect (AVE). The Abstinence Violation Effect (AVE) The transition from initial lapse to full-blown relapse is determined by dissonance conflict and self-attribution. Dissonance is created by a conflict between a self-image as someone who no longer smokes or drinks and the current behaviour (e.g. smoking/drinking). This conflict is exacerbated by a disease model of addictions, which emphasizes ‘all or nothing’, and minimized by a social learning model, which acknowledges the likelihood of lapses. Having lapsed, the individual is motivated to understand the cause of the lapse. If this lapse is attributed to the self (e.g. ‘I am useless, it’s my fault’), this may create guilt and self-blame. This internal attribution may lower self-efficacy, thereby increasing the chances of a full-blown relapse. However, if the lapse is attributed to the external world (e.g. the situation, the presence of others), guilt and self-blame will be reduced and the chances of the lapse remaining a lapse will be increased. Marlatt and Gordon developed a relapse prevention programme based on cognitive behavioural techniques to help prevent lapses turning into full-blown relapses. This programme involved the following procedures: Self-monitoring (what do I do in high-risk situations?). Relapse fantasies (what would it be like to relapse?). Relaxation training/stress management. Skills training (‘How will I say “No” to a drink?’). Contingency contracts (‘When offered a cigarette I will...’). Cognitive restructuring (learning not to make internal attributions for lapses). In the image below the transition from ‘high risk situation’ through to ‘no coping response’ and the ultimate ‘AVE’ can be seen running across the middle. A relapse prevention programme would use the techniques described above to help the individual deal with each of these stages. How these procedures relate to the different stages of relapse is illustrated in Figure 7.3. 180 PART TWO Staying Well: Health Beliefs, Behaviour and Behaviour Change Self-monitoring Relaxation training Contract Behaviour Stress management Reminder card assessment Imagery (what to do if you slip) High-risk No coping Decreased Lapse Abstinence situation response self-efficacy Initial violation effect Positive use of Cognitive outcome substance dissonance expectancies Internal attributions Relapse Skills fantasies training Relapse rehearsal Education Programmed Cognitive about effects relapse restructuring of substance Decision matrix Figure 7.3 Relapse prevention intervention strategies SOURCE: Adapted from Marlatt and Gordon (1985) Relapse prevention has been used in a multitude of different contexts as a means to change behav- iour either on its own or as part of a complex intervention. For example, Roske et al. (2008) explored the impact of a smoking cessation intervention using relapse prevention techniques in women post-pregnancy. The results showed that the intervention predicted both non-smoking and improved self-efficacy by six months. But by one year the intervention group showed smoking levels similar to the control group. Likewise, Bowen et al. (2014) conducted a randomized control trial with 286 participants and found that compared with psycho-education, mindfulness-based relapse prevention led to participants having a significantly lower risk of relapse for drug use and heavy drinking. Learning theory (along with cognitive theory) therefore forms the basis of many interventions to change behaviour. Some of these take the form of behavioural strategies with their emphasis on reinforcement, modelling and associative learning. Many incorporate both cognitive and behavioural strategies such as CBT (with its emphasis on behaviour change) and relapse prevention (with its empha- sis on sustaining change and preventing relapse). 3 SOCIAL COGNITION THEORY Social cognition theory was described in Chapter 2 and emphasizes expectancies, incentives and social cognitions (e.g. Bandura 1986). Expectancies include beliefs such as ‘a poor diet can cause heart dis- ease’, ‘if I changed my diet I could improve my health’ and ‘I could change my diet if I wanted to’. Incentives relate to the impact of the consequences of any behaviour and are closely aligned to rein- forcements. For example, a healthy diet would be continued if an individual lost weight or had more energy but stopped if they became bored. Finally, social cognitions reflect an individual’s represen- tations of their social world in terms of what other people around think about any given behaviour. These constructs form the basis of social cognition models such as the theory of planned  CHAPTER 7 Changing Health Behaviours 181 behaviour (TPB) and were described in Chapter 2 in the context of predicting how people behave and the intention-behaviour gap. This approach has been used to develop interventions to change cog- nitions as a means to change subsequent behaviour and to close the intention-behaviour gap. These different types of intervention are shown in Figure 7.4 and will now be considered. Plans Social cognition Implementation model based intentions Social cognition theory Information giving Figure 7.4 Behaviour change interventions derived from social cognition theory SOCIAL COGNITION MODEL BASED INTERVENTIONS Sutton (2002b, 2010) described a series of steps to develop an intervention based upon the TPB, although he argued that the steps could also be applied to other social cognition models. Step 1: Identify Target Behaviour and Target Population Sutton argued that it is crucial that the target population and behaviour are clearly defined so that all measures used can be specific to that behaviour and population. This is in line with Ajzen’s (1988) notion of correspondence or compatibility. For example, the behaviour should not just be ‘healthy eating’ but ‘eating lettuce with my sandwich at lunchtime in the work canteen’. Accordingly the target behaviour should be defined in terms of action (eat healthily), target (lettuce), time (lunchtime) and context (at work). Step 2: Identify the Most Salient Beliefs about the Target Behaviour in the Target Population Using Open-Ended Questions Sutton then suggests that those developing the intervention carry out an elicitation study to identify the most salient beliefs about the target behaviour in the target population being studied. Elliot et al. (2005) designed an intervention to encourage drivers’ compliance with speed limits and asked ques- tions such as ‘What do you think are the advantages of keeping within the speed limit while driving in a built-up area?’ The most common beliefs are known as modal beliefs and form the basis for the analysis. A modal belief in this situation might be ‘to avoid accidents with pedestrians’. Step 3: Conduct a Study Involving Closed Questions to Determine which Beliefs Are the Best Predictors of Behavioural Intention. Choose the Best Belief as the Target Belief To further help to decide which beliefs to target in the intervention, Sutton (2002b, 2010) suggests carrying out a quantitative study including the salient beliefs identified in Step 2 involving the target population. These data can then be analysed to explore the best predictors of behavioural intentions as a means to decide whether all or only some of the TPB variables need to be included in the intervention. 182 PART TWO Staying Well: Health Beliefs, Behaviour and Behaviour Change Step 4: Analyse the Data to Determine the Beliefs that Best Discriminate between Intenders and Non-Intenders. These Are Further Target Beliefs Next, Sutton suggests that the same data set be used to assess which beliefs (including those identified in Step 2) differentiate between either intenders versus non-intenders or those who either do or do not carry out the target behaviour. These are now the key beliefs to be addressed in the intervention. Step 5: Develop an Intervention to Change these Target Beliefs Finally, Sutton describes how the intervention should then aim to change these beliefs which mostly involves giving strong messages to contradict and change the target population’s beliefs. Using this approach, Elliot et al. (2005) used messages to target individuals’ beliefs about driving speed as follows: Target belief Keeping to 30 mph will make it difficult to keep up with the traffic. Strong message Many drivers think that if they keep to the speed limit they will have difficulty keeping up with the traffic. However, this is a perception rather than a reality for the most part. Consider what driving in a 30 mph area is typically like. Even on larger 30 mph roads, there are roundabouts, traffic lights, pedestrian crossings and other things that make it necessary for traffic to slow down or stop. If a vehicle in front starts to pull away from you, you will often find that by maintaining a speed of 30 mph you will catch up with that vehicle further up the road, because they have had to stop or slow down. They will have saved no significant amount of time and they will have gained little or no advantage. From this perspective the TPB can be used as a framework for developing a behaviour change intervention. However, as Sutton (2002b, 2010) points out, although this process provides clear details about the preliminary work before the intervention, the intervention itself remains unclear. Hardeman et al. (2002) carried out a systematic review of 30 papers which used the TPB as part of an intervention and described a range of frameworks that had been used. These included persuasion, information, increasing skills, goal-setting and rehearsal of skills. These have recently been developed and integrated into a causal modelling approach for the development of behaviour change programmes (Hardeman et al. 2005). Sutton (2002b) indicates that two additional frameworks could also be useful. These are guided mastery experiences, which involve getting people to focus on specific beliefs, and the ‘elaboration likelihood’ model (Petty and Cacioppo 1986), which involves the presentation of ‘strong arguments’ and time for the recipient to think about and elaborate upon these arguments. Studies have also used a range of methods for their interventions including leaflets, videos, lectures and discussions. Evidence for Social Cognition Model Based Interventions There are some problems with using social cognition models for interventions, as follows. How to change beliefs. As Hardeman et al. (2002) found from their systematic review, although many interventions are based upon theory, this is often used for the design of process and out- come measures and to predict intention and behaviour rather than to design the intervention itself. Using the TPB for behaviour change interventions describes which beliefs should be changed but not how to change them. Does behaviour change? A TPB-based intervention assumes that changing salient beliefs will lead to changes in behaviour. However, studies indicate that there is an attenuation effect whereby any changes in beliefs are attenuated by the other variables in the model which reduce their impact upon behaviour (Armitage and Conner 2001; Sniehotta 2009). Further, although there is some evidence that theory-based interventions are successful, whether the use of theory relates to the success of the intervention remains unclear. For example, Hardeman et al. (2002) reported that the use of the TPB to develop the intervention was not predictive of the success of the intervention.  CHAPTER 7 Changing Health Behaviours 183 Do they miss other important factors? Sniehotta (2009) described the ‘bottleneck’ whereby interventions using the TPB assume that all changes in behaviour will be mediated through inten- tions. This, he argues, misses the opportunity to change other relevant factors that may influence behaviour directly, such as changes in the environment, and which do not need to pass through behavioural intentions. Social cognition models have therefore been used to develop behaviour change interventions. To date, however, although they provide a clear structure for evaluating an intervention, the actual intervention and the means to be used to change beliefs require further attention. MAKING PLANS AND IMPLEMENTATION INTENTIONS Much research indicates that although an individual may make an intention to carry out a behaviour this intention is not always translated into practice. This is known as the intention–behaviour gap (see Chapter 2) and appears to result from intenders who do not act rather than non-intenders who do act (Sheeran 2002). Research has highlighted a number of ways that this gap can be closed and in 1993 Gollwitzer defined the notion of implementation intentions which involve the development of simple but specific plans, after intentions, as to what an individual will do given a specific set of environmental factors. Therefore implementation intentions describe the ‘what’ and the ‘when’ of a particular behav- iour. For example, the intention ‘I intend to stop smoking’ will be more likely to be translated into ‘I have stopped smoking’ if the individual makes the implementation intention ‘I intend to stop smoking tomorrow at midday when I have finished my last packet’. Further, ‘I intend to eat healthily’ is more likely to be translated into ‘I am eating healthily’ if the implementation intention ‘I will start to eat healthily by having a salad tomorrow at lunchtime’ is made. Implementation intentions are quite simi- lar to the notion of SMART goals that are used in other disciplines, and which stand for goals that are S – Specific, M – Measurable, A – Achievable, R – Reasonable and T – Timebound. Evidence for Implementation Intention Interventions Some experimental research has shown that encouraging individuals to make implementation inten- tions can actually increase the correlation between intentions and behaviour for a range of behaviours such as adolescent smoking (Conner and Higgins 2010), adult smoking (Armitage 2007b), fruit con- sumption (Armitage 2007a), fruit and vegetable consumption (Gratton et al. 2007), taking a vitamin C pill (Sheeran and Orbell 1998), reducing alcohol intake (Armitage 2009a) and reducing dietary fat (Armitage 2004). Gollwitzer and Sheeran (2006) carried out a meta-analysis of 94 independent tests of the impact of implementation intentions on a range of behavioural goals and concluded that imple- mentation intentions had a medium to large effect on goal attainment. In 2013, Bélanger-Gravel et al. conducted a meta-analysis of 26 studies investigating the effects of implementation intentions on physical activity. The overall effect size was small to medium and they concluded that implementa- tion intentions were particularly effective among student and clinical populations. Implementation intentions therefore provide a simple and easy way to promote and change health-related behaviours. There have, however, been some criticisms of this approach, as follows: Do people make plans when asked to? Research indicates that between 20 and 40 per cent of people do not make implementation intentions when they are asked to (Skar et al. 2008). This must influence the effectiveness of any intervention. The impact of existing plans. Implementation intention interventions ask people to make plans and then explore the impact of this on their behaviour. This assumes that people have not already made plans. Sniehotta et al. (2005) suggest that we need to differentiate between spontaneous plans and those made in response to the interventions. Do people own their plans? If people have made their own spontaneous plans they may well feel more ownership of these plans, making them more motivated to carry them out. This is not usually assessed in intervention studies. 184 PART TWO Staying Well: Health Beliefs, Behaviour and Behaviour Change Is all behaviour change volitional? Central to a social cognition model approach to behaviour and the use of implementation intentions is the assumption that behaviour is volitional and under the control of the individual. Much behaviour, however, may be either habitual or in response to environmental changes. These aspects are not addressed within this framework. Not all plans are the same. Sniehotta (2009) argues that research often treats all types of plans as the same but that it is important to differentiate between action plans and coping plans. Action plans involve choosing the behaviour that will achieve the goal (the where, when and how of the behav- iour) and are what have been called implementation intentions. In contrast, coping plans prepare an individual for successfully managing high-risk situations in which strong cues might encourage them to engage in unwanted habits (without intention) or new unhealthy behaviours (with inten- tion). Sniehotta et al. (2006) explored the relative impact of action and coping plans in promoting physical activity post-heart attack and concluded that those in the combined planning group did more activity than those in either the action planning group or the usual care group. INFORMATION-GIVING If a person believes ‘I smoke but I am not at risk of getting lung cancer’ or ‘I eat a high fat diet but my heart is healthy’, then the obvious first starting point to change their behaviour would be to improve their knowledge about their health. This has been the perspective of health education and health promotion campaigns for decades and has resulted in information provision through leaflets, billboards, TV advertisements and group-based seminars and lectures. Some research has evaluated the impact of information-giving using a range of mediums. For example, O’Brien and Lee (1990) manipulated knowl- edge about pap tests for cervical cancer by showing subjects an informative video and reported that not only did the video improve knowledge but that the resulting increased knowledge was related to future healthy behaviour. Further, Hammond et al. (2003) examined the effectiveness of the warning labels on cigarette packets and showed that the intention to stop smoking in the next six months and the number of quit attempts was higher in those who reported reading, thinking about and discussing the labels with other people. Similar results were also found in adolescents who were either established or occasional smokers (White et al. 2008; Germain et al. 2010). The provision of information is often incorporated into more complex interventions such as CBT, relapse prevention and psychoeducational interventions with people in rehabilitation (e.g. Dusseldorp et al. 1999; Sebregts et al. 2000; Rees et al. 2004). In 2017, Collins et al. carried out a systematic review of six systematic reviews investigating the impact of providing risk information regarding cardiovascular disease. They concluded that while increasing the accuracy of risk perception in adults, there was no evidence that information giving reduced the incidence of cardiovascular disease. Generally it is accepted that giving information is not sufficient to change behaviour but that it is a useful and necessary adjunct to any other form of behaviour change strategy. Social cognition theory has therefore informed much research on predicting and explaining health- related behaviours (see Chapter 2). It has also been the basis for behaviour change interventions, particularly through the use of the TPB and planning in the form of implementation intentions. In addi- tion, interventions often use information and education as a means to improve knowledge and change cognitions. Such approaches are not without their problems, however, particularly in terms of their narrow focus on intentions and behaviour and the focus on behaviour as a response to intentions and other related cognitions. 4 STAGE MODELS Strategies to change behaviour based upon both learning theory and social cognition theory concep- tualize behaviour as a continuum and change behaviour by encouraging people to move along the continuum from unhealthy to healthy ways of acting. In contrast, stage models of behaviour such as  CHAPTER 7 Changing Health Behaviours 185 the Stages of Change (SOC) and the health action process approach (HAPA) emphasize differences between people who are at different stages (see Chapter 2). Stage models have influenced behaviour change interventions in two ways: the use of stage-matched interventions and the development of moti- vational interviewing. STAGE-MATCHED INTERVENTIONS A stage model approach to behaviour highlights how people show different levels of motivation to change their behaviour at different stages. Therefore, someone at the pre-contemplation stage is less likely to attend a smoking cessation clinic or wear a nicotine replacement patch than someone at the contemplation or action stages. A stage approach has often been combined with the many strategies derived from learning and cognitive theory or social cognition theory described above so that interven- tions can be targeted to people according to where they are in the process of change. This has taken the form of either tailored or stage-matched interventions. Participants are initially asked to rate their motivation as a means to assess their stage and then the intervention is delivered accordingly. At times this results in people being refused entry into the study as they are at the pre-contemplation stage and deemed not ready to change. Overall, it means that interventions tend to be more effective as the intervention makes more sense to the individual and those who would not have responded to the inter- vention are removed from the study. Stage-matched interventions have been used across a number of behaviours such as smoking cessation (Di Clemente et al. 1991; Aveyard et al. 2006) and cervical cancer screening (Luszczynska et al. 2010). For example, Lu et al. (2019) carried out a nurse-led smoking cessa- tion stage-matched intervention in patients with coronary heart disease or diabetes and concluded that it resulted in successful smoking reduction and abstinence at 3 and 6 months. Stage-matched interven- tions are often used in conjunction with a range of intervention approaches such as CBT, counselling, implementation intentions and planning. MOTIVATIONAL INTERVIEWING (MI) If people are at a stage when they are unmotivated to change their behaviour, then there seems little point in offering them an intervention or including them in a study, particularly as motivation is a consistently good predictor of behavioural intentions and behaviour (e.g. Jacobs et al. 2011). Motivational interviewing (MI) was developed by Miller and Rollnick (2002) as a way to help people consider changing their behaviour and to increase their motivation to change. From a stages of change perspective it takes people from a pre-contemplation to a contemplation stage in the behaviour change process. MI therefore doesn’t show people how to change but encourages them to think about their behaviour in ways that may make them realize that they should change. MI was originally used with people with addictions but is now used across all health care settings and has become a core part of the toolkit of any health professional. The process of MI is based upon the idea that cognitive dissonance is uncomfortable and that people are motivated to get out of a state of dissonance by changing their cognitions (Festinger 1957). For health-related behaviours, conflicting beliefs such as ‘My drug addiction lost me my job’ and ‘I like taking drugs’ or ‘My weight makes it difficult for me to move’ and ‘I like eating a lot’ cause cognitive dissonance and are uncomfortable. The aim of MI is to encourage people to focus on these conflicting beliefs and therefore feel the discomfort more strongly. Questions asked would include: ‘What are some of the good things about smoking/eating a lot/taking drugs?’ ‘What are the not so good things about smoking/eating a lot/taking drugs?’ The client is then encouraged to elaborate on the costs and benefits of their behaviour which are then fed back to them by the health professional to highlight the conflict between these two sets of cognitions: ‘So your smoking makes you feel relaxed but you are finding it hard to climb stairs?’ ‘So taking drugs helps you cope but you have lost your job because of them?’ 186 PART TWO Staying Well: Health Beliefs, Behaviour and Behaviour Change Next, the client is asked to describe how this conflict makes them feel and to consider how things could be different if they changed their behaviour. It is hoped that by focusing on their cognitive dissonance they will be motivated to change both their cognitions and behaviour as a means to resolve this dissonance. Obviously it is hoped that they will change towards being healthier, although this may not always be the case as the process could encourage people to see the benefits of their behaviour and decide to continue as they are. Miller and Rollnick (2002) are very clear that their approach should be non-confrontational and should encourage people to think about the possibility of change rather than persuading them to change. This is particularly important for clients who may have already met much frustration and anger at their behaviour from other professionals or family members and may be very reluctant to speak openly. Miller and Rollnick also emphasize that professionals using MI should be empathic and non-judgemental and should assume that the client is responsible for the decision to change when and if they want to make that decision. A systematic review shows that MI is an effective tool for use by non-specialists for drug abuse treatment (Dunn et al. 2001). Research also shows that MI is effective across a number of areas including promoting attendance at a drug treatment programme (Heslop et al. 2001), enhancing medication compliance in people with schizophrenia (Bellack and DiClemente 1999), treating eating disorders (Killick and Allen 1997), reducing problem drinking in inpa- tients with psychiatric problems (Hulse and Tait 2002), promoting healthy eating (Resnicow et al. 2001), promoting a short-term increase in physical activity in a primary care setting (Harland et al. 1999) and training pharmacists to deliver smoking cessation (Caponnetto et al. 2017). Recently, however, several systematic reviews have found mixed results for the effectiveness of MI. For example, Morton et al. (2015) conducted a systematic review to investigate the impact of MI on health behaviour change in primary care settings. They included 33 papers focusing on physical activity, dietary behaviours and/ or alcohol intake and found that around 50 per cent of studies reported a positive effect of MI on the health behaviours. However, they also concluded that the efficacy of MI remains unclear due to the inconsistency of MI descriptions and intervention components. Likewise, Lindson et al.’s (2019) systematic review of 37 studies evaluated the efficacy of MI for smoking cessation and found mixed results and Michalopoulou et al. (2022) concluded from their systematic review and meta-analysis of 46 studies suggested that there is no evidence that MI increases effectiveness of behavioural weight management programmes in controlling weight. Stage models have therefore influenced behaviour change interventions through the use of tailored or stage-matched interventions and the development of motivational interviewing as a means to move people to a stage where they might consider entering an intervention to change their behaviour. 5 THE ROLE OF AFFECT One of the main criticisms of many psychological theories of behaviour and the strategies used to change behaviour is that they do not address an individual’s emotions and consider people to be rational processors of information (van der Pligt et al. 1998; van den Berg et al. 2005). Some studies, however, have included a role for affect and this has taken various forms including fear appeals, visualization and self-affirmation interventions. These are illustrated in Figure 7.5. USING FEAR APPEALS In recognition of the role that emotion plays in behaviour, many health promotion campaigns include fear appeals which are designed to raise fear as a means to change how people behave. For example, in the 1980s tombstone images were used to promote awareness about the dangers of AIDS, pictures of emaciated and wasted people were used in the 2000s to discourage drug use and cigarette warnings on packets show images of cancer or describe the problems of impotency or harm to unborn children.  CHAPTER 7 Changing Health Behaviours 187 Visualization Changing affect Fear appeals Self-affirmation Figure 7.5 Behaviour change interventions based on a changing affect Fear appeals typically provide two types of message relating to fear arousal and safety conditions as follows: 1 Fear arousal which involves: There is a threat: ‘HIV infection’, ‘lung cancer’. You are at risk: ‘Unsafe sex or sharing needles puts you at risk of HIV’. The threat is serious: ‘HIV kills’, ‘Lung cancer kills’. 2 Safety conditions which involve: A recommended protective action: ‘Use condoms’, ‘Don’t share needles’, ‘Stop smoking’. The action is effective: ‘Condoms prevent HIV’, ‘Stopping smoking prevents lung cancer’. The action is easy: ‘Condoms are easy to buy and easy to use’. Together, fear arousals and safety conditions are designed to generate an emotional response (i.e. fear) and offer a simple way to manage the threat (i.e. behaviour change). However, research indicates that it is not clear whether, when or how fear appeals work. For example, originally it was believed that fear had a U-shaped impact on behaviour with maxi- mum change resulting from moderate fear, while low fear caused no effect and high levels of fear resulted in denial, defensiveness and inaction (Janis 1967). Subsequent research, however, indicates that a more linear rela- tionship may exist with greater levels of fear being the most effective at changing behaviour, although this might be due to the kinds of fear that can be generated in the laboratory setting (Ruiter et al. 2001). For example, Tannenbaum et al. (2015) conducted a meta-analysis on 127 papers investigating the impact of fear appeals on Fear appeals: this gravestone attitudes, intentions and behaviour change. They con- advertisement was one of the first cluded that fear appeals positively influence attitudes, health promotion attempts after the intentions and behaviours and are more effective when identification of the HIV virus they depict high levels of fear, include an efficacy mes- SOURCE: © Department of Health, Reproduced sage and stress severity and risk. Fear appeals also under the Open Government Licence v3.0. seemed to be more effective for women and when they https://www.nationalarchives.gov.uk/doc/ recommended one-time only behaviours. open-government-licence/version/3/ 188 PART TWO Staying Well: Health Beliefs, Behaviour and Behaviour Change The Problem of Blocking The key problem with fear appeals and the process of arousing strong emotions is that when aroused, people tend to block the information they are hearing. Therefore, when presented with messages try- ing to change their behaviour, many people resist, using a number of strategies such as avoidance, ignoring and finding fault in the arguments used, or criticizing the mode of presentation (Jacks and Cameron 2003; Harris and Epton 2009). They therefore think ‘what do you know?’, ‘scientists are always changing their minds’, ‘this leaflet is poorly designed’, ‘the font is too small on that poster’, ‘this health professional is too young’ or ‘this health professional is too old’. An example of this is smokers’ abil- ity to continue to smoke even when the words ‘smoking kills’ are written on their packet of cigarettes. In fact research suggests that those least persuaded by risk data are often those most at risk and that they can either ignore unwelcome information or find reasons for rejecting it (Sherman et al. 2000; Jacks and Cameron 2003). This process of blocking creates a problem for those trying to change behaviour as the message cannot get through. Research has highlighted three potential strategies to counteract this tendency to block: the use of visual images, self-affirmation and focusing on benefits to someone else. Using Visual Images The saying ‘a picture paints a thousand words’ reflects the belief that visual images may be more effec- tive at conveying information or changing beliefs compared to language-based messages. This forms the basis of most advertising, marketing and health education campaigns and is central to the use of diagrams and illustrations throughout education. Some research has explored the impact of visual images in health research (see Chapter 9 for a discussion of imagery and changes in illness cognitions). For example, Hammond et al. (2003) examined the effectiveness of the cigarette warning labels and reported an association between reading and discussing the labels with a higher intention to stop smoking, more quit attempts and a reduction in smoking. Shahab et al. (2007) also reported that showing smokers images of their carotid arteries with a plaque compared to an artery without a plaque increased intentions to stop smoking in those with higher self-efficacy and Kang and Lin (2015) showed that visual fear appeals reduced optimistic bias in smokers. Further, Karamanidou et al. (2008) reported that showing patients with renal disease a plastic container in which they could see how their phosphate binding medication would work in their stomachs changed their beliefs about treatment. In a similar vein, Lee et al. (2011) used a web-based intervention to show participants images of heart disease (with or without text) and concluded that imagery caused more changes than text alone but that a combination of the two forms of information was the most effective. To further unpack the impact of images, Byrne et al. (2019) compared full-colour graphic warning images (GWI) about the harms of smoking vs black and white GWIs vs prominent text-only warnings vs brand images. The results showed that both youth and adult smokers paid more attention to full-colour GWIs than black-and-white GWIs; that for adults, images (regardless of colour) generated more negative affect than text only, and that text only generated more negative affect than brand imagery; and that in young smokers the colour of the image made no difference, but they reported greater negative affect after the warning images compared to the brand images. Images therefore seem to change cognitions and behaviour but to date little is known about the mechanisms behind this process. However, researchers have begun to theorize about this process and have suggested that images may be processed more rapidly than text and may be more memorable over time. In addition, images may also have a greater impact upon affect than text which in turn influences behaviour (Cameron 2008, 2009; Cameron and Chan 2008). Accordingly, visual images may be a means to prevent blocking when people are presented with emotional information. Using Self-affirmation Self-affirmation theory is grounded in the idea of ‘self-integrity’ and argues that people are inherently motivated to maintain their self-integrity and their sense of self as being ‘adaptively and morally ade- quate’ (Steele 1988). If a person’s integrity is challenged by information indicating that their behaviour  CHAPTER 7 Changing Health Behaviours 189 is damaging, then they resist this information as a means to preserve their sense of self. This per- spective provides a framework for understanding the process of blocking. It also highlights a way to encourage people to stop blocking and respond to the message in the desired way. In particular, self affirmation theory indicates that resistance can be reduced if the individual is encouraged to enhance their self-integrity by affirming their self-worth by focusing on other factors that are core to how they see themselves but unrelated to the threat (Harris and Epton 2009). A self-affirmation intervention can take many forms and studies have used methods such as providing positive feedback on a test, asking participants to rate themselves on a series of key values, writing an essay on their most important value or asking a series of questions about a universally valued construct such as kindness (e.g. ‘Have you ever forgiven another person when they have hurt you?’). Therefore, if presented with information that threatens their sense of self, they behave defensively and either ignore or reject it. However, if given the opportunity to self-affirm in another domain of their lives, then their need to become defensive is reduced. For example, if a smoker thinks that they are a sensible person, when confronted with a message that says that smoking is not sensible, their integrity is threatened and they behave defensively by blocking the information. If given the chance, however, to think about another area in which they are sensible, then they are less likely to become defensive about the anti-smoking message. Epton et al. (2015) conducted a meta-analysis of 144 experimental tests of self-affirmation on health message acceptance, intentions to change, and subsequent behaviour. They found that, in general, deploying self-affirmation had a positive effect on all outcomes although effect sizes were small but comparable to others found in meta-analyses of other health behaviour interventions. In particular, they showed that self-affirmation can increase message acceptance for information relating to caffeine consumption, smoking, sun safety, alcohol intake and safe sex (e.g. Sherman et al. 2000; Harris and Napper 2005; Harris et al. 2007). It can change affect (Harris and Napper 2005; Harris et al. 2007) and attitudes (Jessop et al. 2009) and in general cause increases in behavioural intentions (Harris et al. 2007). The review also indicates that self-affirmation interventions can change behav- iour in the short term (e.g. immediately taking a leaflet or buying condoms) but to date there is little evidence on longer-term changes in behaviour. Further, those most at risk (e.g. heavy drinkers, heavy smokers) seem to be more responsive to self-affirmation interventions than those less at risk. Focusing on benefit to someone else When someone is asked (or told) to change their behaviour they can feel affronted as it challenges their sense of self and personal integrity. They can therefore block this information and may denigrate the messenger by thinking ‘what do you know?’, ‘scientists are always changing their minds’, ‘the leaflet is poorly designed’. Some people also resort to their sense of freedom and might think ‘it’s my body I can do what I like’, ‘it’s my choice’ and may even decide to ‘live fast die young’. One possible way to combat this ‘inner libertarian’ is by encouraging people to think about someone else rather than them- selves. Interestingly, the smoking ban could be implemented once we started to understand the risks of passive smoking; the inner libertarian voice saying ‘it’s my body’ no longer worked. Likewise, the law enforcing seatbelt wearing in the back of a car could be brought in once advertising showed us that people in the back not wearing seatbelts could kill people in the front (Ogden 2014). To date, there is not much research within health psychology focusing on the benefits of focusing on others but within envi- ronmental psychology some research points to the benefits of moral motives, biospheric and altruistic (self-transcendent) values and moral norms (i.e. being good) as more effective at changing behaviour than selfishly motivated benefits, such as saving money (i.e. being greedy) (eg. Bolderdijk et al. 2013). This would be a useful avenue for research to address. USING AFFECT EFFECTIVELY Health psychology (and even psychology in general) is often criticized for ignoring emotion. Fear appeals use emotion to change behaviour yet can be met with resistance with people blocking 190 PART TWO Staying Well: Health Beliefs, Behaviour and Behaviour Change the information as it challenges their sense of integrity. Visual images, self-affirmation and focusing on others are useful approaches to limit this process of blocking. How fear appeals, visual images, self affirmation and focusing on others can work together is shown below, with the example of someone who is living with obesity: Fear appeal: ‘Being overweight can cause heart disease’. Emotional response: ‘Anxiety’. Resistance: Ignoring the message/thinking ‘research is always wrong’, ‘that leaflet isn’t very well designed’ or ‘it’s my life and I can live it however I want to’. Self-affirmation intervention: ‘Think of times when you have been kind to others’. Visual image: Here is an image of fatty deposits on an artery. Focusing on others: ‘Being healthier would also help your children to be healthier’. Emotional response: ‘I am reassured’, ‘I am a good person’, ‘I am a good parent’. Reaction to fear appeal: ‘Maybe I should lose some weight’. Affect can therefore be used to change behaviour. This can involve fear appeals which generate strong emotions, the use of visual images, the use of self-affirmation and taking the focus away from the self to counter the voice of the ‘inner libertarian’. 6 INTEGRATED APPROACHES So far this chapter has outlined the wide range of different behaviour change strategies based upon four theoretical frameworks: learning and cognitive theory; social cognition theory; stage models and the use of affect. Over recent years there has been a call to integrate these different perspectives to deliver more successful interventions. This has taken four different approaches: creating a science of behaviour change interventions; the use of modern technologies; the use of mass media and understanding sustained behaviour change. CREATING A SCIENCE OF BEHAVIOUR CHANGE INTERVENTIONS Over the past decade there has been a call to improve intervention research in the following ways: to improve the reporting of interventions to make the process more transparent and easier to synthesize and replicate; to identify which aspects of behaviour change interventions are effective; to improve the design and therefore effectiveness of behaviour change interventions (e.g. Abraham and Michie 2008; Michie et al. 2009; Michie and Wood, 2015; West et al. 2010). This call for a science of behaviour change interventions has involved two key approaches: (i) the integration of theories of behaviour change; and (ii) the development of a taxonomy of behaviour change techniques (BCTs). This drive for a science of behaviour change will now be described. 1. The Integration of Theories of Behaviour Change Due to the proliferation of theories of health behaviour (see Chapter 2) and the number of theoretical approaches to behaviour change described in this chapter, Michie and colleagues proposed an integrated approach to behaviour change (e.g. Abraham and Michie 2008; Michie et al. 2011b; Michie et al. 2014b; Atkins et al. 2015; Michie and Wood 2015). This has resulted in the following: The COM-B Following a cross disciplinary review of theories of behaviour and behaviour change including 83 theories and 1,659 constructs, the COM-B was created to reflect a comprehensive and parsimonious approach to behaviour and the factors necessary for behaviour change to occur. The COM-B highlights the key role of Capability (derived from the individual’s psychological or physical ability to enact the  CHAPTER 7 Changing Health Behaviours 191 behaviour), Opportunity (reflecting the physical and social environment that enables the behaviour) and Motivation (describing the reflective and automatic mechanisms that activate or inhibit the behaviour). In turn, these factors predict Behaviour. This is discussed further in Chapter 2 and illustrated in Figure 7.6. Psychological or physical ability to Capability enact the behaviour Physical and social environment that Opportunity Behaviour enables the behaviour Reflective and automatic mechanisms that activate or inhibit Motivation the behaviour Figure 7.6 The COM-B SOURCE: Michie et al. (2011b) The COM-B finds reflection in the models used in a number of domains such as criminology (the opportunity, motive, capability triad) and workplace and environmental interventions (Motivations, opportunities, ability model, MOA) and has been used to predict and explain a multitude of behaviours including physical activity, weight loss, hand hygiene, dental hygiene, diet, smoking, medication adher- ence, prescribing behaviours, condom use, female genital mutilation and hygiene practices during the COVID pandemic (e.g. Jackson et al. 2014; Brown et al. 2015; Bailey et al. 2016; Chadwick and Benelam 2013; Asimakopoulou and Newton 2015; Gibson Miller et al. 2020; Heneghan et al. 2020; Willmott et al. 2021; see Michie and Wood 2015; Atkins et al. 2015 and Michie et al. 2014b for reviews). The Theoretical Domains Framework (TDF) In addition to the COM-B, Michie and colleagues have developed the Theoretical Domains Framework (TDF) as a means to synthesize across the different theories that psychologists use to predict behaviour and design and evaluate behaviour change interventions (see Atkins et al. 2017 for a review). The TDF was initially developed in the context of health professional behaviour and involved the synthesis of 33 theories of behaviour and behaviour change which were clustered into 14 domains (Michie et al. 2005; Cane et al. 2012). It was then extended to patient behaviours in general (e.g. Kolehmainen et al. 2011; Taylor et al. 2013). More recently the researchers have identified 33 theories and 128 theoretical constructs which have also been grouped into a similar structure of 14 domains and a guide has been developed for the use of the TDF in the development of interventions (Davis et al. 2015; Atkins et al. 2017). These domains include social influences, environmental context and resources, physical skills, emotion, memory, attention and decision processes, knowledge, beliefs about consequences and beliefs about capabilities. The COM-B and the TDF have therefore been developed as a means to integrate existing theories of behaviour and behaviour change and provide a framework for developing and evaluating interventions. 192 PART TWO Staying Well: Health Beliefs, Behaviour and Behaviour Change Together they have been used to frame a wide range of interventions focusing on issues such as nutritional adherence in performance sport and cervical screening (e.g. Bentley et al. 2019; O’Donovan et al. 2021). 2. The Development of a Taxonomy of Behaviour Change Techniques (BCTs) As part of the drive to improve the reporting and effectiveness of interventions, Michie and colleagues have also developed a taxonomy of behaviour change techniques (BCTs). To achieve this, they have coded and classified the components of a vast range of protocols and interventions to label which strategies are being used. This process has gone through several permutations and has generated a long and comprehensive list of all the different actions that can be carried out to bring about behav- iour change. The ultimate goal is that these techniques can be matched to their target behaviour and target populations so that researchers can calculate which techniques are most effective at produc- ing change in which behaviours and which populations (Abraham and Michie 2008; Michie et al. 2011; Michie et al. 2014b; Atkins et al. 2015; Michie and Wood 2015). This would mean that if a practitioner wanted to change smoking (the behaviour) in teenagers (the population) then they would be able to identify which techniques to include in their intervention to maximize the chance of success. The most recent version of the taxonomy highlights 93 possible techniques (Michie et al. 2013). Behaviour change techniques (BCTs) include goals and planning, feedback, repetition and substitution, and comparison of behaviour. Recently, Brown et al. (2019) carried out a systematic review of 32 studies to identify which BCT components could be considered ‘promising’, or likely to be effective in future interventions to maintain smoking abstinence after pregnancy. The researchers identified six BCTs as promising: ‘problem solving’, ‘information about health consequences’, ‘information about social and environmen- tal consequences’, ‘social support’, ‘reduce negative emotions’ and ‘instruction on how to perform a behaviour’. They also concluded that tailored self-help approaches, with or without counselling, may be the most effective modes of delivery of these BCTs. The Behaviour Change Wheel (BCW) Michie and colleagues have therefore integrated existing theories of behaviour change to produce the COM-B and the TDF. They have also identified and coded a wide range of BCTs as a means to improve both the reporting and effectiveness of interventions and also to target specific techniques at specific behaviours and populations. In 2011, Michie et al. carried out a synthesis of all the different types of taxonomy as a means to identify essential conditions for behaviour change and how these could be turned into actual behaviour change. From this process, the researchers created a behaviour change wheel with three levels illustrating the translational process from essential conditions, through intervention functions, to policy (see Figure 7.7). Essential conditions: the researchers identified three conditions which are deemed essential for behaviour and behaviour change: capability, motivation and opportunity. These constructs reflect the COM-B. Intervention functions: it is argued that changing behaviour requires a change in these essential conditions and that a series of intervention functions can bring this change about. The nine functions identified in the behaviour change wheel reflect a synthesized version of the many strategies that are used to change behaviour and were derived from a detailed coding process. These reflect the BCTs. Categories of policy: finally, the researchers argue that policy changes are needed to enable the interventions to occur. The end result of the behaviour change wheel would be that the policy enables interventions to occur, which in turn change the essential conditions of behaviour which bring about changes in behaviour. The model can be used to describe and understand why interventions do or do not work. It could also be used to design more effective interventions which could be linked to policy or even used to promote new policy.  CHAPTER 7 Changing Health Behaviours 193 Figure 7.7 The behaviour change wheel SOURCE: Michie et al. (2011b) Case Study: Using the Science of Behaviour Change to Develop and Test an Intervention West and Michie (2015) provide the following simple step-by-step guide to using the science of behav- iour change for developing and evaluating an intervention based upon their longer book (Michie et al. 2014a). This could be to bring about a change in any behaviour (diet, exercise, help seeking, screening, shopping) or for any population (well, ill, old, young, British, African, European). Recently, West et al. (2020) and Lunn et al. (2020) also suggested that the principles of behaviour change could be used to reduce the transmission of COVID-19. The steps are as follows, and reflect moving from the inner ring of the Behaviour Change Wheel outwards: 1 Behavioural target specification: Specify what behaviour needs to change, in what way and for whom. 2 Behavioural diagnosis: Use the COM-B to determine what factors would change the specified behaviour (i.e. Capability, Opportunity, Motivation). 3 Intervention strategy selection: From the COM-B decide what ‘intervention functions’ to use (e.g. Education, Persuasion, Incentivisation, Coercion). 4 Implementation strategy selection: Choose from the possible policy options (e.g. Fiscal policy, Legislation, Regulation). 5 Selection of specific behaviour change techniques: Select the appropriate behaviour change techniques from the taxonomy to design the intervention. 6 Drafting the full intervention specification: Write a detailed intervention. 194 PART TWO Staying Well: Health Beliefs, Behaviour and Behaviour Change These stages may not be linear and may involve cycling back and forth until the task is complete. This process should also involve applying the APEASE criteria to make sure the intervention is suitable and practical (Acceptability, Practicability, Effectiveness/cost-effectiveness, Affordability, Safety/ side-effects, Equity). The intervention should then be evaluated using an appropriate methodology. Over the past decade there has therefore been a drive to improve behaviour change intervention research. This has involved the integration of existing theories of behaviour and behaviour change and the classification of behaviour change techniques. This process has culminated in the Behaviour Change Wheel which can be used as a framework to developing and evaluating interventions. MODERN TECHNOLOGIES This chapter has described a number of different theories and strategies to change health-related behaviour. Recent developments in modern technologies have provided the opportunity for new ways to deliver such strategies and new sources of information that may help individuals change what they do. These include the use of ‘ecological momentary interventions’ (EMIs) or app-based interventions (via palmtop computers or smarthones) and web-based interventions. Behaviour change interventions such as smoking bans, taxation and restricting or banning advertising are described in Chapter 3 in the context of addictive behaviours. Ecological Momentary Interventions (EMIs) Traditionally, interventions occurred in the clinical setting with patients attending individual or group- based therapy sessions. It has long been recognized that interventions are more effective if contact between therapist and patient can be extended beyond these interactions and until recently this has taken the form of homework to ensure that the patient takes the ideas discussed back into their day- to-day lives, or telephone helplines so that people can ring up whenever they need extra support when their resilience is weakened and additional motivation is needed. The development of new technolo- gies, such as smartphones and palmtop computers, provides a simple and cost-effective way to extend therapy beyond the consultation and patients become accessible at all times. The term EMI refers to treatments provided to people during their everyday lives (i.e. in real time) and in natural settings (i.e. in the real world). Such treatments/interventions include text messaging and apps and have been used for a wide range of behaviours such as smoking cessation, weight loss, anxiety, alcohol use, dietary change and exercise promotion. They have also been used across a number of different chronic illnesses including obesity (Ogden et al. 2019) diabetes, coronary heart disease (CHD) and eat- ing disorders (see Heron and Smyth 2010 for a comprehensive review). They are particularly useful for hard-to-reach groups, such as adolescents who would usually avoid contact with health professionals. For example, Sirriyeh et al. (2010) explored the impact of affective text messages and instrumental texts for promoting physical activity in adolescents, with those in the intervention groups receiving one text per day over a two-week period. The results showed that all participants increased their activity levels over the course of the study. In addition, affective texts such as ‘exercise is enjoyable’ were par- ticularly effective at changing behaviour in those who were most inactive at baseline. For their review of the evidence, Heron and Smyth (2010) identified 27 interventions using EMIs to change behaviour and drew three conclusions. First, at the most practical level EMIs can be easily and successfully deliv- ered to the target group. Second, this new approach is acceptable to patients, even those who are hard to reach such as adolescents, and third, EMIs are effective at changing a wide range of behaviours. Recent systematic reviews indicate that text messaging can be effective for smoking (Whittaker et al. 2019), physical activity (King et al. 2008), a reduction in calorie intake (Joo and Kim 2007), weight man- agement (Alamnia et al. 2022), pain management (Fritsch et al. 2020) and diabetes self-management (Sahin et al. 2019), although most changes seem to be in the shorter rather than longer term and reviews often conclude that effect sizes can be small. Recently, research has also used virtual representations (i.e. digital twins) to promote behaviour change as these encourage participants to further engage with  CHAPTER 7 Changing Health Behaviours 195 the image on the app or who they are and who they could become (see Taylor et al. 2022 for a review). They represent a combination of a wide range of behaviour change techniques as well as visual imag- ery and information tailored to the individual. EMIs are a relatively new approach but offer a simple and cost-effective means to change behaviour for a wide section of the population. Web-based Interventions New developments in technologies have also led to the use of web-based interventions so that patients who may be unable or unwilling to attend face-to-face consultations can now engage in a range of therapeutic strategies from their own home to fit in with their own time frame. For example, packages of web-based interventions have been developed for patients with a range of chronic illnesses such as diabetes, asthma, coronary heart disease, AIDS/HIV to deliver treatments such as cognitive behav- iour therapy (CBT), relapse prevention, education and goal-setting. Many also address psychological problems such as depression and anxiety, obsessive compulsive disorder (OCD) and body image distur- bance. Rosser et al. (2009) carried out a systematic review of novel technologies for the management of chronic illnesses (both psychological and physical) and identified that most utilized web-based inter- ventions (53 per cent), with other technologies being interactive CD-ROM programmes, online message boards, video presentations, email