Protection motivation theory and health behaviour : conceptual review, discussion of limitations, and recommendations for best practice and future research

This is a self-archived version of an original article. This version may differ from the original in pagination and typographic details. Author(s): Title: Year: Version: Copyright: Rights: Rights url: Please cite the original version: CC BY 4.0 https://creativecommons.org/licenses/by/4.0/ Protection motivation theory and health behaviour : conceptual review, discussion of limitations, and recommendations for best practice and future research © 2024 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group Published version Balla, Jessica; Hagger, Martin S. Balla, J., & Hagger, M. S. (2024). Protection motivation theory and health behaviour : conceptual review, discussion of limitations, and recommendations for best practice and future research. Health Psychology Review, Early online. https://doi.org/10.1080/17437199.2024.2413011 2024 Health Psychology Review ISSN: (Print) (Online) Journal homepage: www.tandfonline.com/journals/rhpr20 Protection motivation theory and health behaviour: conceptual review, discussion of limitations, and recommendations for best practice and future research Jessica Balla & Martin S. Hagger To cite this article: Jessica Balla & Martin S. Hagger (17 Oct 2024): Protection motivation theory and health behaviour: conceptual review, discussion of limitations, and recommendations for best practice and future research, Health Psychology Review, DOI: 10.1080/17437199.2024.2413011 To link to this article: https://doi.org/10.1080/17437199.2024.2413011 © 2024 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group Published online: 17 Oct 2024. Submit your article to this journal View related articles View Crossmark data Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=rhpr20 REVIEW ARTICLE Protection motivation theory and health behaviour: conceptual review, discussion of limitations, and recommendations for best practice and future research Jessica Balla a and Martin S. Hagger a,b,c,d a Department of Psychological Sciences, University of California, Merced, CA, USA; b Health Sciences Research Institute, University of California, Merced, CA, USA; c Faculty of Sport and Health Sciences, University of Jyväskylä, Jyväskylä, Finland; d School of Applied Psychology, Griffith University, Brisbane, Australia ABSTRACT Protection motivation theory is a pre-eminent health behaviour theory purposed to predict participation in health protection and risk behaviours. It has been widely applied across multiple behaviours, populations and contexts. In this conceptual review, we summarise research applying the theory and identify shortcomings and evidence gaps that limit reported inferences and impede theory and intervention development. Accordingly, we provide recommendations for best practices and suggestions for future research to resolve these limitations. Limitations identified include a dearth of comprehensive theory tests, sparse evidence of theory sufficiency, a lack of studies including additional constructs, overuse of correlational and crosssectional research designs, a paucity of intervention studies and tests of theory-consistent mechanisms of action, few tests intrapersonal and environmental moderators of theory effects and measurement concerns. We provide recommendations to address these limitations including conducting comprehensive theory tests in support of nomological validity; incorporating past behaviour and other constructs to establish theory sufficiency and extend its scope; adopting crosslagged panel and factorial experimental research designs to test directional effects, permit better causal inference and test mechanisms of action; testing effects of moderators to identify conditions that may affect theory applicability and developing measurement standards for study constructs and adopting non-self-report behaviour measures. ARTICLE HISTORY Received 17 March 2024 Accepted 1 October 2024 KEYWORDS Social cognition; behaviour change; health behaviour; theory; behavioural theory; behaviour change interventions Epidemiological research has indicated that regular participation in health-promoting behaviours (e.g., adhering to health screening, regular physical activity participation, following a healthy diet), and long-term cessation, moderation, or avoidance of behaviours that present a risk to health (e.g., limiting sedentary activity, avoiding excessive alcohol consumption, abstinence from smoking tobacco), is consistently related to positive physical and mental health (Ford et al., 2011; Shaw & Agahi, 2012; Warren Andersen et al., 2018). However, population level rates of participation in these health-related behaviours in many countries are insufficient to confer these stipulated health benefits (Bruni et al., 2022; Centers for Disease Control and Prevention, 2021b; Haug et al., 2009), while participation rates for risk behaviours remain at levels likely to pose ongoing threats © 2024 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The terms on which this article has been published allow the posting of the Accepted Manuscript in a repository by the author(s) or with their consent. CONTACT Martin S. Hagger [email protected] HEALTH PSYCHOLOGY REVIEW https://doi.org/10.1080/17437199.2024.2413011 to health (Centers for Disease Control and Prevention, 2021a; Shmulewitz et al., 2021; Silveira et al., 2022). Governmental and community health organisations have, therefore, formally designated promotion of health behaviours and prevention of risk behaviours through intervention a priority (Centers for Disease Control and Prevention, 2022; Spring et al., 2013; World Health Organization, 2021). Accordingly, these organisations have identified the development and promulgation of behaviour change interventions aimed at fostering uptake and maintenance of health behaviours and cessation of risk behaviours as a key means to address this priority, particularly in at-risk populations. Such interventions are more likely to be optimally efficacious in changing behaviour if they are based on theoretical approaches that offer insight into the reliable determinants of health promoting and health risk behaviours and the processes involved (Bishop et al., 2015; Michie, 2008; Protogerou & Johnson, 2014; Rothman, Simpson, et al., 2020; Webb et al., 2010). In particular, researchers have applied theories from the behavioural sciences, particularly psychology, to identify these determinants and processes, as well as potential moderators that may explain contextual variations in determinant effects on behaviour (Hagger et al., 2020a; Johnston et al., 2021; Rothman, Klein, et al., 2020). Knowledge of determinant-behaviour links, and the linked processes and moderators, could highlight potentially modifiable targets for behavioural interventions aimed at promoting behaviour change. Further, these links may also inform work on the methods or techniques that may be efficacious in changing or activating these determinants and the mechanisms of action by which the techniques lead to behavioural enactment (Carey et al., 2018; Connell et al., 2019; Hagger et al., 2020; Sheeran et al., 2016, 2023). These techniques can then form the content of behaviour change interventions for dissemination to targeted populations by various means (e.g., messages highlighting risks, prompts to set goals, provision of feedback on progress, exercises specifying adoption and practice of self-regulatory skills; Dombrowski et al., 2016; Hagger & Hardcastle, 2014). Theories adopting social cognition perspectives are pre-eminent among the theories that have been applied to identify the determinants of health and risk behaviours, and the relevant processes involved (Conner & Norman, 2015). Common to these theories is the assumption that individuals’ decisions to perform a given target behaviour in future are a function of their processing of social information, summarised in the beliefs they hold with respect their future performance of the behaviour (Ajzen, 1991; Conner & Norman, 2015). These beliefs are assumed to be fundamental sources of information that individuals use to form behavioural intentions, a key motivational precursor of the uptake and maintenance of behaviour (Armitage & Conner, 2000). A prominent theory of this type is protection motivation theory (Rogers, 1975), a theory specifically purposed to identify the determinants of behaviours that confer health benefits or reduce health risk and the decisional processes involved. Central to the theory is protection motivation, a construct equivalent to behavioural intention, which is designated as the most proximal correlate of health protective behaviour. Accordingly, protection motivation is proposed to serve as a mediator of the effects of sets of beliefs or appraisals concerning health risks or threats, personal capacity to perform the given behaviour, and the perceived utility of the target behaviour on performance of the behaviour in future. The theory also specifies the role of other key factors likely to be implicated in individuals’ decision to engage in health protective behaviours, such as maladaptive responses that may undermine protection motivation. The theory has been widely applied in research studies to predict health behaviours across multiple populations and contexts (Norman et al., 2015; Orbell et al., 2020; Prentice-Dunn & Rogers, 1986), and meta-analytic syntheses of this research has provided generalised support for theory-stipulated relations among its constructs across multiple health behaviours (Floyd et al., 2000; Milne et al., 2000). This broad empirical support notwithstanding, a number of limitations in the research methods employed to test theory predictions, as well as boundary conditions in theory conceptualisation and specification, have been identified. These limitations and boundary conditions place caveats and restrictions on the interpretation and generalizability of many theory tests available 2 J. BALLA AND M. S. HAGGER in the current literature, and potentially serve to hinder the progress of research adopting the theory to identify behavioural determinants and develop behaviour-change interventions in health contexts (e.g., Cismaru & Lavack, 2007; Milne et al., 2002; Weinstein, 2007). Accordingly, in the current conceptual review of applications of the theory we aim to identify these limitations and boundary conditions, outline how they delimit inferences and affect the quality of current research findings, and provide guidelines on potential alternative research methods and study designs, and conceptual modifications, that may contribute to resolving them. The current review has value in that it is expected to set the agenda for future researchers seeking to develop studies that will address these issues so as to improve the scope and quality of the evidence for theory predictions and extend inferences that can be drawn from research applying it in health behaviour contexts. Protection motivation theory: overview and key hypotheses Protection motivation theory was purposed to identify the determinants of behaviours aimed at managing or coping with health threats, with a focus on illustrating how fear appeals may impact cognitions and subsequent health behaviours (Rogers, 1975). The basic constructs and hypothesised effects of the theory are summarised in Figure 1. A central hypothesis of the theory is that an individual’s protection motivation is the most proximal determinant of the performance of health behaviours aimed at managing, or offering ‘protection’ from, health threats. Protection motivation is a motivational construct indicating an individual’s readiness to adopt or engage in a behavioural response to a given health threat or health-promoting opportunity akin to the intention construct specified in other social cognition theories (see Ajzen, 1991; Fishbein et al., 2001; Triandis, 1977). In the theory, protection motivation is proposed to be a function of two parallel sets of beliefs or appraisals with respect to health threats and behavioural means to manage them: threat appraisals and coping appraisals. Threat appraisals represent beliefs about risks or threats to health (e.g., beliefs about the health risks associated with continuing to smoke cigarettes), and encompass perceived severity and perceived susceptibility or vulnerability, and the extent to which the ‘maladaptive response’, that is, not performing the behaviour, is rewarding. Coping appraisals represent beliefs about the efficacy of the behaviour in promoting desirable health outcomes, known as response efficacy, and the perceived costs of performing the adaptive response, known as response costs. A subsequent revised version of the theory introduced self-efficacy from Bandura’s (1986) social Figure 1. Protection motivation theory. Note: Solid arrowed lines represent key theory predictions and broken lines represent moderation effects. HEALTH PSYCHOLOGY REVIEW 3 cognitive theory as an additional determinant of protection motivation (Maddux & Rogers, 1983). Self-efficacy reflects beliefs regarding personal capacity to perform the requisite behaviour. Protection motivation serves a pivotal role in the theory as it is predicted to mediate relations between the social cognition constructs, namely, the sets of threat and coping appraisals and behaviour. The mediation effects indicate that individuals’ appraisals and associated beliefs with respect to the target behaviour informs the intensity of their motivation to perform it in future and the extent to which they are subsequently prepared to pursue it (see Perugini, 2005). Threat and coping appraisals are also expected to be informed by individuals’ processing of available interpersonal information about the behaviour derived from personal experience (e.g., past behaviour), and from environmental (e.g., social context, messaging) and intrapersonal (e.g., personality) sources. In addition, threat and coping appraisals are hypothesised to affect individuals’ adoption of alternate behavioural responses that run counter to effective threat management behaviours that would be expected to offer protection from the health threat. These maladaptive coping responses are typically adopted by individuals to cope with health threatening information, usually through mitigating the emotional upheaval caused by knowledge of the threat, but are not expected to ultimately lead to adaptive outcomes. Typically, such responses encompass coping procedures such as distraction, emotion suppression, avoidance, and denial, all of which generally tend to be focused on reducing felt negative emotions associated with the threat response (for a discussion see Carver et al., 1989). Given such responses tend to be alternatives to behaviours aimed at promoting adaptive health outcomes, they are expected to be negatively associated with protection motivation. Such a relationship likely reflects individuals prioritising the mitigation of the psychological distress associated with the threat because that is seen as more proximally valuable relative to the adoption of protective behaviours aimed at managing the threat, hence the maladaptive label. The predicted associations between the theory constructs and protection motivation have been hypothesised to occur through multiple mechanisms. Specifically, associations between threat and coping appraisals and behaviours are proposed to be mediated by maladaptive coping as well as protection motivation, and effects of these appraisals on protection motivation are proposed to be moderated by maladaptive coping. The mediated effects reflect the tendency for individuals to respond to high perceptions of threat with emotion-focused coping strategies like avoidance or denial, which are proposed to be negatively related to protection motivation for the adaptive health behaviour. The moderation effects suggests that effects of threat and coping appraisals on protection motivation and behaviour are likely to be undermined among those who tend to adopt maladaptive coping strategies relative to those who do not engage in such behaviours. Hundreds of studies have applied protection motivation theory to predict multiple health behaviours in numerous contexts and populations (e.g., Babazadeh et al., 2017; Guo et al., 2015; Helmes, 2002; Norman et al., 2015; Taylor & May, 1996). Large-sample tests of the theory, often assumed to produce the most reliable estimates of effects, have supported the key hypothesised effects of the theory in behavioural contexts such as physical activity (Plotnikoff et al., 2009, 2010), healthy eating (Chamroonsawasdi et al., 2021), and vaccination (Li & Sun, 2021; Wu et al., 2022), and highlighted its potential value in identifying key behavioural correlates and the processes involved. Similarly, metaanalytic syntheses of this research have broadly corroborated key theory-implied associations among the appraisal constructs, protection motivation, and the coping and behavioural outcomes across studies, and indicate that the theory constructs individually explain substantive variance in protection motivation and health behaviour across contexts, behaviours, and populations (Floyd et al., 2000; Milne et al., 2000). For example, both meta-analyses demonstrated small-sized positive correlations between perceived susceptibility and severity, and between protection motivation and behaviour. Medium-sized negative correlations have also been observed between perceived response costs and protection motivation and behaviour, as well as positive correlations between self-efficacy, response efficacy, and protection motivation and behaviour with highly variable effect sizes, and suggest that associations of coping appraisals and protection motivation and behaviour tend to be larger than for threat appraisals. In addition, these analyses both highlighted the 4 J. BALLA AND M. S. HAGGER paucity of research at the time examining maladaptive response rewards, maladaptive coping responses, response costs, and fear. A key caveat of these meta-analytic findings is that they confine their analysis to zero-order or bivariate correlations among theory constructs. This is in contrast to primary studies that typically adopt multivariate analytic models and, therefore, provide more robust tests of theory predictions because they estimate the unique effects of each theory construct on protection motivation while simultaneously accounting for the effects of the other constructs (Floyd et al., 2000; Milne et al., 2000; Norman et al., 2015). As a consequence, a meta-analysis of research on the theory that adopts multivariate analytic techniques to estimate unique theory effects across the extant literature represents a key type of research that may fill this gap. This approach has been previously applied in meta-analyses of other social cognition theories such as the theories of planned behaviour (e.g., Hamilton et al., 2020; McEachan et al., 2011) and the reasoned action approach (e.g., Hagger et al., 2018; McEachan et al., 2016) and provides a template for procedures that could be usefully applied to research on protection motivation theory. We look to future research to conduct such an analysis, which will not only provide more precise estimates of the unique effects of theory constructs and their associated variability across the literature, but provide the opportunity to update the effects identified by Milne et al. (2000) and Floyd et al. (2000) as well as the opportunity to conduct more extensive moderator analyses. Alongside research derived from correlational data, there has also been experimental and intervention studies adopting randomised controlled designs testing the efficacy of the theory in guiding efforts to change behaviour and promote adaptive outcomes in health contexts (e.g., Gong et al., 2009; Khiyali et al., 2017; Mccullock & Perrault, 2020). It is also important to note that research adopting these designs is more prevalent for protection motivation theory relative to other social cognition models (e.g., the theory of planned behaviour; Hagger, 2019). Such experiments and interventions adopt strategies or techniques that seek to change health behaviours through change in, or activation of, the salient constructs of the theory (e.g., messaging emphasising health risks targeting change in threat appraisals, educational videos highlighting effectiveness of the target behaviour targeting change in response efficacy, prompting successful practice targeting change in self-efficacy). Such research has additional value beyond the previously cited correlational research insofar as they enable evaluation of the extent to which techniques targeting change in specific theory-relevant constructs affect subsequent and concomitant change in behavioural outcomes. Such research permits inference of causal effects and change for theory effects which is contraindicated in research adopting correlational designs. Studies adopting experimental or intervention designs and targeting change in key constructs such as threat and coping appraisals have demonstrated efficacy in changing behaviour in multiple behavioural domains such as sunscreen use, condom use, and physical activity (e.g., Gaston & Prapavessis, 2014; Gong et al., 2009; McClendon & Prentice-Dunn, 2001). Taken together, a substantive body of research has provided support for key predictions of protection motivation theory (Taylor & May, 1996; Wurtele & Maddux, 1987), and the theory has demonstrated efficacy in guiding interventions targeting behaviour change in numerous behavioural contexts (Fruin et al., 1992; Malmir et al., 2018; Prestwich et al., 2008). This generalised support notwithstanding, various shortcomings or limitations of research on the theory have been noted as well as some important boundary conditions of the theory itself. In the next sections, we expand on these limitations and boundary conditions and how they serve to narrow the scope of inferences that researchers can draw regarding theory efficacy and predictive validity and stymie the extent to which it can effectively guide intervention and be broadly generalisable. Alongside this, we set out alternative research designs and methodological innovations that could potentially resolve these issues. Importantly, we provide recommendations of the specific kinds of studies needed to effectively address these limitations and contribute towards the improvement of the predictive validity of the theory. These issues, including their respective resolutions and suggestions for future reading, are summarised in Table 1. Our review follows a similar approach to other conceptual HEALTH PSYCHOLOGY REVIEW 5 Table 1. List of limitations and issues in protection motivation theory research identified in the current review with suggested recommendations and further reading. Limitation/issue Recommendation Further reading 1. Comprehensive theory tests are seldom performed. Perform tests of the theory which encompass most or, preferably, all proposed constructs of the theory, with appropriate conceptual rationale, and their expected relationships in simultaneous tests to confirm nomological validity of the theory. Hagger et al. (2017); Maddux and Rogers (1983); Wang et al. (2019) 2. Few studies measure maladaptive coping responses, with a paucity of tests of mediation and moderation effects involving maladaptive coping. Maladaptive coping should be measured as standard when testing the theory, along with tests of both mediation and moderation effects involving this construct. Brown et al. (2005); Ho (2000); Rippetoe and Rogers (1987) 3. Need for widespread examination of the interaction between threat and coping appraisals. Routinely incorporate tests of these interaction effects in studies. Cismaru and Lavack (2007); Pechmann et al. (2003); Rogers (1975) 4. Past behaviour is seldom included in tests of the theory to provide evidence for its sufficiency and capture other unmeasured constructs. Include a measure of past behaviour as a predictor of all theory constructs to test theory sufficiency and effects of unmeasured constructs. Ajzen (1991); Hodgkins and Orbell (1998); Ouellette and Wood (1998) 5. Few studies test effects of additional constructs not originally encompassed within the theory on protection motivation and behaviour. Include additional constructs from other theories when testing theory effects such as affective and implicit attitude measures, again, with appropriate conceptual rationale, to increase scope of theory. Ajzen (1991); Bryan et al. (1997); Millar (2011) 6. The majority of research testing protection motivation theory adopts correlational and cross-sectional study designs. Adopt longitudinal panel, experimental, and intervention designs wherever possible to facilitate directional and causal inferences. Hagger et al. (2020a, 2020b); Milne et al. (2000) 7. Current experimental and intervention research on the theory tends to target multiple constructs in single condition designs. Utilize factorial experimental and intervention designs that isolate effects of individual behaviour change techniques allowing evaluation of the unique and interactive effects of each technique, and, potentially, intervention mechanisms of action. Hagger et al. (2020); Orbell et al. (2020) 8. Current studies tend not to adopt longitudinal designs that allow for inference of directional effects. Longitudinal, cross-lagged panel designs should be routinely adopted to examine direction and reciprocity in effects among theory constructs while controlling for stability. Dorsey et al. (1999); Gollob and Reichardt (1987); Hagger and Hamilton (2023); Lindwall et al. (2011); Van der Velde and Van der Pligt (1991). 9. Little research on effects of candidate moderator variables on theory effects, such as individual difference constructs and socio-structural variables. Conduct more systematic study of these types of moderators, including examples such as socio-economic status and personality traits, to support generalizability of theory effects. Amireault et al. (2008); Brouwers and Sorrentino (1993) 10. Use of standardised measures of theory constructs are not always adopted. Application of formal classification procedures to ensure standardised theory measures are developed and widely adopted. Ajzen (1991); Ajzen and Fishbein (1980); Hagger and Chatzisarantis (2009); Starfelt Sutton and White (2016) 11. Salient beliefs of the target population regarding the health threat or behaviour are seldom used as a basis for development of theory construct measures. Conduct open-ended, formative research with the target population to solicit beliefs and guide subsequent development of construct measures. Ajzen (2002); Norman et al. (2015); Searle et al. (2000) 12. A majority of studies adopt selfreport measures of behaviour. Employ non-self-report measures of behaviour whenever possible (e.g., observation, other wearable devices), to reduce biases associated with selfreport measures. Gaston and Prapavessis (2014); Hall et al. (2018); Norman et al. (2015); Sheeran et al. (2017) 6 J. BALLA AND M. S. HAGGER reviews in the health psychology domain (e.g., Hagger & Orbell, 2022; Swann et al., 2021) and those specific to protection motivation theory in other domains (e.g., Clubb & Hinkle, 2015; Westcott et al., 2017), in that we provide an overview of the theory and current literature informed by the available primary research and evidence reviews, comment on and critically analyse current work, and make recommendations to address the limitations or issues arising from the review. Specifically, our review will highlight the need for (a) research addressing the paucity of comprehensive tests of theory, particularly tests of theory-stipulated interaction and mediation effects such as those concerning maladaptive coping and the interaction of threat and coping appraisals; (b) studies examining additional constructs and variables within tests of the theory, including past behaviour to account for theory sufficiency and implicit and affective variables; (c) studies that adopt intervention or experimental designs that adopt manipulations or techniques targeting change in theory constructs to test theory-aligned mechanisms of action; (d) research examining the temporal stability of theory constructs and their long term prediction; (e) studies that address the dearth of research examining moderators of theory-implied effects, including socio-structural factors and personality traits; and (f) research examining effects of measurement issues that moderate theory effects including conceptualisation of theory constructs, the need to solicit salient beliefs regarding the target behaviour in the population of interest, the essentiality of adopting conditional measures of threat appraisals, and the importance of using non-self-report measures of behaviour. We propose an agenda suggesting research needed to address these limitations, particularly research that adopts designs that expand the scope of the inferences that can be made regarding the theory predictions. We encourage researchers to ensure their theory tests are optimally inclusive through, for example, inclusion of, and control for, past behaviour in theory tests, adoption of longitudinal panel, factorial experimental, and intervention designs that afford more robust causal inferences and tests of mechanisms of action, testing of moderators that may account for observed differences in theory effects dependent on the population and context, and adoption of rigorous standards to systematize and improve measurement of theory constructs. Addressing the limitations and boundary conditions of protection motivation theory Comprehensive tests of the theory Researchers testing the predictive validity of social cognition theories have tended to examine the fit of a proposed network of theory-stipulated relationships specified a priori with correlational crosssectional or longitudinal data from studies in which theory constructs and behavioural outcomes have been measured on one or more occasions (Hagger et al., 2017). Such tests enable confirmation or rejection of the specified pattern of unique relations among theory constructs in a single simultaneous model test, particularly where constructs are likely to be intercorrelated. It also allows for additional analyses to be conducted to examine theory-implied processes, such as mediation effects. In the context of protection motivation theory, however, studies have tended not to conduct full theory tests that encompass all constructs, the proposed relations among them, and the associated mediation and moderation effects. These include the mediating effect of protection motivation on relations between threat and coping appraisals and behavioural outcomes, or potential interaction effects such as the interaction between self-efficacy, an example of a coping appraisal, and threat appraisals on protection motivation. Instead, researchers have tended to confine their analyses to a core subset of theory constructs (i.e., perceived susceptibility, severity, self-efficacy, response efficacy, and response costs) and predictions applied in novel contexts (e.g., COVID-19 preventive behaviours; Okuhara et al., 2020), often neglecting other key constructs (e.g., fear, maladaptive coping responses, maladaptive response rewards; Milne et al., 2000) and failing to test key theory predictions. Or they have focused on examining whether the introduction of a few additional constructs accounts for additional variance in protection motivation or behaviour alongside subsets of HEALTH PSYCHOLOGY REVIEW 7 simultaneously, and a control group would receive neither. This is a similar, but more simplified, example to the study conducted by Wurtele and Maddux (1987) employing the 2 × 2 × 2 × 2 factorial design, where groups of participants systematically received combinations of messages targeting change in four theory constructs. Alongside affording a test of the effect of each technique on the outcome of interest (e.g., protection motivation and health behaviour participation) and the theory-stipulated mechanisms of action through mediation as suggested earlier, such designs would test the extent to which parallel presentation of messages targeting two or more theory constructs led to greater change in the targeted outcomes beyond the effects of each alone. To date the adoption of such designs to test main and interactive effects of protection motivation theory interventions on outcomes, with accompanying mechanism tests, are rare. Researchers should be encouraged to conduct such tests, using designs such as those specified above, to provide more robust evidence not only of causal effects in the theory, but also the mechanisms of action and interactive effects of isolated techniques that target theory constructs. Stability, directionality, and reciprocity in theory effects There is also a need for studies that examine effects of construct stability and directional patterns of prediction in protection motivation theory predictions over time. Studies that adopt longitudinal research designs that measure theory constructs and protection motivation and behavioural outcomes on multiple occasions may provide a potential solution. Pre-eminent among such designs are variants of cross-lagged panel designs, particularly those that have been recently advocated by research methodologists (e.g., Orth et al., 2021; Usami, 2021). Such designs have seldom been utilised in studies testing protection motivation theory predictions. These designs entail administering measures of theory constructs (e.g., threat and coping appraisals, self-efficacy), protection motivation, and behavioural outcomes on multiple occasions (e.g., Lindwall et al., 2011). Such designs would allow researchers to account for the extent of construct change over time attributable to their stability, that will likely affect associations between the constructs and the motivational and behavioural outcomes. In addition, these types of design allow for the better inference of the direction of the proposed effects through the estimation of cross-lagged effects among theory constructs and these outcomes. For example, collecting panel data on protection motivation theory would afford researchers with means to test whether protection motivation predicts subsequent behaviour as proposed, whether behaviour predicts protection motivation over time, or whether both effects are supported, that is, reciprocal effects. Adopting longitudinal panel designs to test protection motivation theory hypotheses would also enable tests of the moderating effects of the stability in intention (or protection motivation) on the intention-behaviour relationship. Intention stability could be calculated using a number of indices, such as the sum of the absolute difference of intention measures at two time points (e.g., Sheeran & Abraham, 2003), and, as such, would not require the use of any additional construct measures. The moderating effects of intention stability is primarily derived from research on other social cognition theories, particularly the theory of planned behaviour. Relatively stable intentions have been found to consistently lead to an increased likelihood of intention translation into subsequent behaviour (Ajzen, 1991) as indicated by larger intention-behaviour effects when intentions are relatively stable (Conner & Godin, 2007; Sheeran & Abraham, 2003). By implication, this moderating effect would also be expected to in protection motivation theory and would indicate protection motivation stability as a key variable of interest for researchers testing theory effects and may provide an additional condition on which protection motivation-behaviour relations depend (Conner & Godin, 2007). In addition, panel designs also provide the opportunity to establish whether the within-occasion effects among theory constructs are stable over time, known as stationarity (Gollob & Reichardt, 1987). For example, this would enable a test of whether the association between threat or coping appraisals and protection motivation remains relatively consistent, or is highly variable, over time. Where temporal variability in a relationship is high, it may be that the stability and cross-lagged 14 J. BALLA AND M. S. HAGGER effects are insufficient in maintaining the relationship between key constructs over time and the effects may eventually reduce to the null, known as entropy (Hagger & Hamilton, 2023; Hertzog & Nesselroade, 1987). However, in the case of protection motivation theory, consistent with longitudinal tests of other social cognition theories for which there is a substantive evidence base comprising longitudinal designs, a truly entropic pattern of effects is unlikely. Maintenance of the withinoccasion effects even when stability and cross-lagged effects are less than perfect is likely due to other extraneous unmeasured constructs that might promote stability. These may include factors in the environment that offer support, such as resource availability or social support, or intraindividual factors, such as individual difference traits like conscientiousness or self-control. Some studies adopting longitudinal panel designs to test protection motivation theory effects have reported reciprocal effects among some of its constructs, while others have found null effects. For example, Dorsey et al. (1999) found reciprocal relations between self-efficacy and a health risk behaviour, drinking alcohol in excess. By contrast, Van der Velde and Van der Pligt (1991) found no evidence for reciprocal relationships among maladaptive coping and threat or coping appraisals for AIDS preventive behaviours. These inconsistencies indicate that further research is needed in order to draw definitive conclusions regarding stability, directionality, and reciprocity in effects across populations, contexts, and behaviours. A systematic review may not be indicated to draw this conclusion, given the relative dearth of studies adopting these types of design to test protection motivation theory predictions, as well as research on the predictions of other social cognition theories stressing this concern (see, for example, Hagger & Hamilton, 2023; Kothe et al., 2019). There is also need for research testing additional temporal processes such as stationarity and entropy. Finally, researchers need to consider the type of panel design that should be utilised to test these proposed effects, such as random-intercept cross-lagged panel designs which, to some extent, superseded standard panel designs by accounting for intraindividual stability in constructs (Usami, 2021). The utility of research adopting panel designs is that they may facilitate future research syntheses that enable tests of these effects across the current literature and their consistency as well as the extent to which they may vary according to key moderating conditions, as has been done recently for other social cognition theories (e.g., Hagger & Hamilton, 2023). Effects of moderator variables Theories such as protection motivation theory, in common with other social cognition theories, assume that their predictions represent generalisable effects that should hold across individuals, behaviours, and contexts. However, researchers have also acknowledged the theory-stipulated effects may vary due to key contextual, individual difference, and environmental factors (see Ajzen, 1991; Balla et al., 2024; Hagger, 2025; Hagger & Hamilton, 2023). This acknowledgement is based on the premise that individuals’ processing of social information prior to making behavioural decisions is dependent on variables that likely determine the relative size or strength of the proposed predicted effects in the theory and reflect potential boundary conditions or auxiliary assumptions that determine when the theory predictions are likely to operate or hold. Moderators do not, therefore, necessarily invalidate theory predictions, but serve to qualify its predictions and the conditions likely to determine their operation, and could be considered ‘auxiliary assumptions’ (see Trafimow, 2012). Testing the effects of moderator variables in theories like protection motivation theory, therefore, is a crucial step to identifying the conditions in which the theory will be most efficacious in describing behaviour. While there has been some attempts to test moderators of effects in applications of protection motivation theory in the health domain (e.g., Plotnikoff & Trinh, 2010; Zhang et al., 2019), these analyses have tended to focus on sample characteristics (e.g., age, gender; Guo et al., 2015; Luo & Mou, 2022; Plotnikoff et al., 2009) with relatively few studies systematically testing effects of other moderators such as individual differences (e.g., psychological traits such as need for cognition), and structural or environmental variables (e.g., access to health-related resources). HEALTH PSYCHOLOGY REVIEW 15 Research examining the moderating effects of these variables on theory effects afford researchers capacity to draw some key conclusions with respect to the generalizability of its effects or whether they are conditional on certain factors. For example, tests of moderators enable evaluation of whether or not the theory predictions hold regardless of the moderator. Such tests also enable identification of conditions that might affect the relative contribution of the different theory constructs to the prediction of protection motivation or behaviour. In addition, the tests may also assist in evaluation of whether intervention effects based on the theory vary across moderator levels, and the extent to which they operate through their nominated mediators, that is, whether the moderator affects the intervention mechanism of action (see also Rothman & Sheeran, 2021). Taken together, such moderator analyses may contribute evidence to support or disconfirm the generalizability of theory effects and identify potential conditions in which researchers may expect theory effects to vary or place bounds on its applicability. To illustrate the importance of systematically examining the effects of moderators of protection motivation theory predictions, we outline two types of moderator variable that have not been widely examined in applications of the theory: socio-structural variables and personality. Researchers have only recently begun to examine how socio-structural variables (e.g., income, education) are implicated in the processes by which belief-based constructs from social cognition theories like protection motivation theory relate to motivational and behavioural outcomes. This represents an important shift away from treating such variables as mere covariates to be ‘factored out’ of analyses testing theory predictions toward examining them as key influences on the mechanisms represented by theory effects. For example, researchers have indicated that average personal or household income, a key indicator of socio-economic status within social structures, moderates relations between social cognition constructs (e.g., perceived behavioural control or self-efficacy, attitudes) and behaviour (e.g., physical activity participation) in tests of social cognition theories in health contexts (Schüz et al., 2020). In addition, socio-structural variables indicating socio-economic status (e.g., household income, employment status) have been identified as moderators of the intention-behaviour relationship in multiple behaviours (e.g., breastfeeding initiation, physical activity; Conner et al., 2013), such that smaller intention-behaviour effect sizes were observed in those with lower income or do not have regular employment. These effects indicate that individuals at the lower end of the socio-economic spectrum are less likely to follow through on their intentions. Low-income groups may have less access to resources (e.g., capacity to afford healthy food) or facilities (e.g., access to gym or sport facilities) that serve as barriers to action despite high motivation. As an illustration of the potential moderating effects of socio-structural variables on the predicted effects between constructs in protection motivation theory, Amireault et al. (2008) found that individuals reporting higher income status exhibited larger positive effects of perceived behavioural control, a construct equivalent to the self-efficacy construct from protection motivation theory, on leisure-time physical activity behaviour. Similarly, Schüz et al. (2020) revealed a positive moderation effect of socio-economic status on the relationship between attitude, a construct parallel the response efficacy and response costs constructs from protection motivation theory as measured in this study, and participation in multiple health behaviours, such that those from high socio-economic status backgrounds more likely to act in accordance with their attitudes. This pattern of effects has been surmised to represent a key auxiliary prediction of the theory. For example, individuals from lower socio-economic backgrounds have limited access to resources or have intensive occupational commitments, which means they will have less access to facilities or may have to work long hours in their vocation, both of which are key barriers to physical activity. Thus, even if they believe in their capability of performing physical activity, or that physical activity will result in certain outcomes, such perceptions are less likely to lead them to perform the behaviour because of the constraints reflected in their social conditions. Findings from these studies suggest that such moderating effects may operate in protection motivation theory, but there is very little research explicitly testing them. There is, therefore, need for researchers to systematically examine the 16 J. BALLA AND M. S. HAGGER potential for such socio-structural variables to moderate theory effects by including relevant measures of these variables alongside the theory constructs when testing the theory. Similarly, personality constructs and other traits have been examined as candidate moderators of effects of protection motivation theory constructs on motivational and behavioural outcomes. Such moderators test a generalised premise that the extent to which individuals are motivated to act in accordance with their appraisals and beliefs is conditional on intrapersonal conditions that may affect their decision making (for similar examples see De Bruijn et al., 2009; Rhodes, Courneya, & Hayduk, 2002). As an example in the context of protection motivation theory, uncertainty orientation, a construct that reflects how individuals typically represent and respond to uncertainty in their environment, has been found to moderate the relationship between perceived threat and acceptance of disease screening and disease information seeking (Brouwers & Sorrentino, 1993). Uncertainty-oriented individuals, who tend to be motivated to resolve uncertainty, may be more likely to seek out information regarding diagnosis of a disease under conditions of both high appraisals of threat and high appraisals of ability to cope, in line with expectations outlined in protection motivation theory. Certainty-oriented individuals, who tend to be more likely to act in situations in which uncertainty is low, however, may be motivated to avoid information regarding their diagnosis of a disease under these same circumstances. Counter to theory hypotheses, even if such individuals believe they have high personal capacity to receive screening for a disease, and that a disease would be personally threatening to their well-being, these beliefs would be less likely to lead to higher intentions or likelihood of enacting the behaviour due to the inherent uncertainty of the diagnosis and the potential threat of this diagnosis to their health. These results are empirically demonstrated and discussed further in Brouwers and Sorrentino (1993). A number of other personality traits also show promise as moderators in social cognition theories, such as conscientiousness or trait self-control, which may moderate the relationships between appraisals and health behaviour performance as well as the relationship between intention and behaviour (e.g., Conner & Abraham, 2001; Rhodes, Courneya, & Jones, 2002). However, findings of these effects are less equivocal for personality. For example, Rhodes et al.’s (2022) systematic review of moderators of the intention-physical activity behaviour relationship describe a consistent moderation effect for conscientiousness, but not for other traits such as agreeableness and openness to experience. It is clear that research examining effects of these moderators on protection motivation theory effects is in its infancy, and researchers are encouraged to consider testing such moderators to better capture the intrapersonal conditions on which theory predictions depend and provide a larger evidence base of these effects. Beyond providing important knowledge on the conditional basis of theory predictions, such evidence may also provide indication of which segments of the population that may be more responsive or receptive to certain communications and interventions aimed at changing behaviour by changing theory beliefs and where such interventions may be less efficacious without accounting for pervading intrapersonal conditions. Measurement issues A final concern in research testing protection motivation theory hypotheses is the lack of consistency in the measurement and operationalisation of theory constructs, an issue which has also been highlighted in tests of other social cognition theories (e.g., Hagger et al., 2020b). Specifically, the items selected to measure theory variables may impact observed effects in studies testing theory predictions. For example, effects of coping appraisals on protection motivation and behaviour in the theory may vary when the items used focus on outcomes that may not be salient to the individual. As an illustration, Wulfert and Wan (1993) observed larger effects of response efficacy on condom use when the items used to measure the construct focused on pregnancy prevention rather than disease prevention in a sample of undergraduate students. Similarly, measures that do not have appropriate correspondence with outcome measures such as protection motivation or behaviour may also influence theory effects (Ajzen, 1991). For example, Courneya (1994) found that the HEALTH PSYCHOLOGY REVIEW 17 effects of measures of intention, a construct synonymous with protection motivation, on physical activity behaviour were larger when the scale format (e.g., dichotomous or continuous response scales) of the intention and behaviour measures corresponded closely. Alongside this, researchers may adopt measures that vary substantially in terms of item content but label those measures similarly as the same construct. For example, measures of maladaptive response rewards have referenced either intrinsic or extrinsic rewards, or both (e.g., MacDonell, 2013), or measures of financial and monetary benefits of not performing the target behaviour (e.g., Eberhardt & Ling, 2021). Such issues have potential to introduce method variance in studies testing theory effects and present problems to researchers when making comparisons across studies. This is also likely to introduce additional variability to the averaged sizes of theory effects in quantitative syntheses of research, such as meta-analyses. It is therefore important that researchers seeking to compare theory effects take measurement issues into account when doing so and consider the consequences that such measurement variation could have when testing the theory and in drawing inferences. One means to allay the increased error variance would be to develop and apply a formal classification procedure that provides standardised definitions of theory constructs and prototype measures ensure measures use conform. An illustration is provided in guidelines provided for the development of measures of constructs of the theories of reasoned action and planned behaviour. Such guidelines provide standardised definitions and prototype measures of theory constructs to minimise variability measure content and operationalisation in research testing the theory (e.g., Ajzen, 1985, 2002; Ajzen & Fishbein, 1980). Others have adopted a ‘meta-theoretical’ approach by classifying constructs across theories into a sets of constructs with standardised labels that capture constructs typically included the theories (e.g., Cane et al., 2012; McMillan & Conner, 2007; Peters & Crutzen, 2022). The virtue of this approach has been summarised by researchers conducting meta-analyses of studies testing the effects of these theories in health contexts, observing that many studies have followed the stipulated guidelines thus minimising the observed variability in the measures used across studies (Hagger & Chatzisarantis, 2009; Starfelt Sutton & White, 2016). It is also important that measures are appropriately tailored to make explicit reference to the specific risks or health threats and protection behaviours relevant to the target population, that is, there is appropriate correspondence between measures of theory constructs and outcomes. This has been a recognised feature of other social cognition theories and has been incorporated as a key measurement principle in guidelines for developing theory construct measures (e.g., Ajzen, 1991, 2002). This practice ensures appropriate specificity of the measures with respect to the target outcomes and ensures they are appropriate for the target population, which is expected to improve precision in predictive tests of theory hypotheses. It also informs intervention such that persuasive communications based on the theory target relevant outcomes of relevance to the target population. One means to promote correspondence is to adopt open-ended formative research to elicit the salient outcomes and risks from groups of participants from the target population and use the outcomes and risk identified to develop construct measures (Ajzen, 1991; Francis et al., 2004; Hamilton & White, 2010). Such research typically involves using an open-ended survey or pilot interview research in which participants are prompted to list the relevant outcomes or risks with respect to the health threat or protective behaviour (e.g., Norman et al., 2015). These are then content analysed for the most frequently occurring outcomes and risks, and used in the development of measurement items (for further details of this approach see Ajzen, 2002). While this practice has been frequently, albeit not universally, used to develop corresponding measures for social cognition theories like the theory of planned behaviour, this has not been the norm for research on protection motivation theory. That said, there is research that has provided precedent for this practice to be modelled in future research applying the theory. For example, when evaluating their child’s perceived vulnerability of worsening vision impairment over time with compliance with treatment for amblyopia, Searle et al. (2000) used a qualitative interview pilot study to elicit parents’ reported beliefs with 18 J. BALLA AND M. S. HAGGER respect to this condition. Content analysis of the interviews revealed specific beliefs with respect to the uncertainty surrounding the risks of the outcome (child’s worsening vision) and the importance of focusing on adherence to treatment. However, such pilot work is not common practice in research on protection motivation theory and is strongly advocated in the early formation stages of study measures to promote less error variability and greater measurement precision in research testing the theory. A related ongoing measurement issue is the need for conditional measures of threat appraisals when testing theory predictions. Conditional measures entail prompting members of the target population to evaluate the health risks of failing to perform the target protection behaviour, the consequences of performing the target behaviour, and the health threat of the illness or condition taking recent and current behaviour into account (e.g., ‘Considering that you are fully vaccinated against COVID-19, how susceptible do you think you are to contracting COVID-19 in the future?’). This is an issue that has been recognised in previous research applying the theory, but studies have tended not to implement these types of measure when applying the theory (Norman et al., 2015). Instead, researchers have tended to adopt generic measures of constructs that do not make reference to salient health risk and consequences or account for current behaviour. Use of conditional measures of threat appraisals may be particularly useful in capturing perceived vulnerability, as opposed to measures that do not make reference to the relevant personal health threat and past performance of the protection behaviour (e.g., ‘How susceptible do you think you are to contracting COVID-19 in the future?’), by minimising individuals’ tendency to under – or over-estimate their risk (Garcia et al., 2018). Use of measures of this type may help explain some of the inconsistent findings observed in theory tests, such as why some studies have observed positive relationships between threat appraisals and protection motivation or behaviour, while others have observed negative relationships. Measures that specify the health threat against which participants judge their risk in the context of prior risk and behaviour may account for this variability in risk appraisal effects across participants. Finally, an over-reliance on self-report behaviour measures has been a long-standing concern and acknowledged limitation of research testing the predictions of social cognition theories, including protection motivation theory (Hessing et al., 1988). Although self-reported behaviour measures have been shown to correlate with non-self-report measures lending some support for their concurrent validity, the correlations tend to be modest in size (Dang et al., 2020). Numerous limitations may be responsible including poor recall, acquiescence bias, demand characteristics, and low reliability (Norman et al., 2015; Sheeran et al., 2017). To illustrate, meta-analyses of the theory of planned behaviour have found larger effects of the theory constructs on behaviour in studies adopting self-reported behavioural measures when compared with those that employed non-self-report measures (e.g., observation devices such as accelerometers for physical activity or micro-electromechanical systems (MEMS) for medication adherence; Armitage & Conner, 2001; McEachan et al., 2011). This implies that the additional error variance introduced in effect size estimates in studies employing self-report behavioural measures may be inflating theory effects. Non-self report behaviour measures have been suggested to minimise these biases such as use of surveillance cameras or trained coders to observe behaviour (e.g., Lapinski et al., 2013), accelerometers and other wearable devices to measure physical activity (e.g., Gaston & Prapavessis, 2014; Wou et al., 2018) or sleep (e.g., Hall et al., 2018), and MEMS devices on pill bottles for measuring medication adherence (e.g., Wu et al., 2008). Of course, no behavioural measure can be truly considered as providing ‘objective’ or unbiased behavioural assessment, as all are associated with some degree of bias. For example, behavioural observation is limited by the attention and coding of the observer, and the use of wearable and MEMS devices have limitations that include data loss and potential to affect the behaviour of the participant in themselves (e.g., Finkelstein et al., 2016). A further downside is that such measures can be costly and thus impractical for use in some studies. That said, such measures are typically associated with fewer of the biases associated with self-report measures. As a consequence, adoption of non-self-report measures should be considered best practice in theory tests to provide HEALTH PSYCHOLOGY REVIEW 19 greater precision in measurement and minimise measurement error and, when this is not possible, concurrent validity data on the self-report measures adopted with non-self-report measures should ideally be reported. Summary and conclusion Promoting performance of health-enhancing behaviours and the reduction of health-risk behaviours through intervention is a recognised strategy for the prevention of chronic disease development and associated maladaptive health outcomes. Behavioural interventions should be informed by theories derived from the behavioural sciences, particularly psychology, to be optimally effective in changing health-enhancing behaviours. Protection motivation theory is a prominent theory that has been employed to identify the determinants of health and health risk behaviour that might be modified through intervention. While the theory predictions have been generally supported in predictive and experimental research there are outstanding issues that hinder its utility in identifying the determinants of behaviour and the mechanisms that underpin behaviour change. In this review we aimed to highlight these issues and outline potential solutions that serve as an agenda for future research on the theory. The key issues highlighted were: a lack of comprehensive tests of the theory and associations among its constructs; failure to routinely account for past behaviour and test theory sufficiency; a relative absence of wide-spread integration of additional constructs in theory tests; a lack of research adopting of study designs that test longitudinal effects, interactions, and mechanisms of action within the theory; a paucity of tests of moderators that may account for observed variation in theory effects across studies; and adoption of measurement practices that hinder precision of theory tests (e.g., low measurement correspondence, use of self-report behaviour measures). To account for these evidence gaps, we propose that future research applying the theory in health behaviour contexts should attempt to conduct rigorous tests of the theory encompassing most or, preferably, all theory constructs with an appropriate conceptual rationale for the selection, testing their direct and interactive effects, and including past behaviour to test theory sufficiency and as a potential moderator of the appraisal-protection motivation relationship. Future studies should also consider including other constructs that represent associated processes that may extend the scope of theory predictions. In addition, there is a need for studies testing causal and longitudinal effects in the theory consistent with its predictions (e.g., the effects of appraisals on protection motivation and behaviour), and mechanisms of action of interventions based on the theory. For example, we recommend the use of factorial experimental designs in which multiple theory constructs such as appraisals are manipulated along and simultaneously to examine their direct and interactive effects on behaviour. Further, we advocate adoption of the rigorous, recently advocated forms of longitudinal panel design to account for stability and assess directional and cross-lagged theory effects. Researchers should also consider evaluating effects of moderators such as individual traits as well as socio-structural variables in theory tests. Finally, standards for the measurement of protection motivation theory variables should be implemented, such as advocating for the adoption of conditional and specific measures of appraisals, widespread use of standard classifications for the theory variables, and widespread adoption non-self-report measures of behaviour. Note 1. Although this is a feature of studies applying protection motivation theory, the same is true for research testing social cognition theory predictions more broadly. For example, research examining the theory of planned behavior in health behavior contexts has largely neglected testing the proposed moderating effects of perceived behavioral control on the effects of attitudes and subjective norms on intentions, and the effect of intention on behavior (see Hagger et al., 2022; La Barbera & Ajzen, 2021). Disclosure statement No potential conflict of interest was reported by the author(s). 20 J. BALLA AND M. S. HAGGER References Ajzen, I. (1985). From intentions to actions: A theory of planned behavior. In J. Kuhl & J. Beckmann (Eds.), Action Control (pp. 11–39). Springer. Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179–211. https://doi.org/10.1016/0749-5978(91)90020-T Ajzen, I. (2002). Constructing a TPB questionnaire: Conceptual and methodological considerations. http://people.umass. edu/~aizen/pdf/tpb.measurement.pdf. Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social behavior. Prentice-Hall. Amireault, S., Godin, G., Vohl, M.-C., & Pérusse, L. (2008). Moderators of the intention-behaviour and perceived behavioural control-behaviour relationships for leisure-time physical activity. International Journal of Behavioral Nutrition and Physical Activity, 5(1), 7. https://doi.org/10.1186/1479-5868-5-7 Armitage, C. J., & Conner, M. (2000). Social cognition models and health behaviour: A structured review. Psychology & Health, 15(2), 173–189. https://doi.org/10.1080/08870440008400299 Armitage, C. J., & Conner, M. (2001). Efficacy of the theory of planned behaviour: A meta-analytic review. British Journal of Social Psychology, 40(4), 471–499. https://doi.org/10.1348/014466601164939 Babazadeh, T., Nadrian, H., Banayejeddi, M., & Rezapour, B. (2017). Determinants of skin cancer preventive behaviors among rural farmers in Iran: An application of protection motivation theory. Journal of Cancer Education, 32(3), 604–612. https://doi.org/10.1007/s13187-016-1004-7 Bagozzi, R. P. (1981). Attitudes, intentions, and behavior: A test of some key hypotheses. Journal of Personality and Social Psychology, 41(4), 607–627. https://doi.org/10.1037/0022-3514.41.4.607 Balla, J., Polet, J., Kokko, S., Hirvensalo, M., Vasankari, T., Lintunen, T., & Hagger, M. S. (2024). Predicting adolescents’ physical activity intentions: Testing an integrated social cognition model. International Journal of Behavioral Medicine, 31 (1), 41–54. http://doi.org/10.1007/s12529-023-10156-3 Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Prentice-Hall, Inc. Bishop, F. L., Fenge-Davies, A. L., Kirby, S., & Geraghty, A. W. A. (2015). Context effects and behaviour change techniques in randomised trials: A systematic review using the example of trials to increase adherence to physical activity in musculoskeletal pain. Psychology & Health, 30(1), 104–121. https://doi.org/10.1080/08870446. 2014.953529 Brouwers, M. C., & Sorrentino, R. M. (1993). Uncertainty orientation and protection motivation theory: The role of individual differences in health compliance. Journal of Personality and Social Psychology, 65(1), 102–112. https://doi.org/ 10.1037/0022-3514.65.1.102 Brown, S. P., Westbrook, R. A., & Challagalla, G. (2005). Good cope, bad cope: Adaptive and maladaptive coping strategies following a critical negative work event. Journal of Applied Psychology, 90(4), 792–798. https://doi.org/10.1037/ 0021-9010.90.4.792 Bruni, L., Serrano, B., Roura, E., Alemany, L., Cowan, M., Herrero, R., Poljak, M., Murillo, R., Broutet, N., Riley, L. M., & De Sanjose, S. (2022). Cervical cancer screening programmes and age-specific coverage estimates for 202 countries and territories worldwide: A review and synthetic analysis. The Lancet Global Health, 10(8), e1115–e1127. https:// doi.org/10.1016/S2214-109X(22)00241-8 Bryan, A. D., Aiken, L. S., & West, S. G. (1997). Young women’s condom use: The influence of acceptance of sexuality, control over the sexual encounter, and perceived susceptibility to common STDs. Health Psychology, 16(5), 468– 479. https://doi.org/10.1037/0278-6133.16.5.468 Cane, J., O’Connor, D., & Michie, S. (2012). Validation of the theoretical domains framework for use in behaviour change and implementation research. Implementation Science, 7(1), 37. https://doi.org/10.1186/1748-5908-7-37 Carey, R. N., Connell, L. E., Johnston, M., Rothman, A. J., De Bruin, M., Kelly, M. P., & Michie, S. (2018). Behavior change techniques and their mechanisms of action: A synthesis of links described in published intervention literature. Annals of Behavioral Medicine, 53(8), 693–707. https://doi.org/10.1093/abm/kay078 Carver, C. S., Scheier, M. F., & Weintraub, J. K. (1989). Assessing coping strategies: A theoretically based approach. Journal of Personality and Social Psychology, 56(2), 267–283. https://doi.org/10.1037/0022-3514.56.2.267 Centers for Disease Control and Prevention. (2021a). Data and statistics—Smoking and tobacco use. https://www.cdc. gov/tobacco/data_statistics/index.htm. Centers for Disease Control and Prevention. (2021b). FastStats—Exercise or physical activity. https://www.cdc.gov/nchs/ fastats/exercise.htm. Centers for Disease Control and Prevention. (2022). Promoting health for adults. https://www.cdc.gov/chronicdisease/ resources/publications/factsheets/promoting-health-for-adults.htm. Chamroonsawasdi, K., Chottanapund, S., Pamungkas, R. A., Tunyasitthisundhorn, P., Sornpaisarn, B., & Numpaisan, O. (2021). Protection motivation theory to predict intention of healthy eating and sufficient physical activity to prevent diabetes mellitus in Thai population: A path analysis. Diabetes & Metabolic Syndrome: Clinical Research & Reviews, 15(1), 121–127. https://doi.org/10.1016/j.dsx.2020.12.017 Cismaru, M., & Lavack, A. M. (2007). Interaction effects and combinatorial rules governing Protection Motivation Theory variables: A new model. Marketing Theory, 7(3), 249–270. https://doi.org/10.1177/1470593107080344 HEALTH PSYCHOLOGY REVIEW 21 Clubb, A. C., & Hinkle, J. C. (2015). Protection motivation theory as a theoretical framework for understanding the use of protective measures. Criminal Justice Studies, 28(3), 336–355. https://doi.org/10.1080/1478601X.2015.1050590 Connell, L. E., Carey, R. N., De Bruin, M., Rothman, A. J., Johnston, M., Kelly, M. P., & Michie, S. (2019). Links between behavior change techniques and mechanisms of action: An expert consensus study. Annals of Behavioral Medicine, 53(8), 708–720. https://doi.org/10.1093/abm/kay082 Conner, M., & Abraham, C. (2001). Conscientiousness and the theory of planned behavior: Toward a more complete model of the antecedents of intentions and behavior. Personality and Social Psychology Bulletin, 27(11), 1547– 1561. https://doi.org/10.1177/01461672012711014 Conner, M., & Godin, G. (2007). Temporal stability of behavioural intention as a moderator of intention–health behaviour relationships. Psychology & Health, 22(8), 875–897. https://doi.org/10.1080/14768320601070449 Conner, M., McEachan, R., Jackson, C., McMillan, B., Woolridge, M., & Lawton, R. (2013). Moderating effect of socioeconomic status on the relationship between health cognitions and behaviors. Annals of Behavioral Medicine, 46(1), 19– 30. https://doi.org/10.1007/s12160-013-9481-y Conner, M., & Norman, P. (2015). Predicting and changing health behaviour: A social cognition approach. In M. Conner & P. Norman (Eds.), Predicting and changing health behaviour: Research and practice with social cognition models (3rd ed., pp. 1–29). McGraw-Hill Education. Courneya, K. S. (1994). Predicting repeated behavior from intention: The issue of scale correspondence. Journal of Applied Social Psychology, 24(7), 580–594. https://doi.org/10.1111/j.1559-1816.1994.tb00601.x Courneya, K. S., & Hellsten, L.-A. M. (2001). Cancer prevention as a source of exercise motivation: An experimental test using protection motivation theory. Psychology, Health & Medicine, 6(1), 59–64. https://doi.org/10.1080/ 13548500125267 Dang, J., King, K. M., & Inzlicht, M. (2020). Why are self-report and behavioral measures weakly correlated? Trends in Cognitive Sciences, 24(4), 267–269. https://doi.org/10.1016/j.tics.2020.01.007 De Bruijn, G.-J., Brug, J., & Van Lenthe, F. J. (2009). Neuroticism, conscientiousness and fruit consumption: Exploring mediator and moderator effects in the theory of planned behaviour. Psychology & Health, 24(9), 1051–1069. https://doi.org/10.1080/08870440802428241 Dodge, C. E., Fisk, N., Burruss, G. W., Moule Jr, R. K., & Jaynes, C. M. (2023). What motivates users to adopt cybersecurity practices? A survey experiment assessing protection motivation theory. Criminology & Public Policy, 22(4), 849–868. https://doi.org/10.1111/1745-9133.12641 Dombrowski, S. U., O’Carroll, R. E., & Williams, B. (2016). Form of delivery as a key ‘active ingredient’ in behaviour change interventions. British Journal of Health Psychology, 21(4), 733–740. https://doi.org/10.1111/bjhp.12203 Dorsey, A. M., Miller, K. I., & Scherer, C. W. (1999). Communication, risk behavior, and perceptions of threat and efficacy: A test of a reciprocal model. Journal of Applied Communication Research, 27(4), 377–395. https://doi.org/10.1080/ 00909889909365546 Eberhardt, J., & Ling, J. (2021). Predicting COVID-19 vaccination intention using protection motivation theory and conspiracy beliefs. Vaccine, 39(42), 6269–6275. https://doi.org/10.1016/j.vaccine.2021.09.010 Finkelstein, E. A., Haaland, B. A., Bilger, M., Sahasranaman, A., Sloan, R. A., Nang, E. E. K., & Evenson, K. R. (2016). Effectiveness of activity trackers with and without incentives to increase physical activity (TRIPPA): A randomised controlled trial. The Lancet Diabetes & Endocrinology, 4(12), 983–995. https://doi.org/10.1016/S2213-8587 (16)30284-4 Fishbein, M., & Ajzen, I. (2011). Predicting and changing behavior (0 ed.). Psychology Press. Fishbein, M., Triandis, H. C., Kanfer, F. H., Becker, M., Middlestadt, S. E., & Eichler, A. (2001). Factors influencing behavior and behavior change. In A. Baum, T. A. Revenson, & J. E. Singer (Eds.), Handbook of health psychology (pp. 3–17). Lawrence Erlbaum Associates. Floyd, D. L., Prentice-Dunn, S., & Rogers, R. W. (2000). A meta-analysis of research on protection motivation theory. Journal of Applied Social Psychology, 30(2), 407–429. https://doi.org/10.1111/j.1559-1816.2000.tb02323.x Ford, E. S., Zhao, G., Tsai, J., & Li, C. (2011). Low-risk lifestyle behaviors and all-cause mortality: Findings from the national health and nutrition examination survey III mortality study. American Journal of Public Health, 101(10), 1922–1929. https://doi.org/10.2105/AJPH.2011.300167 Francis, J. J., Eccles, M. P., Johnston, M., Walker, A. E., Grimshaw, J. M., Foy, R., Kaner, E. F. S., Smith, L., & Bonetti, D. (2004). Constructing questionnaires based on the theory of planned behaviour: A manual for health services researchers. Centre for Health Services Research, University of Newcastle. Fruin, D. J., Pratt, C., & Owen, N. (1992). Protection motivation theory and adolescents’ perceptions of exercise. Journal of Applied Social Psychology, 22(1), 55–69. https://doi.org/10.1111/j.1559-1816.1992.tb01521.x Garcia, T. A., Fairlie, A. M., Litt, D. M., Waldron, K. A., & Lewis, M. A. (2018). Perceived vulnerability moderates the relations between the use of protective behavioral strategies and alcohol use and consequences among high-risk young adults. Addictive Behaviors, 81, 150–156. https://doi.org/10.1016/j.addbeh.2018.02.001 Gaston, A., & Prapavessis, H. (2014). Using a combined protection motivation theory and health action process approach intervention to promote exercise during pregnancy. Journal of Behavioral Medicine, 37(2), 173–184. https://doi.org/ 10.1007/s10865-012-9477-2 22 J. BALLA AND M. S. HAGGER Gollob, H. F., & Reichardt, C. S. (1987). Taking account of time lags in causal models. Child Development, 58(1), 80. https:// doi.org/10.2307/1130293 Gong, J., Stanton, B., Lunn, S., Deveaux, L., Li, X., Marshall, S., Brathwaite, N. V., Cottrell, L., Harris, C., & Chen, X. (2009). Effects through 24 months of an HIV/AIDS prevention intervention program based on protection motivation theory among preadolescents in the Bahamas. Pediatrics, 123(5), e917–e928. https://doi.org/10.1542/peds.2008-2363 Grindley, E. J., Zizzi, S. J., & Nasypany, A. M. (2008). Use of protection motivation theory, affect, and barriers to understand and predict adherence to outpatient rehabilitation. Physical Therapy, 88(12), 1529–1540. https://doi.org/10. 2522/ptj.20070076 Guo, X., Han, X., Zhang, X., Dang, Y., & Chen, C. (2015). Investigating m-Health acceptance from a protection motivation theory perspective: Gender and age differences. Telemedicine and E-Health, 21(8), 661–669. https://doi.org/10.1089/ tmj.2014.0166 Hagger, M. S. (2019). The reasoned action approach and the theories of reasoned action and planned behavior. Oxford Bibliographies in Psychology. https://doi.org/10.1093/OBO/9780199828340-0240 Hagger, M. S. (2025). Psychological determinants of health behavior. Annual Review of Psychology. https://doi.org/10. 1146/annurev-psych-020124-114222 Hagger, M. S., Cameron, L. D., Hamilton, K., Hankonen, N., & Lintunen, T. (Eds.). (2020a). The handbook of behavior change. Cambridge University Press. Hagger, M. S., Cameron, L. D., Hamilton, K., Hankonen, N., & Lintunen, T. (2020b). The science of behavior change: The road ahead. In M. S. Hagger, L. D. Cameron, K. Hamilton, N. Hankonen, & T. Lintunen (Eds.), The handbook of behavior change (pp. 677–699). Cambridge University Press. Hagger, M. S., & Chatzisarantis, N. L. D. (2009). Integrating the theory of planned behaviour and self-determination theory in health behaviour: A meta-analysis. British Journal of Health Psychology, 14(2), 275–302. https://doi.org/ 10.1348/135910708X373959 Hagger, M. S., Chatzisarantis, N. L. D., & Biddle, S. J. H. (2002). A meta-analytic review of the theories of reasoned action and planned behavior in physical activity: Predictive validity and the contribution of additional variables. Journal of Sport and Exercise Psychology, 24(1), 3–32. https://doi.org/10.1123/jsep.24.1.3 Hagger, M. S., Cheung, M. W.-L., Ajzen, I., & Hamilton, K. (2022). Perceived behavioral control moderating effects in the theory of planned behavior: A meta-analysis. Health Psychology, 41(2), 155–167. https://doi.org/10.1037/hea0001153 Hagger, M. S., Gucciardi, D. F., & Chatzisarantis, N. L. D. (2017). On nomological validity and auxiliary assumptions: The importance of simultaneously testing effects in social cognitive theories applied to health behavior and some guidelines. Frontiers in Psychology, 8, 1933. https://doi.org/10.3389/fpsyg.2017.01933 Hagger, M. S., & Hamilton, K. (2023). Longitudinal tests of the theory of planned behaviour: A meta-analysis. European Review of Social Psychology, 35(1), 198–254. https://doi.org/10.1080/10463283.2023.2225897 Hagger, M. S., Hamilton, K., Phipps, D. J., Protogerou, C., Zhang, C.-Q., Girelli, L., Mallia, L., & Lucidi, F. (2023). Effects of habit and intention on behavior: Meta-analysis and test of key moderators. Motivation Science, 9(2), 73–94. https:// doi.org/10.1037/mot0000294 Hagger, M. S., & Hardcastle, S. J. (2014). Interpersonal style should be included in taxonomies of behavior change techniques. Frontiers in Psychology, 5, 254. https://doi.org/10.3389/fpsyg.2014.00254 Hagger, M. S., Moyers, S., McAnally, K., & McKinley, L. E. (2020). Known knowns and known unknowns on behavior change interventions and mechanisms of action. Health Psychology Review, 14(1), 199–212. https://doi.org/10. 1080/17437199.2020.1719184 Hagger, M. S., & Orbell, S. (2022). The common sense model of illness self-regulation: A conceptual review and proposed extended model. Health Psychology Review, 16(3), 347–377. https://doi.org/10.1080/17437199.2021.1878050 Hagger, M. S., Polet, J., & Lintunen, T. (2018). The reasoned action approach applied to health behavior: Role of past behavior and tests of some key moderators using meta-analytic structural equation modeling. Social Science & Medicine, 213, 85–94. https://doi.org/10.1016/j.socscimed.2018.07.038 Hall, M. H., Brindle, R. C., & Buysse, D. J. (2018). Sleep and cardiovascular disease: Emerging opportunities for psychology. American Psychologist, 73(8), 994–1006. https://doi.org/10.1037/amp0000362 Hamilton, K., Van Dongen, A., & Hagger, M. S. (2020). An extended theory of planned behavior for parent-for-child health behaviors: A meta-analysis. Health Psychology, 39(10), 863–878. https://doi.org/10.1037/hea0000940 Hamilton, K., & White, K. M. (2010). Identifying parents’ perceptions about physical activity: A qualitative exploration of salient behavioural, normative and control beliefs among mothers and fathers of young children. Journal of Health Psychology, 15(8), 1157–1169. https://doi.org/10.1177/1359105310364176 Haug, E., Rasmussen, M., Samdal, O., Iannotti, R., Kelly, C., Borraccino, A., Vereecken, C., Melkevik, O., Lazzeri, G., Giacchi, M., Ercan, O., Due, P., Ravens-Sieberer, U., Currie, C., Morgan, A., & Ahluwalia, N. (2009). Overweight in school-aged children and its relationship with demographic and lifestyle factors: Results from the WHO-Collaborative Health Behaviour in School-aged Children (HBSC) Study. International Journal of Public Health, 54(S2), 167–179. https:// doi.org/10.1007/s00038-009-5408-6 Helmes, A. W. (2002). Application of the protection motivation theory to genetic testing for breast cancer risk. Preventive Medicine, 35(5), 453–462. https://doi.org/10.1006/pmed.2002.1110 HEALTH PSYCHOLOGY REVIEW 23