Are you cooking your meat enough? The efficacy of the Theory of Planned Behavior in predicting a best practice to prevent salmonellosis

https://doi.org/10.1016/j.foodres.2011.06.028Get rights and content

Abstract

Salmonella is one of the most common pathogens in the European Union which can spread in the domestic environment due to incorrect food handling practices. This study has investigated the efficacy of the Theory of Planned Behavior (TPB) to uncover the processes that lead to the formation of the intentions and thus to a specific behavior, i.e. consuming meat and fresh sausages only if well cooked. Two parallel longitudinal studies comprising Italian young adults and adults were carried out. In a two week period, participants were administered two questionnaires measuring behavioral intentions and their antecedents in the first wave, and self-reported behavior toward the target behavior in the second one.

Results showed the superior predictive power of the TPB plus past behavior. From a cognitive perspective, the samples differed: the actual behavior of young adults derives from intentions and perceived control, whereas the behavior of adults depends only on past behavior. Findings suggest that food risk-communication should consider the differences in behavior models.

Highlights

► A best practice to prevent salmonellosis is consuming meat only if well cooked. ► TPB is used to reveal processes leading to intentions’ and behavior’s formation. ► Two longitudinal studies with Italian young adults and adults were carried out. ► Results showed the superior predictive power of the TPB plus past behavior. ► Food risk-communication should consider cognitive differences emerged in samples.

Introduction

Food safety management is a major public health concern as it may be responsible for foodborne illnesses (World Health Organization, 2002). Salmonella, in particular, is one of the most common pathogens in the European Union (Pangloli et al., 2008). Although its diffusion has been steadily decreasing, in 2008 salmonellosis remained, behind campylobacteriosis, the second most commonly reported human zoonoses, accounting for 131,468 confirmed human cases in 27 EU countries, with a notification rate of 26.4 per 100,000 inhabitants (European Food Safety Authority [EFSA], 2010). In Italy, in 2008, 6728 cases of salmonellosis were reported, contracted especially by males, aged 0–14. In the same year, 2411 Italians were generally infected by foodborne diseases (Ministero Italiano della Salute, 2011).

People infected by Salmonella may develop symptoms such as diarrhea, fever, abdominal cramps, headache and emesis (Baird-Parker, 1990). The clinical problems that ensue for the individuals (Helms, Vastrup, Gerner-Smidt, & Molbak, 2003) and the consequential economic burden (Altekruse et al., 1997, Lopalco et al., 2000) are relevant issues for society.

Humans can contract foodborne diseases from a number of sources, including a wide range of domestic and wild animals, as well as a variety of foodstuffs of both animal and plant origin. Food contamination by Salmonella can occur along the entire food chain, from the farm to the domestic environment. Transmission in fact, often occurs through direct contact with infected animals, primarily those in rural farms, or during food preparation, because of inadequate storage temperatures, cross-contamination of ready-to-eat foods or insufficient cooking (European Food Safety Authority, 2010, Parry et al., 2004, Scott, 2003). In 2008, eggs and egg products were reported to be the main food vehicle of salmonellosis outbreaks in EU, but this pathogen was often also detected in fresh chicken, turkey and pork meat (EFSA, 2010). Specifically, some recent research has recognized that undercooked meat can result in a higher concentration of pathogens – Salmonella included – in food before consumption. As a consequence, it is important to avoid eating raw meats and one of the best ways to prevent this risk is to eat well cooked meats and sausages (e.g., Busani et al., 2009, Cibin et al., 2009). Additionally, literature highlights that people generally give little weight to the cooking process which instead ought to be regarded as an important factor of prevention (Arzenton et al., 2009). Thus, this behavior deserves greater attention.

Especially in the domestic environment, wrong beliefs on the storage, the handling and the preparation of food can expose consumers to the risk of foodborne diseases (Desmarchelier, 1996, Milton and Mullan, 2010, Motarjemi and Käferstein, 1997, Redmond and Griffith, 2003). So far, research has investigated consumer perceptions and behavior in relation to foodborne illnesses, focusing both on the way people perceive and manage food risks and on the correspondence between risk perception, the knowledge of food risks, and the practices carried out at home (Abbot et al., 2009, Jay et al., 1999, Jevšnik et al., 2008, Sanlier, 2009, Worsfold and Griffith, 1997). Actually, although food hygiene precautions in domestic kitchens are regarded as an important means of prevention (Milton and Mullan, 2010, Redmond and Griffith, 2003), recent research has shown that consumers may behave completely differently and unsafely. Even if they claim to know about food safety behavior, they do not put it into practice (McCarthy et al., 2007, Mullan and Wong, 2009, Wilcock et al., 2004). Moreover, people who have experienced food poisoning behave no differently than those who have never contracted a foodborne disease (Parry et al., 2004).

The above reasons highlight the need to fully understand the cognitive factors that underlie domestic food practices. When striving to correct wrong cognitions and behaviors by means of educational and risk communication campaigns, it is important to investigate how intentions on best practices develop (Jacob et al., 2010, Redmond and Griffith, 2003). In fact, communication campaigns promoting safer food handling practices, which assess how much the targeted group knows about microbial pathogens, and if they feel at risk, have proven to be effective and beneficial in reducing the incidence of foodborne diseases, including Salmonella (Arzenton et al., 2005, Tiozzo et al., 2011).

The theoretical framework adopted herein is the Theory of Planned Behavior (TPB; Ajzen, 1991, Ajzen and Madden, 1986). It represents one of the most important social cognition models in the health psychology field and can be considered a basis for understanding the determinants of behavior. Any intervention whose goal is to improve public health should be based on these principles (Abraham et al., 2008, Conner and Sparks, 2005). TPB posits that the proximal predictor of volitional behavior is behavioral intention (for a formal description of the theory, see e.g. Fig. 1 in the Results section). The latter construct represents the motivation to act: a conscious plan to exert efforts to carry out a behavior. As a general rule, the stronger the intention to engage in a behavior, the more likely its performance will be. Another direct influence on behavior is the perception of behavioral control (PBC). Although the motivation to carry out the action is present, the performance may depend, at least to some degree, on the availability of opportunities and resources. These factors represent the individual's control over the behavior. Nevertheless, since actual control is difficult to measure, perceived behavioral control is preferred: that is, the individual's perception of the ease or difficulty in performing the target behavior.

According to the TPB (Ajzen, 1991), the intention in turn, is affected by three conceptually independent antecedents. The first is the individual's attitude toward the behavior, thus having a favorable or unfavorable appraisal of the behavior. The second is the subjective norm, which refers to the perceived social pressure to perform or not to perform the target action. The last antecedent of intention is the previously mentioned PBC. Past experience, anticipated difficulties or facilitating conditions, are considered. Intentions to act are generally stronger when the individual's attitude toward the action is favorable, one perceives a social pressure to act and when the person feels that he or she can control the behavior.

The TPB has attracted a great deal of attention. It has been applied to a multitude of behavioral domains, including health-relevant behaviors. A meta-analysis of the TPB (Armitage & Conner, 2001) found that the model accounts for 39% of the variance in behavioral intention, and 27% of the variance in behavior for a variety of target actions. A following meta-analysis of meta-analyses (Conner & Sparks, 2005) specifically centered on health-related behavior studies showed that intention and PBC explained 25.6% of the variance in behavior, whereas attitude, subjective norm and PBC accounted for 33.7% of the variance in intentions.

Despite its usefulness and parsimony, the critical issue of the ‘intention–behavior gap’ in the TPB still remains: good intentions are not always translated into behavior. This explains the numerous developments of the TPB proposed in literature, involving either re-conceptualization or addition of predictors to enhance the explanation of behavior (for a review, see Conner & Sparks, 2005). Several investigators proposed a distinction between instrumental component of attitude and the affective component (e.g., Bagozzi et al., 2001, Hagger and Chatzisarantis, 2005). The former represents the evaluative and instrumental nature of attitude measured by semantic differential, using such items as worthless/valuable, and harmful/beneficial. Affective attitude has a more experiential quality and is measured by such items as pleasant/unpleasant and enjoyable/unenjoyable. Although these components may be highly correlated, they can be discriminated empirically and have different functions (Breckler and Wiggins, 1989, Eagly and Chaiken, 1993).

Research conducted using the TPB framework has yielded evidence on the predictive power of past behavior on future behavior (see Ouellette & Wood, 1998, for a meta-analytic review). Conner and Armitage (1998) showed that past behavior explained a further 7.2% of variance in intentions (across 12 studies), over and above the contribution of TPB variables. Likewise, past behavior accounted for a mean 13% of variance in behavior (across five studies), after controlling for intentions and PBC.

Consistent findings have showed that past behavior may be the best predictor of future behavior, rather than the cognitions described in the TPB (Ajzen, 2002b, Ouellette and Wood, 1998, Rhodes and Courneya, 2003). Ouellette and Wood (1998) pointed to frequency of past behavior as a measure of habit strength. Once established, such a habit may control subsequent behavior without deliberate cognitive mediation. Nevertheless, it must be underlined that a behavior does not necessarily become habitual just because it has been carried out repeatedly. Indeed, in contrast to Ouellette and Wood, 1998, Ajzen, 2002b suggested that the association between past and later actions merely demonstrates that the behavior is stable over time. That is, if the factors that had determined the action in the past remain unchanged, they are also likely to exert their effect in the present. The residual effect of prior behavior, after taking into account the TPB variables, may be explained in several ways. First of all, it can be considered a test to support the theory: the significant effect of past behavior on the future one may reflect the influence of the factors that were present at both times of measurement but which have not been taken into account by the investigators (Bagozzi, 2004, Ajzen, 2002b, Bamberg et al., 2003, Rhodes and Courneya, 2003).

Secondly, the residual effect of past behavior may simply be a measurement confound due to the common method variance shared by the measurements — the items used to measure prior and current behavior are often very similar (Ajzen, 2002b). Finally, past behavior may have a statistical predictive value, more than a theoretical merit, because of temporal instability of TPB cognitions: some research have found that past behavior effect is weaker when intentions are well-formed and stable, whereas it becomes stronger in conditions of unstable intentions (Conner et al., 2000, Sheeran et al., 1999).

The TPB was also used to predict and explain various behaviors that may be associated with food consumption (Lobb et al., 2007, Verbeke and Vackier, 2005) and a great number of generic foodborne diseases. The research focused in particular on food handling and results were generally consistent with Conner and Armitage's (2001) meta-analysis. TPB was found to account for 34% of variance in intentions of hand hygiene malpractice in catering establishments (Clayton & Griffith, 2008), 48% in intentions of food safe handling practice (Seaman & Eves, 2010), and even 79% in intentions and 87% in behavior of hand hygiene practice in hospitals (Jenner, Watson, Miller, Jones, & Scott, 2002). Interestingly, some research tested the effect of past behavior as an additional predictor of the intentions and behavior. For instance, Mullan and Wong (2010) used TPB to predict 33% of intentions and 15% of behavior in a food hygiene intervention; the inclusion of past behavior into the model increased the variance accounted for, respectively of 6% and 4%. Mullan and Wong (2009) also found that TPB explained 66% of variance in intention and 21% in behavior of preparing food hygienically. The addition of past behavior accounted for an extra 3% for intention and, more importantly, for an extra 19% in behavior.

To our knowledge, there have been no studies, within the TPB framework, on a specific practice to prevent salmonellosis, and on the behavior of consuming thoroughly cooked meat and fresh sausages. As previously mentioned, food poisoning in the domestic environment is often due to insufficient cooking (EFSA, 2010). Research has been underscoring the importance of cooking food adequately, regarding it as a fundamental behavioral practice that ought to be evaluated and promoted in food safety education (Medeiros, Hillers, Kendall, & Mason, 2001). There is no literature that focuses specifically on the cognitive determinants of this behavioral practice. Our study therefore strives to bridge this gap, uncovering the processes that lead to the formation of the intention and subsequently to the specific behavior. In particular, we will test the explanatory and predictive power of the TPB, and the TPB plus the addition of past behavior.

Considering the previous points mentioned regarding past behavior, we believe that the inclusion of a measure of past behavior into the TPB framework may benefit the present research. Since there is little in literature concerning social cognition models' explaining our target behavior, our study could be regarded as an initial effort to explain the effect of past behavior in this practice. Because food behaviors are often habitual (see e.g., Verbeke & Vackier, 2005), we assume that even in the case of this best practice past behavior may be conceived as habitual response. Thus, it should not have a causal effect on the deliberative intentions but just on the behavior (see Rhodes & Courneya, 2003). However, alternative explanations which are empirically driven must also be considered.

Two longitudinal studies were carried out, one which considered a sample of young adults (19–29 years), the other a sample of adults. Results of the two samples were then compared. This choice stemmed from previous results. It demonstrates that food safety knowledge of young consumers is commonly insufficient and their practices are inappropriate (Byrd-Bredbenner et al., 2007). Additionally, findings suggest that the likelihood to engage in risky food handling generally tends to be higher among young people than adults (Li-Cohen and Bruhn, 2002, Sanlier, 2009, Wilcock et al., 2004). The hypothesis that the processes that underlie intentions and behavior to consume meat and fresh sausages well cooked may be different in young people and adults will therefore be tested.

Section snippets

Participants and procedure

Two studies were carried out in parallel. All participants were asked to take part in a large longitudinal study about food security; for both samples, we recruited people who claimed to be meat consumers. In both waves, the instrument used was a questionnaire.

The institutions promoting the research project granted ethical clearance for the study. Participants took part on a voluntary basis. They were informed about the scope of the research and were assured that all responses would remain

Descriptive statistics

In both samples, except for the subjective norm and actual behavior, the items of each scale were averaged to form reliable composite scores. Scale reliability was satisfactory and higher than .77 in Sample 1 (young adults) and .64 in Sample 2 (adults). Means, standard deviations, and the mean differences between the samples are shown in Table 1.

As it can be noted, the two samples claimed to have a positive attitude, both affective and evaluative, toward the target behavior. The means are

Discussion

This research has used the TPB framework (Ajzen, 1991) to investigate a food safety practice – consuming meat and fresh sausages only if well cooked – to prevent the risk of salmonellosis infection. Since Salmonella remains one of the most important pathogens of foodborne diseases (Pangloli et al., 2008), it becomes particularly relevant to fully understand the cognitive determinants of domestic food practices from the public health viewpoint. To date, to our knowledge, the cognitive

Conclusion

This longitudinal research focused on a target behavior – the consumption of meat and fresh sausages only if well cooked – which literature on domestic food safety has neglected till present, despite its critical role in preventing salmonellosis (e.g., Busani et al., 2009, Cibin et al., 2009). The TPB was used to explain the processes that lead to the formation of intentions and actual behavior in two samples (young adults and adults) that may differ in their cognitions.

Findings confirm the

Acknowledgment

This study was part of the research activities promoted by the Istituto Zooprofilattico Sperimentale delle Venezie as part of the research project RC 18/2004, funded by the Italy's Ministry of Health.

References (72)

  • A.E. Lobb et al.

    Modelling risk perception and trust in food safety information within the Theory of Planned Behavior

    Food Quality and Preference

    (2007)
  • M. McCarthy et al.

    Who is at risk and what do they know? Segmenting a population on their food safety knowledge

    Food Quality and Preference

    (2007)
  • N. Michaelidou et al.

    Modeling the factors affecting rural consumers' purchase of organic and free-range produce: A case study of consumers' from the Island of Arran in Scotland, UK

    Food Policy

    (2010)
  • B.A. Mullan et al.

    Hygienic food handling behaviors: An application of the Theory of Planned Behavior

    Appetite

    (2009)
  • B.A. Mullan et al.

    Using the Theory of Planned Behavior to design a food hygiene intervention

    Food Control

    (2010)
  • E.C. Redmond et al.

    Consumer food handling in the home: A review of food safety studies

    Journal of Food Protection

    (2003)
  • N. Sanlier

    The knowledge and practice of food safety by young and adult consumers

    Food Control

    (2009)
  • B. Tiozzo et al.

    Development and evaluation of a risk-communication campaign on salmonellosis

    Food Control

    (2011)
  • W. Verbeke et al.

    Individual determinants of fish consumption: Application of the Theory of Planned Behaviors

    Appetite

    (2005)
  • A. Wilcock et al.

    Consumer attitudes, knowledge and behavior: A review of food safety issues

    Trends in Food Science and Technology

    (2004)
  • J.M. Abbot et al.

    Comparison of food safety cognitions and self-reported food-handling behaviors with observed food safety behaviors of young adults

    European Journal of Clinical Nutrition

    (2009)
  • C. Abraham et al.

    Health psychology: Topics in applied psychology

    (2008)
  • Ajzen, I. (2002a). Constructing a Theory of Planned Behavior questionnaire. Retrieved January 24, 2011, from...
  • I. Ajzen

    Residual effects of past on later behavior: Habituation and reasoned action perspectives

    Personality and Social Psychology Review

    (2002)
  • S. Altekruse et al.

    Emerging foodborne diseases

    Emerging Infectious Diseases

    (1997)
  • C.J. Armitage et al.

    Efficacy of the Theory of Planned Behaviour: A meta-analytic review

    British Journal of Social Psychology

    (2001)
  • V. Arzenton et al.
  • V. Arzenton et al.

    Sicurezza alimentare: Percezioni e comportamenti a confronto per un'efficace comunicazione del rischio (Triveneto, 2008) [Food safety]

  • R.P. Bagozzi

    Consumer action: The role of intentionality, emotion, sociality, and agency

  • R.P. Bagozzi et al.

    A general approach to representing multi-faceted personality constructs: Application to state self-esteem

    Structural Equation Modeling

    (1994)
  • R.P. Bagozzi et al.

    Decisions to donate bone marrow: The role of attitudes and subjective norms across cultures

    Psychology & Health

    (2001)
  • S. Bamberg et al.

    Choice of travel mode in the Theory of Planned Behavior: The roles of past behavior, habit, and reasoned action

    Basic and Applied Social Psychology

    (2003)
  • P.M. Bentler

    Comparative fit indexes in structural models

    Psychological Bulletin

    (1990)
  • S.J. Breckler et al.

    On defining attitude and attitude theory: Once more with feeling

  • L. Busani et al.

    Infectious by Salmonellas in Italy

  • C. Byrd-Bredbenner et al.

    Development of food safety psychosocial questionnaires for young adults

    Journal of Food Science Education

    (2007)
  • Cited by (28)

    • Adopt proper food-handling intention: An application of the health belief model

      2021, Food Control
      Citation Excerpt :

      Firstly, this research only focuses on consumers' intention to proper food-handling, not actual behavior. Although intention is assumed to be the immediate antecedent of practice (Ajzen, 2002) and proper food-handling intention is the strongest predictor of actual practices (Mari, 2012), intention is not the actual behavior. Future research can consider explaining the actual food-handling behavior based on HBM.

    • Determinants of consumer food safety self-protection behavior– an analysis using grounded theory

      2020, Food Control
      Citation Excerpt :

      Interventions based on the theoretical frameworks have been proved more effective than non-theory-based intervention (Young, Thaivalappil, Greig, Meldrum, & Waddell, 2018). Specifically, various social psychology frameworks have been applied to explore and explain safely food handling behaviors, such as theory of planned behavior, health belief model, health action process approach (Hanson & Benedict, 2002; Hassan, Dimassi, & Karam, 2018; Lazou, Georgiadis, Pentieva, Mckevitt, & Iossifidou, 2012; Mari, Tiozzo, Capozza, & Ravarotto, 2012; Mullan, Allom, Sainsbury, & Monds, 2016; Mullan, Wong, & Kothe, 2013). However, these frameworks based on the research results of western consumers may have lower explanatory power in the Chinese context (Bai et al., 2014).

    • Theory of planned behavior on food waste recycling

      2020, Waste Biorefinery: Integrating Biorefineries for Waste Valorisation
    View all citing articles on Scopus
    View full text