Chlamydia is the most commonly diagnosed sexually transmitted infection in England  and also across Europe . It is frequently asymptomatic , and without treatment can lead to serious health consequences for women such as pelvic inflammatory disease, infertility and ectopic pregnancy. There is also growing evidence that it can cause infertility in men [4, 5].
There is evidence that young people, who are disproportionately affected by chlamydia , may be underestimating the risk of infection. Studies have for example identified that important knowledge gaps exist [7–9] and that unhelpful beliefs may be influencing chlamydia risk appraisals in ways that reduce motivation to adopt protective behaviour .
A number of social cognition models include risk appraisal as a primary motivator of behaviour. These include the health belief model , the precaution adoption process  and protection motivation theory . These models suggest that changing risk appraisals could be an effective strategy in motivating protective behaviour such as condom use.
The motivational hypothesis  asserts that preventative behaviour is the result of the desire to reduce one’s risk. It makes logical sense that to be motivated to take precautionary behaviour one has to perceive the consequences of inaction as serious and likely to occur. Systematic reviews examining the predictive relationship between risk appraisal and health behaviour have however largely found this relationship to be small [14, 15], or absent in the case of sexual behaviour specifically . This suggests that an intervention aiming to increase chlamydia risk appraisals will have either no effect or a weak effect on condom use intentions and behaviour.
The existing body of evidence however suffers serious problems which may have served to underestimate the relationship between risk appraisal and future behaviour. Following a review of nearly 60 studies examining this relationship, Weinstein and colleagues  concluded that a high proportion had serious conceptual and methodological flaws. These included the failure to control for past behaviour and the use of correlational data to examine the relationship between risk appraisal and behaviour. The use of correlational data to examine the relationship appears to be particularly common in studies of sexual behaviour . Although cross-sectional designs always limit conclusions that can be drawn regarding cause-effect relationships, they are especially problematic when examining the motivational hypothesis. This is because risk appraisals are both a determinant and a product of risk behaviour. According to the motivational hypothesis, it is the belief that taking precautionary behaviour will be effective in reducing feelings of risk that motivates an individual to act. It follows that once preventative action is taken, feelings of risk are reduced. This means that when taking concurrent measures of risk appraisal and behaviour, a negative relationship should be expected. For this reason, Weinstein and colleagues  advise using longitudinal data, where risk appraisal is measured at time one, behaviour is measured at time two, and analysis controls for behaviour at time one.
A further methodological problem is that studies examining the motivational hypothesis have largely used ‘unconditional’ measures of risk. Unconditional measures are those which ask individuals to rate the probability of an adverse event occurring without indicating, for example, whether this is if they do or do not use a condom, whether this is with a long-term partner or a casual sexual partner. Using conditional measures of risk is preferable, as these questions elicit risk appraisals based on such moderating factors. Conditional measures of risk are not only more conceptually accurate but enable the relationship between risk and behaviour to be examined in a consistent and interpretable manner . Weinstein and colleagues  advise that when testing the motivational hypothesis, that conditional measures of risk are taken in which individuals are asked to either rate their perceived vulnerability to the health threat if they continue with their existing levels of behaviour or if they take no precautionary action.
The present study will assess the efficacy of an intervention to change risk appraisals which overcomes the problems of much previous work in this area. The intervention is a single Sex and Relationships (SRE) lesson produced for the Health Protection Agency (HPA) and made available to teachers across ten European Union countries on their e-Bug website. The lesson was developed using Intervention Mapping [18, 19] which provides a framework for developing theory- and evidence-based interventions. A detailed report on the development and content of the intervention can be accessed using this link (http://www.healthinterventions.co.uk/projects.aspx?section=10&item=78).
In addition to strategies to increase chlamydia risk appraisals, the chlamydia lesson incorporates strategies to raise condom use response efficacy (the perceived effectiveness of condoms in reducing the threat of chlamydia) and self-efficacy (beliefs about one’s own ability to use condoms). Protection Motivation Theory (PMT)  was developed based on the observation that increasing risk appraisals amongst individuals who do not believe that they are able to perform behaviour can be ineffective. PMT predicts that as individuals’ feelings of threat increase, that protective behaviour will also increase if they feel able to cope with that threat. If on the other hand they feel they can do nothing or little to change their behaviour, then increased perceptions of risk can be counter-productive leading to avoidance (e.g. avoid thinking about STIs), information derogation (e.g. ‘health related messages are over-hyped’), or threat minimisation (e.g. denying you’re at risk of STIs). Evidence concerning whether threat appraisal and coping appraisal (the combined effect of response and self-efficacy) interact or operate in parallel is inconclusive . Until the nature of this relationship has been substantiated, interventions should seek to raise both threat and coping appraisals.
The proposed study uses a two-arm cluster randomised controlled trial (RCT) with a waiting-list control to test the efficacy of an intervention in increasing young adults’ condom use intention and behaviour through changing their chlamydia risk and coping appraisals. This will provide useful information about the ability of persuasive communication to bring about the desired programme effects. The waiting list control design has been chosen to allow all participants to benefit from the lesson. Cluster randomisation has been chosen to prevent within-school contamination.
The proposed study will use a conditional measure of risk. Given that a large proportion of the sample are not likely to be sexually experienced, participants appraisals of the risk of getting chlamydia if condoms aren’t used will be measured, rather than choosing the alternative ‘if continue with current behaviour’ option. In addition, this will be framed within the context of casual sexual partners for whom STI status is unknown given that this presents the most risky scenario. A further conditional stipulation will be vaginal sex, rather than oral or anal sex. The proposed study will also use an experimental design with follow-up for two months. Examining the motivational hypothesis in the context of an experimental study is ideal since delivery of the intervention should act to ‘destabilise’ existing chlamydia risk appraisals and reduce the predictive relationship between past and future behaviour for those who are already sexually experienced .
Whilst an experimental design affords the best circumstances for observing the motivational hypothesis, increasing chlamydia risk appraisals may not be sufficient to motivate condom use. Chlamydia is a single consequence of unprotected sex and evidence suggests that other outcomes of unprotected sex, such as the risk of pregnancy or making sexual experiences more enjoyable, may exert a more powerful effect on condom use intentions . Other authors  have also drawn attention to the dyadic context of condom use decision making and the possibility that risk appraisals are not powerful enough to overcome other determinants of behaviour such as embarrassment discussing condom use or uncooperative partners. The proposed study will include a measure of condom use intention as well as behaviour. This is necessary because it is anticipated that a large proportion of the sample will be sexually inexperienced requiring the use of intention as the outcome measure in examining the motivational hypothesis. In addition, it will also enable examination of whether an increase in chlamydia risk appraisals is sufficient to increase condom use intentions and/or behaviour. This will enable conclusions to be drawn about whether a failure to change behaviour is the result of a failure to increase condom use intentions or to overcome the effect of stronger factors operating when intention is translated into behaviour.
Primary research question
Secondary research questions
Is the lesson effective in increasing young people’s condom use during vaginal sex with casual sexual partners?
If the lesson is effective, are changes in young people’s condom use intentions or behaviour due to changes in their chlamydia risk appraisals (perceived likelihood and severity) and/or coping appraisals (condom use response efficacy and self-efficacy)?
This study has received ethical approval from Coventry University Ethics Committee.