During the last pandemic influenza antivirals were prescribed both as prophylaxis and treatment. However, adherence rates were suboptimal. This study assessed the effect of theory-based and evidence-based health messages, which promoted the use of antivirals as prophylaxis for pandemic influenza, on intentions to use antivirals. Using hypothetical scenarios, the investigators tested whether written health communications about pandemic flu and recommended preventative medication (i.e. a prophylactic treatment with antivirals) had an effect on study participants' beliefs about the pandemic flu and the advice received, and their intention to adhere to the recommendation. In particular, the investigators assessed the behavioural impact of health messages presented in four different linguistic formats, defined by a 2×2 (agency assignment × attribute framing) factorial design. The originality of this study relies on the attempt to maximise the behavioural impact of written health messages by combining the agency assignment and attribute framings, which have never been tested together, and by systematically targeting specific predictors of adherence intentions through these messages. The findings of this study may be used to improve the behavioural impact of health communications to the general public in case of a pandemic flu outbreak in the UK.
Background: In 2009, a new strain of influenza virus, called pandemic influenza A(H1N1)pdm09, was detected in humans. During the pandemic in 2009/2010, antiviral medications were prescribed both as prophylaxis and treatment for influenza. However, many people, including those with chronic health conditions, did not used the antivirals as prescribed. This study focused on adherence to antivirals used as prophylaxis, as this is a key preventative stage before any vaccine can be available to the public. To ensure protection of public health during public health emergencies, such as pandemic flu, it is essential that the general public follow the health advice received. Therefore, future risk communications about the use of antivirals for pandemic influenza will need to address the barriers and facilitators of adherence. This study investigated the effect of theory-based and evidence-based health messages, which promoted the use of antivirals as prophylaxis for pandemic influenza, on intentions to use antivirals. In particular, using hypothetical scenarios, the investigators tested whether written health communications about pandemic flu and recommended preventative medication (i.e. a prophylactic treatment with antivirals) had an effect on study participants' beliefs about the pandemic flu and the advice received, and their intention to adhere to the recommendation. The linguistic manipulation used in the present study was based on an adaptation of a successful manipulation carried out by McGlone et al. (2013), who found that using the virus agency assignment in medical factsheets increased students' perception of susceptibility to and worry about the virus A/H1N1, perception of the severity of the consequences of catching the virus, perception of efficacy of the vaccine and reported intention to get the vaccine compared to the human agency assignment. The investigators wanted to test whether these results could be replicated in the context of taking antivirals for pandemic flu, using a more representative sample of the general population and controlling for baseline intentions. For what concerns the framing of the side effects, research shows that the way information on medication side-effects is presented affects people's intentions to accept the recommended medications. The same side-effect likelihood can be expressed in a positively framed message (i.e. 90% of people will not develop side-effects) or in negatively framed message (i.e. 10% of people will develop side-effects). Although a previous meta-analysis (Moxey et al. 2003) found no significant framing effect on either actual immunization or behavioural intention, it found that positively framed information increased positive attitudes toward immunization and perception of the benefits of immunization, decreased expectation of side-effects and led to a more realistic understanding of the side-effects of influenza vaccines. In addition, the health messages used in this study targeted specific variables, which have been in the past associated with higher adherence to antivirals for pandemic influenza. In particular, a recently published systematic review (Smith et al. 2016) investigated the factors affecting adherence to antivirals for influenza, including pandemic flu. They included 26 studies, of which 23 assessed pandemic influenza. Although many of the reviewed studies had serious methodological flaws, the review identified several psychosocial factors associated with adherence to antivirals prescribed as prophylaxis for pandemic flu. In particular, 11 studies investigated actual adherence to prophylactic antivirals. These studies showed that actual adherence was associated with previous adherence to other preventative measures for pandemic flu, beliefs that the recommended prophylactic measures were necessary and having discussed taking oseltamivir, a commonly used antiviral, with someone who had not experienced side effects. Although only one study found that the presence of side-effects was a significant predictor of non-adherence, the occurrence of side-effects was the most commonly self-reported reason for the discontinuation of oseltamivir. The review also included 8 studies investigating psychosocial predictors of intention to adhere: for instance, higher perceived susceptibility to the pandemic flu, higher perceived severity of pandemic flu, knowledge of pandemic flu, higher self-efficacy and response efficacy, and having previously accepted the flu vaccine were all associated with higher adherence intentions. Methods: In May/June 2016 the investigated conducted an online experiment. After reading a hypothetical pandemic flu scenario, 349 adult UK residents were randomly allocated to one of four conditions, defined by a 2x2 (agency assignment × attribute framing of side effects) factorial design. Each condition presented messages describing the pandemic flu using linguistic expressions that assigned agency to either the humans (HA: human agency) or the virus itself (VA: virus agency), whilst describing the antivirals side effects in terms of the chances of either experiencing (NF: negative framing) or not experiencing side effects (PF: positive framing). Although presented with different framings the health messages provided equivalent information that targeted specific predictors of adherence. Participants were blinded to group assignment. Intentions to use the antivirals and potential mediating factors were measured. The aim was to understand the potential behavioural impact of the four health messages developed and to better understand the factors influencing adherence intentions.
Study Type
INTERVENTIONAL
Allocation
RANDOMIZED
Purpose
PREVENTION
Masking
SINGLE
Enrollment
349
Linguistic framing used in written health messages. Each version of the messages described the pandemic flu and the efficacy of the antivirals using linguistic expressions that assigned transmission to either humans (HA: 'You can contract the virus when you touch…') or the pandemic flu virus itself (VA: 'It can infect you when you touch…')
Linguistic framing used in written health messages. Each message described the side effects of the antivirals in terms of chances of either experiencing (negative framing: 'Uncommon side effects (10% of people will be affected)') or not experiencing side effects (positive framing: 'Uncommon side effects (90% of people will not be affected)') after using them.
Change in Intentions to Take Antivirals for Pandemic Flu
Mean adherence intentions post- exposure to the health information in the 4 groups. Intentions were measured by self-report items: participants were asked to agree with three statements about their intentions to take antivirals as recommended in the hypothetical scenario (on 9-point scale, where 1=strongly disagree to 9=strongly agree). The scores reported below represent a composite variable 'change in intentions', which reflects the average of the three items that compose it.
Time frame: Straight after exposure to the health messages
Worry of the Pandemic Flu Threat
Measured by self-report items adapted from Witte et al. (2001): participants were asked to agree with two statements about their perceived worry on 9-point scale, from 1=strongly disagree to 9=strongly agree. Higher scores indicate higher levels of reported worry about pandemic flu threat. This is a composite variable reflecting the average of the items that compose it.
Time frame: At 20 minutes (i.e. straight after exposure to the health messages)
Perceived Susceptibility to the Pandemic Flu
Measured by self-report items: participants were asked to state how likely they were to get sick with pandemic flu, had they not taken prophylactic medication o a 9-point scale, where 1=not likely at all, to 9=extremely likely.
Time frame: At 20 minutes (i.e. straight after exposure to the health messages)
Perceived Severity of the Pandemic
Measured by one self-report item (adapted from Witte et al. (2001): participants were asked to agree with a statement about their perception of the severity of the pandemic flu on 9-point scale, from 1=strongly disagree to 9=strongly agree. Higher scores indicate higher levels of perceived severity of the pandemic flu threat.
Time frame: At 20 minutes (i.e. straight after exposure to the health messages)
Perceived Self-efficacy
Measured by one self-report item (adapted from Witte et al. (2001): participants were asked to agree with a statement about their perceived ability to take the antivirals as recommended on 9-point scale, from 1=strongly disagree to 9=strongly agree. Higher scores indicate higher levels of reported self-efficacy.
Time frame: At 20 minutes (i.e. straight after exposure to the health messages)
Perceived Efficacy of the Antivirals
Measured by two self-report items (adapted from Godinho et al. (2016): participants were asked to agree with two statements about their perception of the efficacy of the antivirals against pandemic flu on 9-point scale, from 1=strongly disagree to 9=strongly agree. Higher scores indicate higher levels of perceived efficacy of the antivirals. This is a composite variable reflecting the average of the items that compose it.
Time frame: At 20 minutes (i.e. straight after exposure to the health messages)
Perceived Response Costs
Measured by self-report items adapted from Godinho et al. (2016): participants were asked to agree with four statements about their beliefs about the side effects and negative consequences of using antivirals on 9-point scale, from 1=strongly disagree to 9=strongly agree. Higher scores indicate higher levels of perceived response costs. This is a composite variable reflecting the average of the items that compose it.
Time frame: At 20 minutes (i.e. straight after exposure to the health messages)
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.