Discussion
There were significant correlations between AD severity, caregiver illness perceptions, QoL and SE. Timeline cyclical (whether the AD was believed to be re-occurring) did not significantly correlate with any other variable. This was unexpected as AD can be chronic and episodic.2 This non-significance could be explained by varying item agreement within the timeline cyclical subscale. For example, in an AD adult population, agreement with an IPQ-R item which refers to symptoms changing daily, was significantly lower than other items that are explicitly cyclical.27 Building on existing evidence,16,27 this study found stronger correlations between QoL, emotional representations and consequences than between AD severity and QoL. This suggests that caregiver QoL may be impacted more by a caregiver’s emotional response to their child’s AD and their perceptions of its consequences on the child’s life than severity of the AD.
When asked about causal beliefs 68% of participants in this study believed their child’s AD was hereditary. There is a direct link between cause and control factors28 and ‘hidden causes’ under which genetic/biologic factors are classified are considered the least controllable compared to environmental or behavioural causes.29 This implies that parents who believe their child’s AD is hereditary may feel they have less control over it, thereby lowering their SE.
The hierarchical regression model found that caregiver illness perceptions (consequences, emotional representations and identity), AD severity, and SE predicted 72% of the variance in caregiver QoL. Adding SE at the third step did not explain additional variance. These results might suggest that SE is not a significant predicter of QoL which contrasts with existing research.12,13 A probable explanation is that the lack of additional variance explained by SE is due to multicollinearity, as SE is correlated with, and mediates, illness perceptions. In order to explore this a separate simple linear regression was run with just SE as the predictor, where it was able to explain 25% variance in QoL.
The importance of SE is further shown in the results of the mediation analysis, where SE mediated between all illness perceptions and QoL, except emotional representations. SE was found to fully mediate the relationship between caregiver views on chronicity, illness coherence and identity and QoL. This implies that the effect of these illness perceptions on QoL within those models was transmitted entirely through SE and so SE rather than these illness beliefs may be better targeted in interventions aimed at improving QoL. SE did not mediate between emotional representations and QoL, and therefore interventions should also focus on negative emotions such as fear anger and depression. Not only were negative emotions linked to reduced QoL but can also cause individuals to underestimate their ability thereby reducing SE.30
This study has a number of limitations. The mediation and hierarchical regression models leave a proportion of variance unexplained. Unexplored predictors could include coping strategies, and anxiety and depression, which are common in caregivers of children with AD.9,31 Whilst the current study’s mediation models are rudimentary and oversimplify the mechanisms between illness perceptions and self-efficacy, the results offer preliminary support to the integration of the CSM and SE models. More variables included and controlled for in future longitudinal research would allow for a more thorough examination of the causal mechanisms.
Overall, the study’s results imply that SE, emotional representations and perceptions of consequences are the most important variables to target to improve caregiver QoL. The findings emphasise the need for psychological assessment and treatment of caregivers to be embedded in the dermatological treatment pathway and have several implications. Clinicians need to be aware that caregivers can be less capable of managing their child’s condition if they perceive the consequences more negatively and struggle emotionally. By assisting caregivers to gain a realistic understanding of the consequences and causes of their child’s AD, clinicians may be able to reduce the negative emotional impact. The findings provide a foundation for future interventions targeting these modifiable variables to support better QoL in caregivers who, in turn, can provide optimum care to their children living with AD.