Illness perceptions of adults with eczematous skin diseases: a systematic mixed studies review

Eczema is a very common skin condition. It predominantly comprises of atopic dermatitis (AD, syn. atopic eczema (AE), eczema) and contact dermatitis (CD), which are henceforth referred to as eczematous skin diseases (ESD). AD is usually characterized by chronic or chronically relapsing skin inflammation with onset frequently already in early childhood and is associated with dry skin and intensive pruritus [1, 2]. It is primarily an endogenous disease, but can be pivotally influenced by environmental factors [3]. However, its etiology is not completely understood yet [4]. CD comprises irritant contact dermatitis (ICD) and allergic contact dermatitis (ACD). It is caused by occupational or non-occupational skin exposure to irritants and/or contact allergens, respectively [5]. Avoidance of the causative agent(s) may lead to clearance of skin symptoms. If detection of the causative agent(s) fails or its complete avoidance is not possible, CD may also result in a chronic condition [6].

For the individual patient, the burden of disease is usually high due to soreness or itching, sleep disorders, feelings of stigmatization, restraints on leisure activities, prolonged sick leave from work or school, impaired social contact, and time consuming treatment, just to name a few [4, 7‐11]. Studies indicate a significant and long-term impact on quality of life (QoL) of patients with AD [8, 10, 12], contact dermatitis [7], and (especially work-related) ACD [13‐15]. In addition, several studies found higher rates of anxiety and depression among AD patients [10, 16‐19].

Against this background, coping with ESD is of particular importance. In order to understand coping processes, defined as the processes of adapting to health threats (e.g., adherence to treatment), it is necessary to examine the patients’ perspectives on their illness [20, 21]. These so-called illness perceptions (or illness representations, e.g., people’s understanding of their illness) are essential, because they have a large impact on health behavior, especially with regard to coping responses and illness management [21, 22]. Leventhal’s Common-Sense Model of Self-Regulation of Health and Illness (CSM) [20, 23] is a valuable framework to describe and understand choosing and planning of coping processes. It assumes individual cognitive and emotional representations, both independently, to be key determinants to evaluate health and illness [22, 24]. Several studies report correlations of illness perceptions with pivotal behavioral and quality-of-life outcomes [25]. In their meta-analytic review, Hagger and Orbell [25] found that perceived serious consequences, high identity beliefs and expected chronic timeline of the illness are associated with impaired psychological well-being, role and social functioning, more psychological distress as well as decreased vitality. Additionally, higher psychological well-being, social functioning, and vitality were associated with higher control beliefs [25]. Broadbent et al. [26] systematically reviewed the usage of the Brief Illness Perception Questionnaire (B-IPQ) [27] in various illnesses and outlined strong associations between depression and anxiety together with lower quality-of-life dimensions, and serious perceived consequences, emotional representations, and strong identity. Higher personal and treatment control beliefs were found to be negatively associated with depression and anxiety and, furthermore, positively associated with better quality of life [26].

The illness representations of various skin diseases (for instance psoriasis, including psoriasis arthritis [28‐30], vitiligo [31], and alopecia areata [32]) are thoroughly investigated. However, there is no systematic review on the illness representations of patients with ESD. The aim of this systematic mixed studies review was therefore to describe the representations of health and illness for this patient population.

This review was registered in the International Prospective Register of Systematic Reviews on September 18, 2018 (PROSPERO registration number CRD42018109217) [33]. This report follows the PRISMA statement (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) [34]. A completed PRISMA checklist is attached in Additional file 1.

The aim of this systematic mixed studies review was to provide an overview of the representations of health and illness of patients with eczematous skin diseases. The narrative synthesis of the seven studies that met the inclusion criteria demonstrated that patients with ESD form their own illness perceptions. Besides strong illness identity and long timeline beliefs, we identified various suspected causes for the disease from the studies. Our results furthermore indicate that patients expect severe consequences in various contexts (e.g., social, economic, and occupational) and perceive a strong emotional impact of the skin disease.

As pointed out before, we refer to atopic dermatitis and contact dermatitis as eczematous skin diseases. This distinction, however, is only of limited appropriateness for these complex diseases, since AD is primarily an endogenous disease; it often manifests during childhood and usually has a relapsing course. If there is no remission during childhood or if there is a relapse in adolescence, the disease can occur continuously and may have a chronic course. AD is also frequently associated with other atopic diseases such as allergic asthma and allergic rhinitis [1, 2]. In contrast, exogenous causes can induce CD. Exposure to irritants may lead to irritant CD, which accounts for about 80% of all CD cases. Allergic contact dermatitis occurs after prior sensitization upon exposure to a specific contact allergen. In either instance, avoidance of exposure to allergens and irritants is of particular importance to prevent CD becoming a chronic condition [5, 6]. Because of the difficulty of clinically differentiating ICD and ACD and due to the fact that both may co-exist [5], both diagnoses have been included in the present review. Furthermore, AD and CD may have similar impacts on the lives of the affected individuals, in particular regarding impairment of quality of life, occupational impact, and higher risk for anxiety and depression [7, 8, 10, 12‐16]. When examining the emotional representations in the present review, a key finding was that studies report a strong emotional effect and numerous concerns in nearly all study groups regardless of whether the diagnosis was CD or AD. These findings are consistent with a recent published study including AD patients across nine European countries on emotional consequences [58], which demonstrated that nearly 57% are emotionally burdened due to living with AD. Broadbent et al. [26] have also demonstrated the association of stronger emotional representations with reduced quality of life and higher depression and anxiety, which supports these results.

Despite the heterogeneity of our studies, triangulation revealed strong illness identity among the study participants, in particular patients with AD, meaning that these patients view their illness as highly symptomatic. In their meta-analytic review, Hagger and Orbell [25] investigated intercorrelations between the dimensions of the CSM across various studies and revealed a logical structure of relationships between the cognitive dimensions. They, for example, found positive correlations between illness identity, timeline, and consequences. Furthermore, identity is significant negatively associated to perceived control. In practice, this could imply that participants with a strong illness identity perceive their disease as more chronic, less controllable, and associate their disease with more serious consequences. However, it has to be considered that Hagger and Orbell [25] report statistical analyses to determine construct and discriminant validity of the model that do not reveal causal relationships (e.g., if patients perceive more serious consequences because of a highly symptomatic disease or vice versa). Since AD being a primarily endogenous and occasionally chronic disease and, as found in our review, is perceived as highly symptomatic (e.g., due to itching), future studies should concentrate on the investigation of illness representations considering different levels of disease severity. In terms of coping behavior, relationships between consequences and identity to maladaptive coping, namely “avoidance/denial” and “expressing emotions,” have been shown [25]. The same authors [25], furthermore, revealed a strong negative association of consequences and identity to the adaptive illness outcome “physical functioning” and positive association to “psychological distress.”

In our review, we assigned the reported causes to two main categories (endogenous and exogenous causes) and six subcategories, namely, psychological factors, medical condition, behavioral factors, environmental factors, domestic factors, and occupational factors. Moss-Morris et al. [40] carried out analyses on the causal items of the IPQ-R, and identified an underlying four-factorial structure of the scale, comprising psychological attributions, risk factors, immunity, and accident or chance. This structure, however, could only be reproduced to a limited extent by Wittkowski et al. [59] in a study with AD patients. In their meta-analytic review, Hagger and Orbell [25] also list various categories that can be used to classify the causes: biological causes, emotional causes, environmental causes, and psychological causes. They consider that categories might overlap, which impedes the interpretation of results [25]. The classification of perceived causes in the present review (see Table 3) is largely congruent with the subscales formulated by Moss-Morris et al. [40] and categories of Hagger and Orbell [25]. However, behavior-related causes (e.g., frequent hand washing, use of emollients) could only be assigned to a category or subscale to a limited extent. They fit best into the subscale “risk factors” postulated by Moss-Morris [40] (cf. item: “My own behavior”), although this item was originally assigned to the subscale “psychological attributions.” Future studies should hence consider a disease-specific adaptation of the cause-scale already during study planning.

Interestingly, we identified several perceived causes in the included studies, which are not represented in the relevant literature, namely, the consumption of red wine or pork. Although these causes cannot be ruled out in individual cases, plausible associations are seldom reported in clinical trials and tend to be overestimated, e.g., in terms of the relationship of hand eczema and alcohol consumption [60, 61]. Our review in addition revealed the widespread assumption that emollients or the ingredients of emollients—despite negative patch test results—could be the causative agent of the skin disease. Although this assumption seems paradoxical at first glance, it has already been described in the literature for protective creams and emollients [62]. As direct application of emollients to inflamed skin is often poorly tolerated, current guidelines, e.g., for the treatment of AD, recommend treatment of acute flare-ups first [63]. It is of particular importance for clinical practice and health care professionals to consider this, because—against the backdrop of the CSM—this example highlights the close connection between illness perceptions and the shown coping behavior (e.g., adherence to treatment recommendations). In practice, health care professionals should be aware of the individualized perceptions of patients, which may not always be consistent with medical facts [20, 22]. Furthermore, Leventhal et al. [23] stated that patients may conceal their reasons for non-adherence in order to avoid conflicts between their own perceptions and the underlying concept of the health care professional.

In summary, our review suggests that patients with eczematous skin diseases have complex illness representations regarding their disease as described by Leventhal et al. [23] in the Common Sense Model. These representations interrelate with the coping behavior and crucial illness outcomes. It seems to be important for medical professionals to consider them during treatment and counseling to ensure patient-centered care.