ABSTRACT
Background
Atopic dermatitis (AD) significantly impacts quality of life, with well‐documented physical and psychological consequences. While disease burden is well characterised, the influence of psychosocial determinants on AD outcomes remains underexplored. This study examines these factors in an Australian paediatric population.
Objectives
To investigate whether children from populations at risk of health inequities—such as those from rural areas, culturally diverse backgrounds, First Nations communities, socioeconomically disadvantaged postcodes, those with neuropsychiatric conditions and ‘Vulnerable’ children (defined as those under state or relative care or with a history of familial abuse or neglect)—experience more severe disease.
Methods
This retrospective cross‐sectional study included children ≤ 16 years with confirmed AD attending a tertiary paediatric referral centre in Melbourne (Aug 2022—Aug 2023). Risk factors were defined by social determinants of health, including economic, cultural and geographic influences. Disease outcomes (severity, hospitalisations, emergency department visits treatment patterns) were analysed using univariate and multivariate methods. A total of 454 at‐risk children were compared to 454 controls.
Results
Children from at‐risk groups were more likely to have severe AD (43.8% vs. 28.3%, p < 0.001), higher annual rates of hospital admissions, increased antibiotic use, and greater prednisolone use for flares (p < 0.001). Multivariate analysis revealed significantly higher odds of severe AD (aMOR 4.72, p < 0.001) and hospitalisation (aIRR 2.73, p < 0.001) in the at‐risk cohort.
Conclusion
Psychosocial determinants of health are associated with increased AD severity and healthcare use in children. These findings highlight the need for targeted, equity‐focused interventions to reduce disparities in AD care in Australia.
Keywords: atopic dermatitis, health equity, healthcare outcomes, healthcare policy, paediatric dermatology
1. Introduction
Atopic dermatitis (AD) is a common inflammatory skin condition, with an estimated global prevalence of 6% among children and adolescents, including 1.1% of adolescents and 0.6% of children experiencing symptoms of severe AD [1]. Chronic, severe AD in infants and young children is often associated with failure to thrive, and the disease imposes a significant burden on individuals and families across all age groups, negatively impacting quality of life [2, 3]. In addition to the physical effects, AD is frequently accompanied by psychological and social comorbidities, further complicating its management [3, 4]. However, despite the well‐documented consequences of AD, there is a limited body of research exploring how psychosocial determinants influence disease outcomes [5, 6].
Health disparities, driven by personal, societal, and systemic factors, contribute to unequal health outcomes in impacted populations [5]. In Australia, limited data exist on the burden of AD among ethnic minorities, its potential misdiagnosis, under‐recognition, and suboptimal treatment [4]. Barriers like poor access to culturally appropriate care and prohibitive out‐of‐pocket treatment costs exacerbate these disparities [4, 5]. Additionally, approximately 28% of the Australian population live in rural or remote areas, where healthcare access is more limited with longer wait times, often leading to worse health outcomes compared with those living in metropolitan areas [7].
A systematic review by Kuo et al. examining North American health disparities in paediatric dermatology highlighted that AD patients who do not see a specialist are more likely to have poor disease control, fewer referrals for patch testing, and higher rates of comorbidities such as asthma, allergies, and eczema herpeticum, particularly among non‐White populations [5]. Social determinants, including low‐income households, single‐parent families, and communities facing social isolation, contribute to the severity of AD [5]. To date, no Australian studies have explored the impact of psychosocial determinants of health on AD.
This study aims to address these gaps by examining the prevalence and characteristics of AD among six groups at risk of systemic health inequities, referred for specialist dermatology care at an Australian tertiary paediatric referral centre [8]. These groups include children from rural communities, First Nations peoples, those living in socioeconomically disadvantaged (SD) areas, children from culturally and linguistically diverse (CALD) backgrounds, those with co‐morbid neuropsychiatric conditions, and children flagged by healthcare providers as ‘vulnerable’ as defined by the Victorian Department of Health [9]. We hypothesise that children in these groups have poorer outcomes compared to those not in these groups. Findings from this study will help inform healthcare planning and resource allocation to improve AD outcomes for populations at risk of health inequities.
2. Methods
2.1. Study Design and Setting
This retrospective cross‐sectional single centre study was conducted at a tertiary paediatric referral centre in Melbourne, Australia. It is the major specialist paediatric hospital in the state of Victoria, extending care to children from Tasmania and southern New South Wales.
2.2. Study Population
This study included patients aged ≤ 16 years with a dermatologist‐confirmed diagnosis of AD of any severity who attended the dermatology outpatient department between August 2022 and August 2023. AD diagnoses were identified using the SlicerDicer validated tool within Epic, the hospital's electronic medical record (EMR) system, which utilises SNOMED clinical terms and ICD‐11 codes. Patients with other forms of atopy and other co‐morbidities were not excluded. The only exclusion criterion was diagnostic uncertainty regarding AD based on clinical notes.
2.3. Exposures
The six risk factors related to systemic health inequities in Australia measured were:
CALD background: Families flagged on the EMR as needing an interpreter or whose preferred language was not English.
Regional, rural, or remote locations: The Modified Monash Model (MMM) categorises Australian areas into seven remoteness categories. Data were collected on patients living in MM 2 to MM 7 areas, which are considered regional, rural, or remote.
SD postcodes: This group was defined using the Australian Bureau of Statistics' (ABS) rankings of postcodes in Victoria, based on the latest Socio‐Economic Indices for Areas (SEIFA) from the 2021 census. Patients living in the top 20 most disadvantaged postcodes out of approximately 3183 in Victoria were included [10].
Aboriginal and Torres Strait Islander peoples: This was derived from demographic data recorded in the EMR.
Mental health diagnosis or neurodevelopmental disorder: Patients included in this group had a confirmed diagnosis of anxiety, depression, obsessive‐compulsive disorder (OCD), post‐traumatic stress disorder (PTSD), autism spectrum disorder (ASD), attention deficit hyperactivity disorder (ADHD), intellectual disability, developmental delay, and/or behavioural problems.
Vulnerable child: The EMR record has an obligatory ‘FYI’ flag or clinical notes documentation indicating a child is under state care, in relative care, or has a history of familial abuse or neglect.
2.4. Participant Selection
Of the 1480 eligible patients identified, 454 patients with at least one risk factor related to health inequity were flagged by the SlicerDicer tool (Figure 1). These patients formed the at‐risk cohort. For the control group, an equal number of patients without any of the six identified risk factors were randomly selected using a computerised randomisation function in the Epic system, ensuring unbiased selection.
FIGURE 1.
Summary of patients in each group at‐risk of experiencing health inequities (total study population; n = 1480).
2.5. Outcomes
Poor AD outcomes were defined as severe AD, an increased number of emergency department (ED) presentations per life year, and/or hospital admissions for AD management per life year compared to the control group. The severity of AD was assessed by dermatology‐trained healthcare professionals, including advanced trainees, consultants or nurse practitioners, based on the clinical signs and symptoms of AD. When available, validated measures such as the SCORing Atopic Dermatitis (SCORAD) and Eczema Area and Severity Index (EASI) were used to quantify severity, with the highest recorded score during the study period determining the classification. Specifically, an EASI score of 7.1 to 21 indicates moderate disease, while a score > 21 indicates severe disease [11]. For SCORAD, a score of ≥ 25 to ≤ 50 indicates moderate disease, and a score > 50 indicates severe disease [12]. When these assessments were unavailable, severity was determined based on documented clinical exam findings and the clinician's impression.
2.6. Other Covariates
Other variables of interest included other demographic characteristics (age, sex and ethnicity when available), age at diagnosis of AD and first visit at the outpatient dermatology, co‐morbidities including other atopic diseases, type of AD (e.g., flexural (classical), discoid), therapies used to treat AD, other concomitant medications including the frequency of antibiotic and prednisolone use for AD flares, total number of patient visits, and missed patient visits to date at the time of data collection.
2.7. Statistical Analyses
All analyses were conducted on STATA v.18. Patient characteristics and outcomes were summarised for the total population and stratified by ‘at‐risk’ status. Patients were considered ‘at‐risk’ if they belonged to one or more of the described six groups. Frequency and proportions were used for categorical variables, and mean and standard deviations, or median and interquartile range, for numerical variables, dependant on normality.
Statistical differences in patient characteristics and outcomes between risk and not‐at‐risk groups were assessed using various univariate tests including Student's t test, Mann–Whitney U test, chi‐squared test and Fisher's exact test as appropriate.
Different univariate and multivariate regression models were utilised to assess the associations between at‐risk and not‐at‐risk groups and study outcomes. Logistic regression was applied for binary outcomes (e.g., recorded infections and prednisolone use), multinomial regression assessed disease severity (a categorical outcome), and Poisson regression evaluated rates of hospital admissions and ED presentations (count outcomes). Confounders, including age and sex, were included in the multivariate models.
Ethical approval for this study was obtained from the Hospital Research Ethics and Governance Office, including approval by an appropriate local First Nations representative. MCC DERP Reference number: 3640.
3. Results
3.1. Participants
Of the patients meeting the inclusion criteria (n = 1480), 14.3% were from regional or rural areas, 8.0% lived in a SD postcode, 7.6% had a neuropsychiatric condition, 6.0% were from CALD backgrounds, 2.8% were identified as ‘Vulnerable Children,’ and 1.5% identified as First Nations (Figure 1). As such, 30.68% (n = 454) were considered at risk. Specifically, 22.77% belonged to at least one at‐risk group, 6.35% belonged to two at‐risk groups, 1.42% belonged to three at‐risk groups, and 0.12% belonged to four at‐risk groups.
3.2. Descriptive Data
The mean age of the sample population (n = 910) was 7.3 years (SD 4.5; range 0–16 years, Table 1). Children in the at‐risk population were, on average, younger than those in the not‐at‐risk population (6.8 vs. 7.9 years; p < 0.001). There was no statistically significant difference in sex between the two groups (Table 1).
TABLE 1.
Summary of patient characteristics, by psychosocial risk status.
| Characteristics | Sample population | Not at risk | At risk | p |
|---|---|---|---|---|
| All patients | 910 | 456 | 454 | |
| Age (years) mean (SD) | 7.3 (4.5) | 7.9 (4.4) | 6.8 (4.4) | < 0.001 |
| Age of onset (years), mean (SD) | 1.1 (2.2) | 1.2 (2.4) | 1.0 (1.9) | 0.207 |
| Duration of AD (years), mean (SD) | 6.3 (4.3) | 6.8 (4.3) | 5.8 (4.2) | < 0.001 |
| Sex | ||||
| Female | 403 (44.3%) | 216 (47.4%) | 187 (41.2%) | 0.061 |
| Male | 507 (55.7%) | 240 (52.6%) | 267 (58.8%) | |
| Eczema type | ||||
| Classical | 677 (74.4%) | 344 (75.4) | 333 (73.4%) | |
| Discoid | 124 (13.6%) | 68 (14.9) | 56 (12.3%) | 0.118 |
| Classical and Discoid | 106 (11.7%) | 43 (9.4) | 63 (13.9%) | |
| Classical and Contact/Allergic Dermatitis | 3 (0.3%) | 1 (0.2) | 2 (0.4%) | |
| Treatments a | ||||
| Topical agents | 904 (99.3%) | 453 (99.1%) | 451 (99.3%) | 0.999 |
| Bleach baths | 614 (67.5%) | 272 (59.7%) | 343 (75.6%) | < 0.001 |
| Wet wraps | 310 (34.1%) | 123 (27.0%) | 187 (41.2%) | < 0.001 |
| Immunomodulators | 131 (14.4%) | 37 (8.1%) | 94 (20.7%) | < 0.001 |
| Biologics | 84 (9.2%) | 39 (8.6%) | 45 (9.9%) | 0.479 |
| UV | 68 (7.5%) | 31 (6.8%) | 37 (8.2%) | 0.438 |
| Other medications a | ||||
| Antihistamines | 534 (58.7%) | 253 (55.5%) | 281 (61.9%) | 0.050 |
| Melatonin | 26 (2.9%) | 4 (0.9%) | 22 (4.9%) | < 0.001 |
| Preventer inhaler | 82 (9.0%) | 37 (8.1%) | 45 (9.9%) | 0.344 |
| Vitamin D | 228 (25.0%) | 111 (24.3%) | 117 (25.8%) | 0.549 |
| Nasal spray | 12 (1.3%) | 6 (1.3%) | 6 (1.3%) | 0.994 |
| Allergy | ||||
| Food allergy | 159 (17.5%) | 82 (18.0%) | 77 (17.0%) | |
| Environmental allergy | 167 (18.4%) | 90 (19.7%) | 77 (17.0%) | 0.554 |
| Both | 169 (18.6%) | 86 (18.9%) | 83 (18.3%) | |
| No allergy | 415 (45.6%) | 198 (43.4%) | 217 (47.8%) | |
| EpiPen/Anaphylaxis | ||||
| Yes | 119 (13.1%) | 65 (14.3%) | 54 (11.9%) | |
| No | 791 (86.9%) | 391 (85.7%) | 400 (88.1%) | 0.291 |
| History of atopic disease | ||||
| Yes | 539 (59.2%) | 274 (60.1%) | 265 (58.4%) | 0.598 |
| No | 371 (40.8%) | 182 (39.9%) | 189 (41.6%) | |
| History of asthma | ||||
| Yes | 222 (24.4%) | 95 (20.8%) | 127 (28.0%) | 0.012 |
| No | 688 (75.6%) | 361 (79.2%) | 327 (72.0%) | |
| Patients with EASI score recorded | 227 (25.0%) | 100 (21.9%) | 127 (28.0%) | 0.035 |
| Patients with SCORAD recorded | 263 (28.9%) | 134 (29.4%) | 129 (28.4%) | 0.746 |
Abbreviations: EASI, Eczema Area and Severity Index; SCORAD, SCORing for Atopic Dermatitis.
Patients can be in multiple groups.
3.3. Outcome Data
Children belonging to at least one at‐risk group exhibited more severe disease (43.8% vs. 28.3%, p < 0.001) and had a higher annual rate of hospital admissions for AD management (p < 0.001, Table 2). However, there was no evidence of a difference in the annual rate of ED presentations for AD between at‐risk and not at‐risk populations (p = 0.014). The use of severity assessment tools like the EASI and the SCORAD was inconsistent, with only 25% and 28.9% of the sample population (n = 910) having these scores recorded, respectively. The at‐risk population also had a greater incidence of recurrent AD infections treated with oral antibiotics (33% vs. 18.9%, p < 0.001) and increased use of oral prednisolone for AD flares (11.9% vs. 5.5%, p < 0.001).
TABLE 2.
Summary of patient outcomes, by psychosocial risk status.
| Patient outcomes | Sample population | Not at risk | At risk | p |
|---|---|---|---|---|
| All patients | 910 | 456 | 454 | |
| Severity | ||||
| Mild | 170 (18.7%) | 120 (26.3%) | 50 (11.0%) | < 0.001 |
| Moderate | 412 (45.3%) | 207 (45.4%) | 205 (45.2%) | |
| Severe | 328 (36.0%) | 129 (28.3%) | 199 (43.8%) | |
| Oral antibiotics taken to treat AD infection | ||||
| No | 492 (54.1%) | 277 (60.8%) | 215 (47.4%) | < 0.001 |
| Once | 182 (20.0%) | 93 (20.4%) | 89 (19.6%) | |
| More than once | 236 (25.9%) | 86 (18.9%) | 150 (33.0%) | |
| Prednisolone use for AD flare | ||||
| No | 722 (79.3%) | 385 (84.4%) | 337 (74.2%) | < 0.001 |
| Once | 100 (11.0%) | 46 (10.1%) | 54 (11.9%) | |
| More than once | 88 (9.7%) | 25 (5.5%) | 63 (11.9%) | |
| ED presentations per life year, mean (SD) | 0.11 (0.33) | 0.08 (0.23) | 0.13 (0.40) | 0.014 |
| Hospital admissions per life year, mean(SD) | 0.04 (0.16) | 0.02 (0.10) | 0.06 (0.21) | < 0.001 |
| Appointments for AD per life year, mean(SD) | 0.88 (0.86) | 0.74 (0.72) | 1.02 (0.95) | < 0.001 |
Abbreviations: AD, atopic dermatitis; ED, emergency department.
3.4. Primary Analysis
When adjusting for age and sex, children in at least one at‐risk group had a 2.46‐fold increased risk (95% CI 1.67–3.64, p < 0.001) of having moderate AD compared to mild AD, and a 4.72‐fold (95% CI 3.10–7.19, p < 0.001) increase in the odds of severe AD compared to mild AD (Table 3). Additionally, these children exhibited a 2.73‐fold greater hospitalisation rate (95% CI 1.98–3.77, p < 0.001) and a 1.42‐fold greater ED attendance rate (95% CI 1.16–1.73, p = 0.001), after adjustments. Results also indicated that children in the at‐risk population (n = 454) had a 1.07‐fold increased risk (95% CI 1.03–1.11, p = 0.001) of receiving more than one course of oral antibiotics to treat a skin infection secondary to AD, as well as a 3.17‐fold increased risk (95% CI 1.93–5.19, p < 0.001) of receiving more than one course of prednisolone to treat an AD flare.
TABLE 3.
Detailed analysis of patient outcomes, with multivariate adjustments for age and sex.
| Patient outcomes | Crude analysis | Adjusted analysis a | ||
|---|---|---|---|---|
| MOR (95% CI) | p | aMOR (95% CI) | p | |
| AD severity | ||||
| Mild | Base outcome | — | Base outcome | — |
| Moderate | ||||
| Not at risk | 1.00 (Ref) | < 0.001 | 1.00 (Ref) | < 0.001 |
| At risk group | 2.38 (1.62–3.48) | 2.46 (1.67–3.64) | ||
| Severe | ||||
| Not at risk | 1.00 (Ref) | < 0.001 | 1.00 (Ref) | — |
| At risk group | 3.70 (2.49–5.51) | 4.72 (3.10–7.19) | < 0.001 | |
| Oral antibiotics taken to treat infection | ||||
| No | Base outcome | — | Base outcome | — |
| Once | ||||
| Not at risk | 1.00 (Ref) | 0.229 | 1.00 (Ref) | 0.097 |
| At risk group | 1.23 (0.88–1.73) | 1.34 (0.95–1.90) | ||
| More than once | ||||
| Not at risk | 1.00 (Ref) | < 0.001 | 1.00 (Ref) | 0.001 |
| At risk group | 2.25 (1.63–3.09) | 1.07 (1.03–1.11) | ||
| Prednisolone use for AD flare | ||||
| No | Base outcome | — | Base outcome | — |
| Once | ||||
| Not at risk | 1.00 (Ref) | 0.170 | 1.00 (Ref) | 0.065 |
| At risk group | 1.34 (0.88–2.04) | 1.50 (0.98–2.30) | ||
| More than once | ||||
| Not at risk | 1.00 (Ref) | < 0.001 | 1.00 (Ref) | < 0.001 |
| At risk group | 2.88 (1.77–4.68) | 3.17 (1.93–5.19) | ||
| Crude IRR (95% CI) | p | aIRR (95% CI) | p | |
|---|---|---|---|---|
| Number of hospitalisations per life year | ||||
| Not at risk | 1.00 (Ref) | < 0.001 | 1.00 (Ref) | < 0.001 |
| At risk group | 2.96 (2.15–4.09) | 2.73 (1.98–3.77) | ||
| Number of ED attendances per life year | ||||
| Not at risk | 1.00 (Ref) | < 0.001 | 1.00 (Ref) | 0.001 |
| At risk group | 1.58 (1.30–1.92) | 1.42 (1.16–1.73) | ||
| Number of recorded appointments per life year | ||||
| Not at risk | 1.00 (Ref) | < 0.001 | 1.00 (Ref) | < 0.001 |
| At risk group | 1.35 (1.28–1.43) | 1.28 (1.21–1.36) | ||
Abbreviations: aIRR, adjusted incidence risk ratio; aMOR, adjusted multinominal odds ratio; IRR, incidence risk ratio; MOR, multinominal odds ratio.
Multivariate analysis adjusted for age and sex.
4. Discussion
This single‐centre study highlights significant disparities in the severity and consequent management of AD among paediatric patients, emphasising the impact of psychosocial determinants of health. Nearly one‐third of the study cohort (n = 1480) belonged to at least one of the at‐risk groups described (n = 454). Children in these groups exhibited more severe disease and had higher rates of hospital admissions as well as increased use of oral antibiotics and prednisolone for AD management, compared to their peers. Specifically, 43.8% of the at‐risk population had severe AD, in contrast to 28.3% in the not‐at‐risk group. These findings underscore the need for targeted interventions and resource allocation to address the healthcare barriers faced by certain Australian populations, aiming to improve AD outcomes for at‐risk children.
Intersectionality offers a valuable framework for understanding how multiple psychosocial factors can interact to compound disease severity and complicate management [13]. In this study, some children in the at‐risk cohort belonged to more than one of the defined groups, indicating that overlapping challenges may affect healthcare access and disease management. Addressing these intersecting determinants is essential for developing comprehensive, equity‐focused strategies that meet the complex needs of these populations.
4.1. First Nations Peoples
Of the study population (n = 1480), 1.5% identified as First Nations Peoples, similar to state census reports (1.0% of Victorians) [14]. It has recently been found that there is a high burden of AD and bacterial skin infections among First Nations children presenting to urban Aboriginal Community Controlled Health Organisations [6, 15]. Data for other Australian Communities remain lacking. While AD morphology and disease burden in First Nations children remain poorly described, studies suggest that symptoms for these children are more severe compared with their non‐First Nations peers, possibly due to some degree of under‐recognition and under‐treatment by health practitioners [4, 15]. Comorbidities in some communities, including scabies and pyogenic infections, can further compromise timely diagnoses and treatments [4, 15]. The burden of disease for remote‐living First Nations Australians is unknown, but it is likely under‐estimated and with worse outcomes compared with urban populations given reduced access to healthcare and systemic health inequities [4, 16]. More research is needed to clarify the prevalence, severity, and treatment outcomes of AD among different urban and rural First Nations populations.
4.2. CALD Background
Among the study cohort, 6% of children were from a CALD background, with Mandarin, Arabic and Vietnamese the most common languages. Children from CALD communities, particularly refugees or immigrants, often face barriers to healthcare such as language and cultural differences, financial constraints, and lower health system literacy [17]. The Australian Institute of Health and Welfare (AIHW) has identified major health data gaps for these children [18].
Cultural factors can hinder research engagement and health‐seeking behaviour, limiting accurate burden‐of‐disease data [17]. Research on AD in Australian ethnic minorities remains scarce, and the burden of disease in this group is likely underappreciated [4]. Immigration continues to outpace emigration, and global studies suggest higher AD prevalence in Asian and Black skin compared to white skin, suggesting a possible future increase in AD burden in Australia [19, 20]. Over 30% of the Australian population was born overseas—most commonly from India, China, the Philippines and the UK [20]. Language and cultural differences may also lead to challenges navigating healthcare systems or reluctance to use certain treatments [4, 17]. Further research is needed to understand and address these barriers in CALD communities.
4.3. Rural Communities
Over 1.2 million people in rural Victoria lack ready access to dermatology services [4, 21]. Many must travel hours to see a specialist, often after delays accessing GPs with dermatology expertise [22]. In our study, 14.3% of children (n = 1480) lived in regional, rural or remote areas, reflecting the relevance of these access issues. While GPs can manage most AD cases with education, emollients, and topical anti‐inflammatories, moderate to severe cases may require systemic therapies or phototherapy, which are often unavailable locally [23]. Barriers such as long travel, financial strain, and time away from work may delay specialist care, worsening disease and increasing reliance on systemic treatment. These findings emphasise the need for improved teledermatology, outreach clinics, and rural provider training to ensure equitable AD care across geographic locations [22, 24].
4.4. Living in a Socioeconomically Disadvantaged Postcode
Socioeconomic status (SES) plays a critical role in shaping the health outcomes of children [10, 25]. In this study, 8% of the cohort lived in a SD area. Financial constraints often create barriers to accessing healthcare for early diagnosis, treatment, and better long‐term disease control [10]. Furthermore, children and adolescents from lower SES backgrounds are two to three times more likely to develop mental health problems than their peers with high SES; this can further complicate AD management, exacerbate disease severity, and compromise health outcomes [26]. These findings underscore the importance of addressing socioeconomic disparities [25]. Targeted healthcare interventions are needed to reduce these inequities and ensure equitable access to effective AD management for all children.
4.5. Neuropsychiatric Co‐Morbidities
AD is frequently associated with concurrent neuropsychiatric conditions such as ADHD, ASD, anxiety, and depression [2, 3]. Despite the growing recognition of the link between mental health and AD, research on how these co‐morbidities impact AD outcomes remains limited. Poor mental health can increase systemic inflammation, reduce treatment adherence, and hinder self‐care efforts, all of which may worsen AD outcomes [3, 27]. An analysis of specific mental health and neurodevelopmental problem subgroups in relation to AD outcomes is beyond the scope of this paper but is warranted. A better understanding of particular barriers to effective AD management for these diverse children and adults is also needed. Additionally, better integration of mental health care into dermatologic treatment plans is warranted.
4.6. Vulnerable Children
Vulnerable children—especially those in out‐of‐home care or exposed to abuse—face increased risk of adverse health outcomes, including AD [28, 29]. In this study, 2.8% were identified as “Vulnerable” by healthcare workers. Concerningly, the number of infants entering out‐of‐home care in Australia continues to rise [29]. I In 2021–2022, around 1 in 32 Australian children were involved with the child protection system, including investigations and placements [30]. These children often experience multiple, intersecting inequities—such as limited healthcare access, caregiver strain, and trauma—that contribute to poorer outcomes [29]. Our findings highlight the need for targeted interventions that support both children and caregivers in managing AD effectively.
5. Limitations
The limitations of this study include its conduct at a single urban tertiary centre, which may not represent other Australian contexts, particularly rural or remote areas where access to healthcare services may differ. There is a potential for selection bias, as the children included were those who had access to primary care and were referred and could access specialist input. As such, this study is not able to estimate the prevalence of AD in certain populations with limited access to healthcare services. Additionally, the heterogeneity within the six at‐risk groups means that not all individuals may be at equal risk for health inequities. Furthermore, the retrospective design posed challenges, as EASI scores and SCORAD assessments for AD severity were not consistently recorded. This reliance on assessor‐dependent documentation may affect the accuracy of AD severity measurements and management decisions.
6. Conclusion
This study supports the hypothesis that psychosocial factors contribute to health disparities in the severity and management of AD. Our findings highlight six groups that require greater attention and resource allocation, as they face multiple barriers to accessing the comprehensive healthcare services needed for managing moderate‐to‐severe AD [8]. Addressing these disparities is crucial for improving outcomes. While differences in treatment patterns and responses among at‐risk groups were beyond the scope of this study, further research on these factors is necessary to identify potential prescribing biases and optimise personalised therapies. Multi‐centre studies across diverse healthcare settings are also needed to validate the respective roles of psychosocial factors in determining AD severity and outcomes in paediatric populations.
Author Contributions
A.C. led the project, including study design input, data collection, analysis interpretation and writing of the original manuscript draft. E.S. conducted the statistical analysis. A.Y. and J.C.S. contributed to the study design, research conception and provided critical manuscript review. A.J.L. assisted with study methodology, design and critical evaluation of the manuscript. C.W. contributed to the critical review of the article. All authors reviewed and approved the final version of the manuscript and agree to be accountable for all aspects of the work.
Ethics Statement
Ethical approval for this study was obtained from the RCH Research Ethics and Governance Office, including approval by an appropriate local First Nations representative. MCC DERP Reference number: 3640.
Conflicts of Interest
A.J.L. has received an investigator‐initiated grant from GlaxoSmithKline (GSK), Sanofi Regeneron and Pfizer for unrelated research, and investigational product (EpiCeram) free of charge from Primus Pharmaceuticals for use in unrelated research. J.C.S. has received grants/research funding for his role as an investigator for AbbVie, Amgen, ASLAN, AstraZeneca, Bristol Myers Squibb, Eli Lilly, Galderma, Janssen, Mayne, Novartis, Pfizer Inc., Pierre Fabre, and Sanofi; has received honoraria for serving on advisory boards for Eli Lilly, GSK, Janssen, LEO Pharma, L'Oréal, Novartis, Pfizer Inc., and Sanofi; and has received honoraria for serving as a speaker for Bioderma, Ego Pharmaceuticals, Pfizer Inc., and Pierre Fabre.
Acknowledgements
The authors have nothing to report. Open access publishing facilitated by The University of Melbourne, as part of the Wiley ‐ The University of Melbourne agreement via the Council of Australian University Librarians.
Funding: This work was supported by Pfizer (70071713).
Data Availability Statement
The authors confirm that the data supporting the findings of this study are available within the article.
References
- 1. Langan S. M., Mulick A. R., Rutter C. E., et al., “Trends in Eczema Prevalence in Children and Adolescents: A Global Asthma Network Phase I Study,” Clinical and Experimental Allergy 53, no. 3 (2023): 337–352. [Google Scholar]
- 2. Rønnstad A. T. M., Halling‐Overgaard A. S., Hamann C. R., Skov L., Egeberg A., and Thyssen J. P., “Association of Atopic Dermatitis With Depression, Anxiety, and Suicidal Ideation in Children and Adults: A Systematic Review and Meta‐Analysis,” Journal of the American Academy of Dermatology 79, no. 3 (2018): 448–456.e30. [DOI] [PubMed] [Google Scholar]
- 3. Courtney A. and Su J. C., “The Psychology of Atopic Dermatitis,” Journal of Clinical Medicine 13, no. 6 (2024): 1602. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Courtney A., Lopez D. J., Lowe A. J., Holmes Z., and Su J. C., “Burden of Disease and Unmet Needs in the Diagnosis and Management of Atopic Dermatitis in Diverse Skin Types in Australia,” Journal of Clinical Medicine 12, no. 11 (2023): 3812. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Kuo A., Silverberg N., Fernandez Faith E., Morgan R., and Todd P., “A Systematic Scoping Review of Racial, Ethnic, and Socioeconomic Health Disparities in Pediatric Dermatology,” Pediatric Dermatology 38, no. Suppl 2 (2021): 6–12. [DOI] [PubMed] [Google Scholar]
- 6. Williams C., Hunt J., Kern J. S., and Dunn R., “A Casemix Study of Patients Seen Within an Urban Aboriginal Health Service Dermatology Clinic Over a Five‐Year Period,” Australasian Journal of Dermatology 62, no. 3 (2021): 331–335. [DOI] [PubMed] [Google Scholar]
- 7. Australian Institute of Health and Welfare , Rural and Remote Health [Internet] (Australian Institute of Health and Welfare, 2024). [Google Scholar]
- 8. https://www.health.gov.au/sites/default/files/documents/2021/12/national‐preventive‐health‐strategy‐2021‐2030_1.pdf.
- 9. Vulnerable Children , “Department of Health. Victoria State Government,” https://www.health.vic.gov.au/populations/vulnerable‐children.
- 10. Australian Bureau of Statistics , Socio‐Economic Indexes for Areas (SEIFA), Australia [Internet] (ABS, 2021). [Google Scholar]
- 11. Hanifin J. M., Baghoomian W., Grinich E., Leshem Y. A., Jacobson M., and Simpson E. L., “The Eczema Area and Severity Index—A Practical Guide,” Dermatitis 33, no. 3 (2022): 187–192. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Faye O., Meledie N'Djong A. P., Diadie S., et al., “Validation of the Patient‐Oriented SCORing for Atopic Dermatitis Tool for Black Skin,” Journal of the European Academy of Dermatology and Venereology 34, no. 4 (2020): 795–799. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Al‐Faham H., Davis A. M., and Ernst R., “Intersectionality: From Theory to Practice,” Annual Review of Law and Social Science 15, no. 1 (2019): 247–265. [Google Scholar]
- 14. Australian Bureau of Statistics , Victoria: Aboriginal and Torres Strait Islander Population Summary [Internet] (ABS, 2022). [Google Scholar]
- 15. Ricciardo B. M., Kessaris H. L., Kumarasinghe P., Carapetis J. R., and Bowen A. C., “The Burden of Atopic Dermatitis and Bacterial Skin Infections Among Urban‐Living Indigenous Children and Young People in High‐Income Countries: A Systematic Review,” Pediatric Dermatology 40, no. 1 (2023): 35–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Haggett M. G. and Tait C. P., “Five Years of Dermatological Disease Requiring Specialist Care in the Kimberley Region of Western Australia,” Australasian Journal of Dermatology 62, no. 2 (2021): 157–161. [DOI] [PubMed] [Google Scholar]
- 17. Khatri R. B. and Assefa Y., “Access to Health Services Among Culturally and Linguistically Diverse Populations in the Australian Universal Health Care System: Issues and Challenges,” BMC Public Health 22, no. 1 (2022): 880. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Australian Institute of Health and Welfare , Australia's Children. Cat. no. CWS. 69 (AIHW, 2020). [Google Scholar]
- 19. Kaufman B. P., Guttman‐Yassky E., and Alexis A. F., “Atopic Dermatitis in Diverse Racial and Ethnic Groups‐Variations in Epidemiology, Genetics, Clinical Presentation and Treatment,” Experimental Dermatology 27, no. 4 (2018): 340–357. [DOI] [PubMed] [Google Scholar]
- 20. “Australia's Population by Country of Birth,” https://www.abs.gov.au/statistics/people/population/australias‐population‐country‐birth/latest‐release.
- 21. Australian Institute of Health and Welfare , Health Workforce [Internet] (Australian Institute of Health and Welfare, 2024), https://www.aihw.gov.au/reports/workforce/health‐workforce. [Google Scholar]
- 22. Whiting G., Magin P., Morgan S., et al., “General Practice Trainees' Clinical Experience of Dermatology Indicates a Need for Improved Education: A Cross‐Sectional Analysis From the Registrar Clinical Encounters in Training Study,” Australasian Journal of Dermatology 58, no. 4 (2017): e199–e206. [DOI] [PubMed] [Google Scholar]
- 23. Goh M. S., Yun J. S., and Su J. C., “Management of Atopic Dermatitis: A Narrative Review,” Medical Journal of Australia 216, no. 11 (2022): 587–593. [DOI] [PubMed] [Google Scholar]
- 24. Smith S. D., Lee A., Blaszczynski A., and Fischer G., “Pharmacists' Knowledge About Use of Topical Corticosteroids in Atopic Dermatitis: Pre and Post Continuing Professional Development Education,” Australasian Journal of Dermatology 57, no. 3 (2016): 199–204. [DOI] [PubMed] [Google Scholar]
- 25. Gupta R. P., de Wit M. L., and McKeown D., “The Impact of Poverty on the Current and Future Health Status of Children,” Paediatrics & Child Health 12, no. 8 (2007): 667–672. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Reiss F., Meyrose A. K., Otto C., Lampert T., Klasen F., and Ravens‐Sieberer U., “Socioeconomic Status, Stressful Life Situations and Mental Health Problems in Children and Adolescents: Results of the German BELLA Cohort‐Study,” PLoS One 14, no. 3 (2019): e0213700. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Farzanfar D., Dowlati Y., French L. E., Lowes M. A., and Alavi A., “Inflammation: A Contributor to Depressive Comorbidity in Inflammatory Skin Disease,” Skin Pharmacology and Physiology 31, no. 5 (2018): 246–251. [DOI] [PubMed] [Google Scholar]
- 28. Lima F., Taplin S., Maclean M., and O'Donnell M., “Infants Entering Out‐Of‐Home Care: Health, Developmental Needs and Service Provision,” Child Abuse & Neglect 149 (2024): 106577. [DOI] [PubMed] [Google Scholar]
- 29. Cashmore J. and Wulczyn F., “Pathways of Care: A Longitudinal Study of Children in Care in Australia: Introductory Article for Special Issue on Pathways of Care Longitudinal Study,” Child Abuse & Neglect 149 (2024): 106586. [DOI] [PubMed] [Google Scholar]
- 30. Australian Institute of Health and Welfare , Child Protection Australia 2021–22 [Internet] (Australian Institute of Health and Welfare, 2024). [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The authors confirm that the data supporting the findings of this study are available within the article.