Abstract
Hand eczema (HE) is a common skin disease, diagnosed through clinical evaluation. However, the increasing need to deliver healthcare remotely, along with the demand for valid and reliable evaluations of HE in research, prompts exploration of the potential benefits of remote assessment. This study assessed the criterion validity and inter- and intra-rater reliability of patient-provided photographs to diagnose and assess the severity of HE. The presence and severity of HE were assessed by 1 of the 2 clinical raters. Following instructions, patients took 4 photographs of their hands. All photographs were assessed twice by the clinical raters and 2 additional raters. In total, 98 patients were included. The sensitivity, specificity, positive and negative predictive value were respectively 85.1%, 79.2%, 92.6%, and 63.3%. Kappa values of 0.35 and 0.66 were found for inter- and intra-rater reliability. The intraclass correlation coefficients based on the Hand Eczema Severity Index (HECSI) were > 0.90. Kappa values for the criterion validity and inter- and intra-rater reliability using the photographic guide were respectively 0.71, 0.58–0.77, and 0.77. The findings indicate that patient-provided photographs are valid and reliable for remote assessment of HE and could be of benefit in healthcare and research by reducing in-person visits and enabling continuous monitoring.
Key words: hand dermatitis, hand eczema, telemedicine, teledermatology, reliability and validity, remote assessment
SIGNIFICANCE
The increasing need to deliver healthcare remotely prompts exploration of the potential benefits of remote assessment of hand eczema. This study found that 85% of patients with hand eczema could be adequately diagnosed using patient-provided photographs, and 79% without hand eczema could be correctly identified as not having the disease. The findings of this study also indicate that patient-provided photographs can be used as a valid and reliable method for severity assessment, using the Hand Eczema Severity Index (HECSI) and the photographic guide. Remote assessment may add value in healthcare and research by reducing in-person visits and enabling continuous monitoring.
Hand eczema (HE) is a common skin disease with a self-reported lifetime and 1-year prevalence of respectively 15% and 9% within the general population (1). However, HE is diagnosed through clinical evaluation, and the diagnostic work-up includes thorough history-taking and examination (2). At the same time, there has been a growing need to deliver healthcare remotely (3). Remote assessment can contribute to improved accessibility to healthcare resources, reduced travel burden, cost savings, and enhanced quality of care (4).
Additionally, there is a demand for valid and reliable evaluations of HE in research. However, the prevalence of HE identified by self-reports is in general considered to be underestimated (5–10), along with varying results regarding validity of self-reported severity (11–16). Yet, real-time examination is often not feasible, which prompts the need for alternative methods such as remote assessment.
In dermatology, remote assessment has been demonstrated as a valid and reliable method for evaluating the presence and/or severity of various diseases, such as psoriasis or atopic dermatitis (AD) (17–21). In HE, remote assessment may be particularly applicable, as the hands represent a defined body surface and are relatively simple for patients to photograph. However, studies on the application of remote assessment of HE are limited and have primarily focused on assessments not involving patient-provided photographs (22–26). Therefore, the aim of this study was to assess the criterion validity and inter- and intra-rater reliability of using patient-provided photographs to diagnose and to assess the severity of HE.
MATERIALS AND METHODS
Design, setting, and study population
This validation study was conducted at the dermatology outpatient clinic at the University Medical Center Groningen (UMCG) in The Netherlands. Adult patients with and without HE with a scheduled visit were eligible. To allow for subgroup analyses for different severity categories (almost clear, moderate, and severe to very severe), an overall equal distribution across these categories was ensured. Exclusion criteria were lack of proficiency in Dutch and inability to photograph both the right and the left hand. Other skin diseases (of the hands) did not serve as exclusion criteria. All patients gave written informed consent. The study was approved by the central ethics review board of the UMCG as a non-interventional study (ref 11092).
Clinical assessment
At the outpatient clinic, all patients were assessed by 1 of the 2 clinical raters (MLAS, LL), which is considered the gold standard (Fig. 1). Both clinical raters were physicians and highly experienced in the assessment of HE (Table SI). They used a standardized form to assess the presence, clinical subtype, and severity of HE. In accordance with clinical practice, clinical raters could ask for clarification on whether potentially temporary skin lesions were caused by specific activities, such as gardening, and on the duration of these lesions’ presence. The clinical subtype was categorized according to the guidelines by Thyssen et al. (2), expanded by 2 categories (“chronic dry fissured HE” and “not classifiable”) based on the guidelines from the Danish Contact Dermatitis Group (27). The severity was based on the Hand Eczema Severity Index (HECSI), a validated instrument taking into account the extent and intensity of clinical signs of HE, ranging from 0–360 points, and the validated photographic guide by Coenraads et al, including 4 groups of photographs representing almost clear, moderate, severe, and very severe HE (28, 29). A standardized instruction guide on how to use the HECSI was provided (Appendix S1). Other characteristics collected were sex, age, and Fitzpatrick skin type (30). All data were entered into the Research Electronic Data Capture tool (REDCap) (31).
Fig. 1.
Overview of the study. Criterion validity for diagnosing hand eczema was based on the first photographic assessment of the identical rater who conducted the clinical assessment. Inter- and intra-rater reliability were based on the photographic assessment of all 4 raters combined. Severity of hand eczema was assessed by using the Hand Eczema Severity Index (HECSI) (28) and the photographic guide by Coenraads et al. (29).
Photography of the hands
At the outpatient clinic, participants took 4 photographs of their hands (left/right, dorsal/palmar), following standardized written and visualized instructions (Appendix S2). The instructions had been tested for understandability among patients with HE, and included recommendations from a photographer experienced in clinical photography. As the study sought to replicate a scenario in which patients take photographs remotely, participants were encouraged to capture the photographs independently. However, if they failed to adhere to the written instructions, additional verbal instructions were provided by the study coordinator (MJB). The reasons for this were documented to provide insight into underlying causes of non-compliance with the instructions. All photographs were taken in a well-lit room at the outpatient clinic, by using a smartphone (iPhone 14 Pro Max [48 megapixels]) (32). All photographs were taken in an upright position, without using flashlight or zooming in or out. Photographs were taken against a blue screen, with a small distance between the hand and the screen to minimize the appearance of shadows. Participants were asked to spread and stretch out the fingers, and make sure that the fingertips and wrists were visible. To minimize the risk of recall bias among raters and to ensure that all parts of the hands were clearly visible, participants were asked to roll up their sleeves and remove jewellery, watches, or other identifiable objects.
Photographic examination
Each set of 4 photographs was assessed twice and independently by the 2 clinical raters, and by 2 additional raters experienced in the assessment of HE (MLAS, LL, EW, KP) (see Table SI). To prevent recall bias, the photographic assessments took place at least 4 weeks after the clinical assessment, and with an interval of at least 4 weeks in between. A standardized form was used to assess the presence of HE and the severity (HECSI and the photographic guide) (28, 29). Prior to the study, a practice session was organized to standardize scoring of the HECSI. The raters received a total of at maximum 25 sets of photographs each time, in a random order.
Sample size calculations
Calculation of the required sample size was performed with power analysis and sample size software (PASS) (33). This calculation was based on the assumed sensitivity and specificity for diagnosing HE using patient-provided photographs, including stratification by severity category based on the photographic guide. In addition, the assumed criterion validity and inter-rater and intra-rater reliability for the severity assessment based on the patient-provided photographs was taken into account (based on intraclass correlation coefficient (ICC) values) (Appendix S3). Based on the assumption of 20% dropout, for example, due poor quality of photographs, a sample size of at least 93 patients would be required, including 24 participants without HE and 69 patients with HE (almost clear: 23; moderate: 23; severe to very severe: 23).
Statistical analysis
Data analyses were performed using the Statistical Products and Service Solutions package version 28 (IBM Corp, Armonk, NY, USA). Criterion validity was assessed by comparing the diagnosis and severity of HE based on patient-provided photographs with the clinical examination (34). This comparison was based on the first round of photographic assessment performed by the identical rater who conducted the clinical assessment. Inter-rater reliability was defined as the degree to which different raters provided consistent assessments based on the photographs (34). Intra-rater reliability was defined as the consistency of a single rater’s assessments based on photographs when evaluated on different occasions (34). Both inter- and intra-rater reliability were based on the photographic assessment of all 4 raters combined. Intra-rater reliability was additionally assessed for each photographic rater separately.
An overview of the statistical procedures conducted is shown in Table I. In the present study, an ICC value of < 0.50 indicates poor, 0.50–0.75 moderate, 0.75–0.90 good, and ≥ 0.90 excellent agreement (35). Kappa values < 0.2 indicate slight, 0.2–0.4 fair, 0.4–0.6 moderate, 0.6–0.8 substantial, and ≥ 0.8 almost perfect agreement (36).
Table I.
Overview of statistical tests used to assess the criterion validity and inter- and intra-rater reliability for diagnosing hand eczema and assessing the severity based on patient-provided photographs
| Item | Diagnosis of hand eczema | Severity of hand eczema | |
|---|---|---|---|
| Hand Eczema Severity Indexa | Photographic guideb | ||
| Criterion validity | Sensitivity (%) Specificity (%) Positive predictive value (%) Negative predictive value (%) |
Intra-class correlation coefficient (ICC) (two-way random; absolute agreement; average measures) Correlation plot Bland–Altman plot |
Weighted kappa Cross-table |
| Inter-rater reliability | Percentage of agreement (%) Fleiss’s kappa |
Intra-class correlation coefficient (ICC) (two-way random; absolute agreement; average measures) Standard error of measurement (SEM) |
Percentage of agreement (%) Weighted kappa Intra-class correlation coefficient (ICC) (two-wayrandom; absolute agreement; average measures) |
| Intra-rater reliability | Percentage of agreement (%) Cohen’s kappa |
Intra-class correlation coefficient (ICC) (two-way mixed; absolute agreement; average measures) Standard error of measurement (SEM) |
Percentage of agreement (%) Weighted kappa Intra-class correlation coefficient (ICC) (two-way mixed; absolute agreement; average measures) |
RESULTS
Study population
In total, 99 patients were eligible for inclusion. Due to an incomplete set of photographs, 1 patient was excluded, leaving a total study population of n = 98 (Table II). The mean age of the included patients was 44.0 ± 17.4 years. More than half were male (55.1%) and most patients had skin type I–III (92.9%). A total of 74 patients (75.5%) were diagnosed with HE, with a mean HECSI score of 46.8 ± 46.0. The majority had chronic dry fissured HE (68.9%). During the photography, most patients receiv-ed additional instructions regarding required distance between the blue screen and the hands (46.9%).
Table II.
Baseline characteristics of total study population, stratified for patients with hand eczema and without hand eczema
| Factor | Total study population (n = 98) n (%) | Hand eczema (n = 74) n (%) | No hand eczema (n = 24) n (%) | p-value |
|---|---|---|---|---|
| Age, years (mean ± SD) | 44.0 ± 17.4 | 43.7 ± 17.1 | 44.9 ± 18.3 | 0.95 |
| Age groups | ||||
| 18–25 years | 15 (15.3) | 12 (16.2) | 3 (12.5) | 1.00 |
| 25–35 years | 24 (24.5) | 17 (23.0) | 7 (29.2) | 0.54 |
| 35–45 years | 12 (12.2) | 10 (13.5) | 2 (8.3) | 0.73 |
| 45–55 years | 11 (11.2) | 7 (9.5) | 4 (16.7) | 0.46 |
| 55–65 years | 23 (23.5) | 19 (25.7) | 4 (16.7) | 0.42 |
| > 65 years | 13 (13.3) | 9 (12.2) | 4 (16.7) | 0.73 |
| Sex | ||||
| Males | 54 (55.1) | 40 (54.1) | 14 (58.3) | 0.71 |
| Females | 44 (44.9) | 34 (45.9) | 10 (41.7) | |
| Skin typea | ||||
| Type I–III | 91 (92.9) | 69 (93.2) | 22 (91.7) | 1.00 |
| Type IV–VI | 7 (7.1) | 5 (6.8) | 2 (8.3) | |
| Diagnosis of hand eczema | 74 (75.5) | 74 (100.0) | – | – |
| Clinical subtype of hand eczemab | ||||
| Hyperkeratotic hand eczema | 7 (7.1) | 7 (9.5) | – | – |
| Acute recurrent vesicular hand eczema | 13 (13.3) | 13 (17.6) | – | – |
| Pulpitis | 3 (3.1) | 3 (4.1) | – | – |
| Nummular hand eczema | 0 (0.0) | 0 (0.0) | – | – |
| Chronic dry fissured hand eczemac | 51 (52.0) | 51 (68.9) | – | – |
| Not classifiable | 0 (0.0) | 0 (0.0) | – | – |
| Severity of hand eczema (HECSI), mean ± SDd,e | 46.8 ± 46.0 | 46.8 ± 46.0 | – | – |
| Severity of hand eczema (photographic guide)f | ||||
| Almost clear | 25 (25.5) | 25 (33.8) | – | – |
| Moderate | 25 (25.5) | 25 (33.8) | – | – |
| Severe to very severe | 24 (24.5) | 24 (32.4) | – | – |
| Additional instructions | ||||
| Removal of jewellery | 11 (11.2) | 7 (9.5) | 4 (16.7) | – |
| Rolling up sleeves | 8 (8.2) | 4 (5.4) | 4 (16.7) | – |
| Hands not completely in front of blue screen | 11 (11.2) | 8 (10.8) | 3 (12.5) | – |
| Not enough distance between blue screen and hand | 46 (46.9) | 30 (40.5) | 16 (66.7) | – |
| Fingers not spread | 7 (7.1) | 6 (8.1) | 1 (4.2) | – |
| Fingers not stretched out | 6 (6.0) | 5 (6.8) | 1 (4.2) | – |
| Smartphone not held in upright position | 3 (3.1) | 3 (4.1) | 0 (0.0) | – |
| Hands not square to camera | 20 (20.4) | 17 (23.0) | 3 (12.5) | – |
| Fingers not visible | 31 (31.6) | 24 (32.4) | 7 (29.2) | – |
| Wrists not visible | 38 (38.8) | 30 (40.5) | 8 (33.3) | – |
| Incorrect number of pictures | 8 (8.2) | 5 (6.8) | 3 (12.5) | – |
| Too much distance between camera and hands | 1 (1.0) | 1 (1.4) | 0 (0.0) | – |
| Made pictures from lateral/medial side hands | 2 (2.0) | 2 (2.7) | 0 (0.0) | – |
| Needed assistance | 12 (12.2) | 8 (10.8) | 4 (16.7) | – |
Skin type was based on published work by Fitzpatrick et al. (30).
Classification of clinical subtype was based on published work by Thyssen et al. and Johansen et al. (2,27).
Chronic dry fissured hand eczema refers to the clinical subtype of hand eczema as described by Johansen et al. (27), rather than the duration of hand eczema.
Severity assessment was based on the Hand Eczema Severity Index (HECSI) by Held et al. (28).
Only among patients with hand eczema.
Severity assessment was based on the photographic guide by Coenraads et al. (29).
n: number; SD: standard deviation.
Diagnosis of hand eczema
Criterion validity. The sensitivity, specificity, positive and negative predictive value of diagnosing HE based on photographs were respectively 85.1%, 79.2%, 92.6%, and 63.3% (Table III). All patients who were incorrectly classified as not having HE actually had almost clear HE, according to the clinical assessment. When focussing on patients with almost clear HE, a sensitivity of 56.0% was found. The sensitivity for moderate and severe-to-very severe HE was 100%.
Table III.
Criterion validity and inter- and intra-rater reliability for diagnosing hand eczema based on patient-provided photographs
| Criterion validitya | Inter-rater reliabilityb | Intra-rater reliabilityb | ||||||
|---|---|---|---|---|---|---|---|---|
| Sensitivity, % (95% CI) | Specificity, % (95% CI) | Positive predictive value, % (95% CI) | Negative predictive value, % (95% CI) | Agreement, % (95%CI) | Fleiss’s kappac | Agreement, % (95%CI) | Cohen’s kappac | |
| Total | 85.1 (77.0–93.2) | 79.2 (63.0–95.4) | 92.6 (86.4–98.8) | 63.3 (46.1–80.6) | 63.3 (53.8–72.8) | 0.35 (0.27–0.43) | 88.8 (85.6–91.8) | 0.66 (0.57–0.76) |
| Severity of hand eczema (photographic guide)d | ||||||||
| Almost clear | 56.0 (36.5–75.5) | – | – | – | 44.0 (24.5–63.5) | 0.21 (0.05–0.37) | 82.0 (74.5–89.5) | 0.54 (0.36–0.73) |
| Moderate | 100.0 (86.0–100.0) | – | – | – | 100.0 (86.0–100.0) | – | 98.0 (95.3–100.0) | – |
| Severe to very severe | 100.0 (86.0–100.0) | – | – | – | 100.0 (86.0–100.0) | – | 97.9 (95.0–100.0) | – |
Criterion validity is based on the first photographic assessment conducted by the identical rater who conducted the clinical assessment.
Inter- and intra-rater reliability are based on the photographic assessments of all raters combined.
For the categories of moderate and severe to very severe hand eczema, there was no variability in the ratings for these categories. Fleiss’s and Cohen’s kappa could therefore not be calculated.
Severity assessment was based on the photographic guide by Coenraads et al. (29).
CI: confidence interval.
Inter- and intra-rater reliability. Fair agreement among all photographic raters was found regarding the diagnosis of HE based on photographs. In addition, fair agreement was found when focusing only on patients with almost clear HE. For cases with moderate and severe to very severe HE, agreement was 100.0%. Substantial intra-rater agreement was found. When looking at patients with almost clear HE, moderate intra-rater agreement was found. The intra-rater agreement was 98.0% and 97.9% among patients with moderate and severe-to-very severe HE separately. The intra-rater reliability by all 4 photographic raters separately ranged from 84.7–90.8% (Table SII).
Severity of hand eczema (HECSI)
Criterion validity. For criterion validity, the use of the HECSI to assess the severity of HE based on patient-provided photographs demonstrated excellent agreement (Table IV). The analyses including patients with almost clear and moderate HE separately showed respectively good and moderate agreement. Good agreement was found among patients with severe to very severe HE only. The correlation between the clinical and photographic assessment is presented in Fig. 2A, and visualized with a Bland–Altman plot in Fig. 2B.
Table IV.
Criterion validity and inter- and intra-rater reliability for severity assessment of hand eczema based on patient-provided photographs using the Hand Eczema Severity Indexa and the photographic guideb
| Hand Eczema Severity Index (HECSI)a | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| Criterion validityc ICC (95%CI) | Inter-rater reliabilityd | Intra-rater reliabilityd | |||||||
| SEM | ICC (95%CI) | SEM | ICC (95% CI) | ||||||
| Total | 0.92 (0.81–0.96) | 10.0 | 0.93 (0.88–0.96) | 7.3 | 0.96 (0.95–0.97) | ||||
| Stratified by photographic guide severity categoriesb | |||||||||
| Almost clear | 0.75 (0.27–0.92) | 4.9 | 0.65 (0.22–0.89) | 4.6 | 0.79 (0.66–0.87) | ||||
| Moderate | 0.69 (0.13–0.88) | 7.0 | 0.81 (0.63–0.91) | 5.7 | 0.86 (0.79–0.91) | ||||
| Severe to very severe | 0.86 (0.60–0.94) | 15.9 | 0.86 (0.68–0.94) | 11.8 | 0.92 (0.88–0.95) | ||||
| Clinical signs HECSI | |||||||||
| Erythema | 0.87 (0.77–0.93) | 0.9 | 0.88 (0.80–0.93) | 0.9 | 0.90 (0.88–0.92) | ||||
| Papulation/induration | 0.79 (0.38–0.90) | 1.0 | 0.85 (0.76–0.90) | 1.0 | 0.85 (0.82–0.88) | ||||
| Vesicles | 0.80 (0.68–0.88) | 0.7 | 0.88 (0.82–0.93) | 0.6 | 0.88 (0.84–0.90) | ||||
| Fissures | 0.82 (0.70–0.89) | 0.7 | 0.82 (0.72–0.89) | 0.6 | 0.88 (0.85–0.91) | ||||
| Scaling | 0.86 (0.75–0.92) | 1.1 | 0.79 (0.56–0.89) | 0.7 | 0.91 (0.89–0.93) | ||||
| Oedema | 0.78 (0.63–0.87) | 0.9 | 0.82 (0.71–0.89) | 0.6 | 0.91 (0.88–0.93) | ||||
| Photographic guideb | |||||||||
| Criterion validityc Weighted kappa (95% CI) | Inter-rater reliabilityd | Intra-rater reliabilityd | |||||||
| Agreement, % (95% CI) | Weighted kappa (95% CI) | ICC (95%CI) | Agreement, % (95% CI) | Weighted kappa (95% CI) | ICC (95% CI) | ||||
| Total | 0.71 (0.58–0.84) | 47.5 (35.0–60.0) | 0.58–0.77 | 0.91 (0.87–0.94) | 80.5 (75.9–85.1) | 0.77 (0.72–0.83) | 0.92 (0.89–0.93) | ||
Severity assessment was based on the Hand Eczema Severity Index (HECSI) by Held et al. (28).
Severity assessment was based on the photographic guide by Coenraads et al. (29).
Criterion validity is based on the first photographic assessment conducted by the identical rater who conducted the clinical assessment.
Inter- and intra-rater reliability is based on the photographic assessments of all raters combined. Weighted kappa values for inter-rater reliability represent the range of agreement scores from the 4 separate pairwise comparisons between raters.
CI: confidence interval; ICC: intra-class correlation coefficient; SEM: standard error of measurement.
Fig. 2.
Criterion validity of patient-provided photographs for assessing hand eczema severity using the Hand Eczema Severity Index (HECSI) (28). (A) Correlation between clinical and first photographic severity assessment of hand eczema; (B) Bland–Altman plot showing relationship between the differences between clinical and first photographic assessment of hand eczema severity.
Inter- and inter-rater reliability. Photographic severity based on the HECSI showed excellent inter- and intra-rater reliability. For the different categories of HE severity the ICC value representing inter-rater reliability ranged from 0.65–0.86, and from 0.79–0.92 for intra-rater reliability. The intra-rater reliability by all 4 photographic raters separately ranged from 0.91–0.97 (Table SIII).
Severity of hand eczema (photographic guide)
Criterion validity. Substantial agreement was found for the severity assessment using the photographic guide (see Table IV). Of the patients with almost clear HE, according to the clinical assessment, 64.3% were correctly classified as having almost clear HE based on the photographic assessment (Table SIV). From the patients with moderate HE, 68.0% were also classified as having moderate HE based on the photographs. Of the patients with severe to very severe HE, a total of 91.7% were classified as having severe to very severe HE based on the photographs.
Inter- and intra-rater reliability. Among all 4 photographic raters agreement was found in 47.5%, with excellent inter-rater reliability. Near perfect agreement, in which at least 3 photographic raters agreed with each other, was found in 81.7%. Weighted kappa values of 0.58–0.77 were found. The agreement between the first and second round of photographic assessments was substantial/excellent. The percentage of intra-rater agreement for all 4 photographic raters separately ranged from 69.6–87.1%, with weighted kappa values of 0.64–0.85 (Table SV).
DISCUSSION
This study evaluated the validity and reliability of diagnosing and assessing the severity of HE based on patient-provided photographs. The findings demonstrated good sensitivity and specificity, and indicate that patient-provided photographs could be a valid and reliable tool for severity assessment using the HECSI and photographic guide.
A few studies have assessed the validity or reliability of remote assessment of HE (22–26); however, they are not directly comparable to the present study, as they did not use patient-provided photographs (22, 23, 25, 26), only included specific populations (e.g., healthcare workers) (22, 23, 25, 26), were limited by small sample sizes (23, 26), did not include or only included a limited number of more severely affected patients (22, 23, 26) or used different instruments (e.g., the Hand Dermatitis Screening Tool [HDST]) (22, 23). Furthermore, substantial variations in study protocols exist (e.g., comparison of the assessments from different raters (22, 24), or focus on worsening vs improvement (26)). Only 2 studies focused on diagnosing HE based on photographs (not patient-provided), using a severity scale as diagnostic tool (the HDST and the photographic guide by Coenraads et al., respectively). In the first study, agreement was obtained regarding diagnosis of HE in about 60% of the participants, with an occupational health nurse conducting in-person assessments and a dermatologist diagnosing based on photographs (22). In the second study, moderate inter-rater agreement was found for the diagnosis of HE, based on photographic assessments conducted by 2 dermatologists (kappa 0.5) (25). Studies on remote assessment using the HECSI showed mixed findings: 1 study reported moderate intra-rater reliability based on photographs from a database containing images of the hands of metal workers (ρ = 0.572) (23), while a second study (using patient-provided photographs), found strong validity (r = 0.898) (24). A third study found that the HECSI may lack intra-rater reliability in detecting subtle skin changes, based on photographs taken by study personnel (26). For the HDST, limited validity was found for severity assessed by a dermatologist using photographs taken by study personnel, compared with clinical assessment by an occupational health nurse (kappa 0.20) (22).
The present study demonstrated a sensitivity of 85.1% and a specificity of 79.2% for diagnosing HE remotely. Notably, all patients with moderate and severe to very severe HE were accurately classified as having HE. Compared with studies using self-reports, which show high variability in sensitivity (7–87%) and specificity (79–100%) (likely due to differences in study populations, e.g., high-risk occupations vs the general population, and different screening tools), the present findings suggest that remote assessment could offer a valuable alternative, particularly because of its high sensitivity (5–10). Although the specificity is somewhat lower, it remains acceptable given that false-positive cases can often be corrected through follow-up assessments, and the specificity remains within the range observed in self-report based methods. In future research, it would be interesting to assess whether the psychometric properties, especially specificity, could be further improved combined with self-reported symptom-based data, such as itching or pain, particularly for individuals with minimal signs that could indicate HE (10, 37).
Additionally, the present study suggests that the severity of HE can be validly and reliably assessed remotely, especially in moderate and severe-to-very severe HE. In line with the present study, 2 studies by Ragamin et al. (2023) and Ali et al. (2022), which assessed the psychometric properties of patient-provided photographs for AD severity, found that severity could be validly and reliably assessed remotely (17, 21). However, in the studies of Ragamin et al. and Ali et al., patients were asked to provide respectively full-body photographs and photographs of lesions selected by a clinical assessor, which may lead to lower validity and reliability when patients must independently decide which lesions to photograph. In contrast, for HE, patients do not need to differentiate between healthy and unhealthy skin, and are less affected by challenges related to hard-to-reach body areas.
In the first place, remote assessment of HE may add value in healthcare. While remote assessment may not be ideal for diagnosing HE in this setting, due to the risk of misidentifying other skin diseases, remote assessment may still be useful for monitoring its fluctuating course. Specifically, in occupational healthcare, remote assessment may support screening programmes, where misclassification is manageable through follow-up.
Second, remote assessment of HE allows for frequent data collection in research, and can serve as an alternative when real-time examination is not feasible. Nevertheless, it should be considered that assessment of large amounts of photographs could be time-consuming. In the future, artificial intelligence (AI) may therefore offer promising alternatives, providing even more reliable estimates of clinical signs (e.g., intensity of erythema) (38, 39).
Strengths and limitations
The major strengths are that the present study assessed criterion validity, alongside inter- and intra-rater reliability of patient-provided photographs, across both diagnosis and severity by using the HECSI and the photographic guide (28, 29). Furthermore, inclusion of an adequate number of patients across different severity categories allowed for sub-analyses. Moreover, raters with varying levels of clinical experience, from 3 different clinical sites in The Netherlands and in Germany, participated. However, some limitations should be taken into account. First, inclusion of patients took place in a tertiary referral centre, which could limit the generalizability of our findings, as this population might be more motivated to ensure high-quality photographs, recognizing potential benefits. Although raters were not strictly randomly selected, they were representative of clinicians assessing HE, supporting the use of the two-way random-effects model while acknowledging some limitation in generalizability. Furthermore, the assessment of the raters may have been influenced by the study’s focus on HE, which could have biased them either towards more readily assigning this diagnosis or, conversely, towards being more critical to avoid overdiagnosis. However, the raters did not receive any feedback on their individual assessments during the study, which may have reduced the risk of systematic bias. In addition, the positive and negative predictive values should be interpreted cautiously, as they depend on disease prevalence, which varies widely across populations and could not be precisely accounted for in our purposive sampling approach. Additionally, limited variability in the data regarding skin type and clinical subtype of HE did not allow for subgroup analyses. Moreover, photographs were taken in a standardized, supervised setting, which may not fully reflect real-world conditions, and the majority of patients needed additional instructions. As most corrections were only minor (e.g., invisibility of a small part of a fingertip), this does not necessarily indicate that the photographs would have been of poor quality otherwise. Nevertheless, health professionals and researchers should be mindful of digital and health literacy challenges, which could not only contribute to difficulties in completing (digital) questionnaires, but also to challenges in taking photographs, and might consider interactive approaches such as instruction videos or real-time digital feedback (40).
Conclusion
The findings of this study indicate that patient-provided photographs are a valid and reliable method to assess the diagnosis and severity of HE. Remote assessment may be of added value in healthcare and research, potentially reducing the need for in-person visits and enabling continuous monitoring. Future research should examine the validity and reliability in different settings, evaluate whether adherence of the instructions on how to take photographs could be improved through interactive approaches, and explore the potential role of AI.
Supplementary Material
ACKNOWLEDGEMENTS
IRB approval status
Reviewed and approved by the Medical Ethical Review Board of the University Medical Center Groningen, reference number: 11092.
Availability of data
Data available on request from the authors.
Conflict of interest disclosures
EW was/is a consultant and/or investigator and/or has received research money from Galderma, German Research Foundation (DFG), German Statutory Insurance (DGUV), Kiniksa, LEO Pharma, Sanofi, and Trevi. KP has received consultancy fees for AbbVie, LEO Pharma, Janssen, and Sanofi Genzyme. MLAS was/is a consultant and/or investigator and/or has received research grants from Sanofi Genzyme, Regeneron Pharmaceuticals, Pfizer, LEO Pharma, Galderma, Amgen, and Incyte. MJB, MMS, UB, and LL have no conflicts of interest to declare.
REFERENCES
- 1.Quaade AS, Simonsen AB, Halling A, Thyssen JP, Johansen JD. Prevalence, incidence, and severity of hand eczema in the general population: a systematic review and meta-analysis. Contact Dermatitis 2021; 84: 361–374. 10.1111/cod.13804 [DOI] [PubMed] [Google Scholar]
- 2.Thyssen JP, Schuttelaar MLA, Alfonso JH, Andersen KE, Angelova-Fischer I, Arents BWM, et al. Guidelines for diagnosis, prevention, and treatment of hand eczema. Contact Dermatitis 2022; 86: 357–378. 10.1111/cod.14035 [DOI] [PubMed] [Google Scholar]
- 3.Loh CH, Chong Tam SY, Oh CC. Teledermatology in the COVID-19 pandemic: a systematic review. JAAD Int 2021; 5: 54–64. 10.1016/j.jdin.2021.07.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Kruse CS, Krowski N, Rodriguez B, Tran L, Vela J, Brooks M. Telehealth and patient satisfaction: a systematic review and narrative analysis. BMJ Open 2017; 7: e016242. 10.1136/bmjopen-2017-016242 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Meding B, Barregård L. Validity of self-reports of hand eczema. Contact Dermatitis 2001; 45: 99–103. 10.1034/j.1600-0536.2001.045002099.x [DOI] [PubMed] [Google Scholar]
- 6.Smit HA, Coenraads PJ, Lavrijsen AP, Nater JP. Evaluation of a self-administered questionnaire on hand dermatitis. Contact Dermatitis 1992; 26: 11–16. 10.1111/j.1600-0536.1992.tb00861.x [DOI] [PubMed] [Google Scholar]
- 7.Bregnhøj A, Søsted H, Menné T, Johansen JD. Validation of self-reporting of hand eczema among Danish hairdressing apprentices. Contact Dermatitis 2011; 65: 146–150. 10.1111/j.1600-0536.2011.01908.x [DOI] [PubMed] [Google Scholar]
- 8.Kovačić J, Franić Z, Bjelajac A, Macan J. Low sensitivity of a self-report questionnaire for hand eczema in hairdressing apprentices. Contact Dermatitis 2024; 91: 506–512. 10.1111/cod.14686 [DOI] [PubMed] [Google Scholar]
- 9.Sedeh FB, Ullum AG, Michaelsdóttir TE, Bang Christensen K, Stückler SG, Jemec GBE, et al. The correlation between self-reported hand eczema and clinically based diagnosis in professional cleaners. Contact Dermatitis 2024; 91: 104–111. 10.1111/cod.14611 [DOI] [PubMed] [Google Scholar]
- 10.Svensson Å, Lindberg M, Meding B, Sundberg K, Stenberg B. Self-reported hand eczema: symptom-based reports do not increase the validity of diagnosis. Br J Dermatol 2002; 147: 281–284. 10.1046/j.1365-2133.2002.04799.x [DOI] [PubMed] [Google Scholar]
- 11.Janstrup AK, Nørreslet LB, Toft-Hansen JM, Ofenloch R, Agner T, Yüksel YT. Assessment of hand eczema severity by patients using the Hand Eczema Severity Index (HECSI). J Eur Acad Dermatol Venereol 2023; 37: 2349–2354. 10.1111/jdv.19329 [DOI] [PubMed] [Google Scholar]
- 12.Van Coevorden AM, Van Sonderen E, Bouma J, Coenraads PJ. Assessment of severity of hand eczema: discrepancies between patient- and physician-rated scores. Br J Dermatol 2006; 155: 1217–1222. 10.1111/j.1365-2133.2006.07531.x [DOI] [PubMed] [Google Scholar]
- 13.Bentz P, Apfelbacher C, Akst W, Molin S, Bauer A, Elsner P, et al. Self-reported versus physician-reported severity of chronic hand eczema: concordance analysis based on data from the German Chronic Hand Eczema Patient Long-Term Management Registry. Acta Derm Venereol 2023; 103. 10.2340/actadv.v103.5383 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Carlsson A, Ganemo A, Anderson CD, Meding B, Stenberg B, Svensson Å. Scoring of hand eczema: good agreement between patients and dermatological staff. Br J Dermatol 2011; 165: 123–128. 10.1111/j.1365-2133.2011.10312.x [DOI] [PubMed] [Google Scholar]
- 15.Cvetkovski RS, Jensen H, Olsen J, Johansen JD, Agner T. Relation between patients’ and physicians’ severity assessment of occupational hand eczema. Br J Dermatol 2005; 153: 596–600. 10.1111/j.1365-2133.2005.06768.x [DOI] [PubMed] [Google Scholar]
- 16.Hald M, Veien NK, Laurberg G, Johansen JD. Severity of hand eczema assessed by patients and dermatologist using a photographic guide. Br J Dermatol 2007; 156: 77–80. 10.1111/j.1365-2133.2006.07565.x [DOI] [PubMed] [Google Scholar]
- 17.Ragamin A, Schappin R, Tan Nguyen N, Nouwen AEM, Hoekstra LF, Schuttelaar MLA, et al. Remote severity assessment in atopic dermatitis: validity and reliability of the remote Eczema Area and Severity Index and Self-Administered Eczema Area and Severity Index. JAAD Int 2023; 13: 184–191. 10.1016/j.jdin.2023.07.019 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Croce EA, Rathouz PJ, Lopes FCPS, Leszczynska M, Diaz LZ, Levy ML, et al. Validation of remote atopic dermatitis severity assessment with the Eczema Area and Severity Index in children using caregiver-provided photos and videos. Pediatr Dermatol 2022; 39: 547–552. 10.1111/pde.15003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.O’Connor DM, Jew OS, Perman MJ, Castelo-Soccio LA, Winston FK, McMahon PJ. Diagnostic accuracy of pediatric teledermatology using parent-submitted photographs a randomized clinical trial. JAMA Dermatol 2017; 153: 1243–1248. 10.1001/jamadermatol.2017.4280 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Ali Z, Robert Zibert J, Dahiya P, Bachdal Johansen C, Grønlund Holm J, Ravn Jørgensen AH, et al. Mild-to-moderate severity of psoriasis may be assessed remotely based on photographs and self-reported extent of skin involvement. JAAD International 2023; 11: 129–136. 10.1016/j.jdin.2023.02.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Ali Z, Chiriac A, Bjerre-Christensen T, Isberg AP, Dahiya P, Manole I, et al. Mild to moderate atopic dermatitis severity can be reliably assessed using smartphone-photographs taken by the patient at home: a validation study. Skin Res Technol 2022; 28: 336–341. 10.1111/srt.13136 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Nichol K, Copes R, Kersey K, Eriksson J, Holness DL. Screening for hand dermatitis in healthcare workers: comparing workplace screening with dermatologist photo screening. Contact Dermatitis 2019; 80: 374–381. 10.1111/cod.13231 [DOI] [PubMed] [Google Scholar]
- 23.Baumeister T, Weistenhöfer W, Drexler H, Kütting B. Spoilt for choice: evaluation of two different scoring systems for early hand eczema in teledermatological examinations. Contact Dermatitis 2010; 62: 241–247. 10.1111/j.1600-0536.2010.01703.x [DOI] [PubMed] [Google Scholar]
- 24.Weigandt WA, Schardt Y, Bruch A, Herr R, Goebeler M, Benecke J, et al. Impact of an eHealth smartphone app on quality of life and clinical outcome of patients with hand and foot eczema: prospective randomized controlled intervention study. JMIR Mhealth Uhealth 2023; 11: e38506. 10.2196/38506 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Parsons V, Williams HC, English J, Llewellyn J, Ntani G, Madan I. The development of a protocol for diagnosing hand dermatitis from photographic images. Contact Dermatitis 2018; 79: 270–275. 10.1111/cod.13053 [DOI] [PubMed] [Google Scholar]
- 26.Zabludovska K, Ibler KS, Jemec GBE, Agner T. Photographic documentation and Hand Eczema Severity Index for severity assessment of hand eczema. Dermatitis 2017; 28: 280–283. 10.1097/DER.0000000000000306 [DOI] [PubMed] [Google Scholar]
- 27.Johansen JD, Hald M, Andersen BL, Laurberg G, Danielsen A, Avnstorp C, et al. Classification of hand eczema: clinical and aetiological types. Based on the guideline of the Danish Contact Dermatitis Group. Contact Dermatitis 2011; 65: 13–21. 10.1111/j.1600-0536.2011.01911.x [DOI] [PubMed] [Google Scholar]
- 28.Held E, Skoet R, Johansen JD, Agner T. The Hand Eczema Severity Index (HECSI): a scoring system for clinical assessment of hand eczema. A study of inter- and intraobserver reliability. Br J Dermatol 2005; 152: 302–307. 10.1111/j.1365-2133.2004.06305.x [DOI] [PubMed] [Google Scholar]
- 29.Coenraads PJ, Van Der Walle H, Thestrup-Pedersen K, Ruzicka T, Dreno B, De La Loge C, et al. Construction and validation of a photographic guide for assessing severity of chronic hand dermatitis. Br J Dermatol 2005; 152: 296–301. 10.1111/j.1365-2133.2004.06270.x [DOI] [PubMed] [Google Scholar]
- 30.Fitzpatrick TB. The validity and practicality of sun-reactive skin types I through VI. Arch Dermatol 1988; 124: 869–871. 10.1001/archderm.124.6.869 [DOI] [PubMed] [Google Scholar]
- 31.Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap): a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 2009; 42: 377–381. 10.1016/j.jbi.2008.08.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Apple Inc . Webpage Apple 2024 [Accessed: 6 November 2024]. Available from: https://www.apple.com/
- 33.NCSS Statistical Software . Webpage PASS: Power Analysis & Sample Size 2024 [Accessed: 6 November 2024] Available from: https://www.ncss.com/software/pass/
- 34.Gagnier JJ, Lai J, Mokkink LB, Terwee CB. COSMIN reporting guideline for studies on measurement properties of patient-reported outcome measures. Qual Life Res 2021; 30: 2197–2218. 10.1007/s11136-021-02822-4 [DOI] [PubMed] [Google Scholar]
- 35.Koo TK, Li MY. A guideline of selecting and reporting intraclass correlation coefficients for reliability research. J Chiropr Med 2016; 15: 155–163. 10.1016/j.jcm.2016.02.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Koch GG, Landis JR, Freeman JL, Freeman DH, Lehnen RG. A general methodology for the analysis of experiments with repeated measurement of categorical data. Biometrics 1977; 33: 133–158. [PubMed] [Google Scholar]
- 37.Yüksel YT, Thyssen JP, Nørreslet LB, Flachs EM, Ebbehøj NE, Agner T. A comparison between self-reported hand eczema and self-reported signs and symptoms of skin lesions indicating hand eczema. Contact Dermatitis 2022; 87: 528–534. 10.1111/cod.14183 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Hald M, Thyssen JP, Zachariae C, Røpke MA, Carstensen JM, Schultz N, et al. Multispectral imaging of hand eczema. Contact Dermatitis 2019; 81: 438–445. 10.1111/cod.13377 [DOI] [PubMed] [Google Scholar]
- 39.Amruthalingam L, Mang N, Gottfrois P, Gonzalez Jimenez A, Maul JT, Kunz M, et al. Objective hand eczema severity assessment with automated lesion anatomical stratification. Exp Dermatol 2023; 32: 521–528. 10.1111/exd.14744 [DOI] [PubMed] [Google Scholar]
- 40.Loman L, Brands MJ, Politiek K, Schuttelaar MLA. Health literacy and associated factors in subjects with hand eczema: a Dutch population-based study. Acta Derm Venereol 2024; 104: adv40079. 10.2340/actadv.v104.40079 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data available on request from the authors.