Patient Preferences for Treatment Attributes in Moderate to Severe Chronic Hand Eczema: A Discrete Choice Experiment Authors: April W. Armstrong, Aseel Bin Sawad, Nisha C. Hazra, Sanjeev Balu, Douglas Maslin, Béa Libchaber, Rui Huang, Marjolaine Gauthier-Loiselle, Annie Guerin Published: 2026 ABSTRACT IntroductionThe treatment landscape for Chronic Hand Eczema (CHE) is evolving; however, evidence on patients’ preferences regarding CHE treatment attributes is sparse. This discrete choice experiment (DCE) assessed how treatment attributes influence preferences among patients with moderate to severe CHE in the USA.MethodsAdults with moderate to severe CHE recruited from an online panel completed a web-based DCE survey (August–September 2025). Seven treatment attributes, identified via literature review, qualitative patient interviews, and clinical expert input, reflected efficacy (improvements in itch, skin appearance, and pain), safety (risks of infections, cancer, and major heart problems), and administration mode/frequency. Patient preferences were analyzed using conditional logistic regression. Attributes’ relative importance and willingness to trade off were calculated. Sensitivity analysis excluding participants who failed ≥ 1 validity test was conducted.ResultsOf the 300 participants with CHE included (mean age: 44.9 years; 50% female; 84.7% employed), most reported multiple CHE subtypes (70.3%). The vast majority of participants reported life impairment due to CHE, including for activities of daily living (96.3%) and work productivity (92.9%). In the DCE, all treatment attributes had a significant impact on treatment preferences (p < 0.001). Attributes with the highest relative importance were related to efficacy: itch improvement (22%) and skin appearance improvement (21%); followed by those related to safety: avoiding risk of cancer (16%) and risk of infections (14%). Other important attributes for patients included pain improvement (13%) and administration mode/frequency (7%). Participants would forgo 28.1, 11.4, and 1.3 percentage points of itch improvement probability to avoid a 1-percentage-point risk of cancer, major heart problems, and infections, respectively.ConclusionsImprovement in efficacy outcomes (itch and skin appearance) and lower risks of certain adverse events (cancer, infections) were the most important treatment attributes for patients with CHE. This study provides insights that may help support shared decision-making to ultimately improve treatment satisfaction and adherence. Key Summary Points Why carry out this study?Chronic Hand Eczema (CHE) is a fluctuating inflammatory skin disease that can impose considerable burden on many aspects of patients’ daily lives.With the evolving CHE treatment landscape, this discrete choice experiment aimed to quantify the extent to which different treatment attributes impact the preferences of patients with moderate to severe CHE in the USA.What was learned from the study?Participants preferred treatments with higher chances of achieving efficacy outcomes, including improvements in itch and skin appearance, while minimizing the risks of adverse events, such as cancer and infections.Participants also preferred topical application twice per day over one injection every 2 weeks, although the mode and frequency of administration was relatively less important in decision-making compared with efficacy and safety attributes.The study findings may help physicians engage with patients in shared decision-making, including discussions of treatment trade-offs, to potentially improve treatment satisfaction, patient engagement, and adherence. Why carry out this study?Chronic Hand Eczema (CHE) is a fluctuating inflammatory skin disease that can impose considerable burden on many aspects of patients’ daily lives.With the evolving CHE treatment landscape, this discrete choice experiment aimed to quantify the extent to which different treatment attributes impact the preferences of patients with moderate to severe CHE in the USA.What was learned from the study?Participants preferred treatments with higher chances of achieving efficacy outcomes, including improvements in itch and skin appearance, while minimizing the risks of adverse events, such as cancer and infections.Participants also preferred topical application twice per day over one injection every 2 weeks, although the mode and frequency of administration was relatively less important in decision-making compared with efficacy and safety attributes.The study findings may help physicians engage with patients in shared decision-making, including discussions of treatment trade-offs, to potentially improve treatment satisfaction, patient engagement, and adherence. Introduction Chronic Hand Eczema (CHE) is a fluctuating inflammatory skin disease involving eczema of the hands and/or wrists that lasts for at least 3 months or with at least two flares in a 12-month period [1]. CHE has a multifactorial pathogenesis encompassing several etiological and clinical subtypes, commonly presenting with skin inflammation, itchiness, pain, redness, flaking, and blistering [2, 3]. These signs and symptoms can impair a patient’s ability to perform work and daily activities [4, 5]. The high visibility of the skin may also contribute to psychosocial burden in patients with skin conditions [6], and many patients with CHE have reported anxiety or depression [7, 8]. A large US survey reported a total lifetime prevalence of hand eczema of 13.6% and a total 1-year prevalence of self-reported physician-diagnosed CHE of 9.6% [9]. The prevalence of CHE can be higher among individuals with occupational wet-work exposure, such as healthcare providers and hairdressers [3, 10]. Patients with CHE may be initially treated with moisturizers/emollients, topical corticosteroids, or topical calcineurin inhibitors, and those refractory to topical treatments may receive phototherapy or systemic therapies [3, 11]. Recently, several small molecule inhibitors and biologics targeting immune signaling pathways have been explored for the treatment of CHE [11]. In July 2025, delgocitinib, a topical pan-Janus kinase (JAK) inhibitor, was the first treatment approved by the US Food and Drug Administration (FDA) specifically for the treatment of moderate to severe CHE in adults who have had an inadequate response to, or for whom topical corticosteroids are not advisable [12, 13]. Other immune signaling inhibitors under clinical investigation include ruxolitinib (JAK1/JAK2 inhibitor) for topical application twice per day, which has been investigated for moderate to severe nonatopic CHE in a phase 2 trial among patients with or without inadequate response to topical corticosteroids [14], as well as dupilumab (interleukin [IL]-4 and IL-13 inhibitor) and tralokinumab (IL-13 inhibitor), both administered as a subcutaneous injection every 2 weeks and have been studied in phase 3 trials of atopic hand eczema, a CHE subtype [15, 16]; dupilumab has also been investigated in severe CHE of mixed subtypes in one phase 2b study [17]. Investigational systemic JAK inhibitors for CHE include abrocitinib (oral JAK1 inhibitor) [18] and gusacitinib (oral JAK/spleen tyrosine kinase inhibitor) [19]. The expanding range of therapeutics under investigation for CHE or its subtypes is reshaping the treatment landscape, offering different mechanisms of action, efficacy and safety profiles, and administration route [11]. An understanding of how patients value different attributes of CHE treatments would be essential to facilitate well-informed treatment discussions between physicians and patients. Discrete choice experiment (DCE) is a research method used to quantify patients’ treatment preferences and benefit–risk trade-offs, enabling assessment of the relative importance of different factors in decision-making [20]. To our knowledge, there has only been one prior US CHE-specific DCE study, which showed that patients rated efficacy improvements as more important than eliminating the risks of specific adverse events (AEs); however, the prior study focused on patients with severe, topical-refractory CHE and a limited set of treatment attributes [21]. Given the sparse evidence on patient preferences regarding CHE treatment attributes and the evolving treatment landscape, an updated DCE study that incorporates multiple efficacy dimensions, clinically relevant safety considerations, and treatment logistics in a broader CHE population is warranted. The present DCE study aimed to expand existing evidence by evaluating CHE treatment attributes relevant to the evolving landscape and to quantify the extent to which different treatment attributes impact the preferences of patients with moderate to severe CHE in the USA. Methods Data Source and Study Population Participants were recruited from online panels of a well-established market research firm, Dynata, including a panel of patients with eczema/atopic dermatitis (AD) and a general population panel to ensure broad coverage of CHE subtypes, based on key eligibility criteria. Data were collected between August and September 2025. Participants were eligible if they were (1) aged at least 18 years; (2) living in the USA at the time of the survey; (3) had self-reported diagnosis of CHE, defined as eczema of the hands that persisted for at least 3 months in the past 12 months or recurred twice or more in the past 12 months; (4) had moderate to severe CHE, with severity defined by inadequate response to topical corticosteroids for CHE treatment or prior and/or current use of systemic therapy (e.g., oral corticosteroids, biologics) or phototherapy for CHE treatment; (5) were able to read and understand English; and (6) had access to appropriate technology for completing a web-based survey. Participants were excluded if they had participated in a clinical trial for eczema/AD during the past 12 months. This study is exempt by the Western Copernicus Group Institutional Review Board under 45 CFR § 46.104(d)(2) because the research only includes interactions involving educational tests, survey procedures, interview procedures, or observations of public behavior; and the information obtained is recorded by the investigator in such a manner that the identity of the human subjects cannot readily be ascertained, directly or through identifiers linked to the subjects. Informed consent was obtained from the participants prior to the start of the study. Study Design Interviews The first phase of the study comprised qualitative interviews with participants to understand the most important CHE treatment attributes from the patient perspective and help prioritize the final list of attributes in the DCE. In this phase of the study, semi-structured, one-on-one qualitative interviews were conducted with 15 eligible participants. Participants were selected such that there was a mix of CHE subtypes, moderate to severe diseases, with and without experience with systemic therapy. Fifteen interviews were considered a sufficient sample size to reach saturation of information for the purpose of informing the selection of attributes and the design of the DCE surveys [22, 23]. During the interviews, participants rated 16 factors related to efficacy, safety, and treatment logistics that could be important when making a treatment decision; these factors were identified on the basis of a literature search of clinical trial results for existing and potential treatments for CHE (including DELTA 1, DELTA 2, and DELTA 3 for delgocitinib [24–26], Smith et al. 2024 for ruxolitinib [27], Voorberg et al. 2023 for dupilumab [17]), US prescribing labels [12, 28–30], and past DCE research in CHE [21]. Potential systemic JAK inhibitors for CHE were not considered, as the results from the abrocitinib trial in CHE were not available at the time of study conduct, and gusacitinib is not yet available in the USA (not approved by the FDA). Each interview lasted approximately 60 min. The findings from the qualitative interviews along with clinical expert input and information from the published literature were used to inform the final attributes and levels for inclusion and design of the DCE in the quantitative phase [22]. The final list of seven treatment attributes, each with two to three levels, included efficacy-related attributes (i.e., itch improvement, skin appearance improvement, pain improvement), safety-related attributes (i.e., risk of infections, risk of cancer, risk of major heart problems), and mode and frequency of administration (i.e., one injection every 2 weeks or topical application twice per day) (Table 1).Table 1Attributes and levels included in the DCEAttributesLevelsItch improvement40 in 100 patients (40%)60 in 100 patients (60%)80 in 100 patients (80%)Skin appearance improvement40 in 100 patients (40%)60 in 100 patients (60%)80 in 100 patients (80%)Pain improvement40 in 100 patients (40%)60 in 100 patients (60%)80 in 100 patients (80%)Risk of infections5 in 100 patients (5%)10 in 100 patients (10%)25 in 100 patients (25%)Risk of cancer (other than non-melanoma skin cancer)0 in 1000 patients (0%)5 in 1000 patients (0.5%)10 in 1000 patients (1%)Risk of major heart problems0 in 1000 patients (0%)5 in 1000 patients (0.5%)10 in 1000 patients (1%)Mode and frequency of administrationOne injection every 2 weeksTopical application twice per dayDCE Discrete choice experiment Table 1Attributes and levels included in the DCEAttributesLevelsItch improvement40 in 100 patients (40%)60 in 100 patients (60%)80 in 100 patients (80%)Skin appearance improvement40 in 100 patients (40%)60 in 100 patients (60%)80 in 100 patients (80%)Pain improvement40 in 100 patients (40%)60 in 100 patients (60%)80 in 100 patients (80%)Risk of infections5 in 100 patients (5%)10 in 100 patients (10%)25 in 100 patients (25%)Risk of cancer (other than non-melanoma skin cancer)0 in 1000 patients (0%)5 in 1000 patients (0.5%)10 in 1000 patients (1%)Risk of major heart problems0 in 1000 patients (0%)5 in 1000 patients (0.5%)10 in 1000 patients (1%)Mode and frequency of administrationOne injection every 2 weeksTopical application twice per dayDCE Discrete choice experiment Attributes and levels included in the DCE DCE Discrete choice experiment DCE Survey The second phase of the study comprised the web-based DCE survey to evaluate preferences for the seven treatment attributes in the DCE. A total of 300 participants were recruited and completed the survey. The sample size and number of attributes conform to the recommendations of the International Society for Pharmacoeconomics and Outcomes Research (ISPOR) Good Research Practices for Conjoint Analysis Task Force [20, 31]. The survey began with a brief introduction and screening questions to confirm participant eligibility. It then collected questions on participant’s demographics, clinical characteristics, and CHE treatment history. The impact of CHE on work productivity and daily activities was assessed using the Work Productivity and Activity Impairment Questionnaire: Specific Health Problem (WPAI:SHP [32]). Finally, the survey included the DCE components to quantify participant’s preference data. Based on the final list of attributes and levels identified in the first phase, 60 choice cards were generated using a D-efficient design in the Ngene software [33]. To limit response burden, the choice cards were divided into six blocks, such that each participant completed ten choice cards. Each choice card displayed a pair of hypothetical treatment profiles, and participants were asked to select the profile that best reflected their preferences between the two profiles (Fig. 1).Fig. 1Example choice card Example choice card All 300 participants also completed two additional choice cards to assess the internal validity of their answers. The validity tests conducted were a dominance test, where all attributes in one treatment profile were better than those in the other option and the participant was expected to select the dominant profile, and a stability test, where a choice card was repeated and the participant was expected to select the same option. Statistical Analyses Participants’ characteristics and the impact of CHE on work productivity and daily activities were descriptively summarized. Continuous variables were reported using means ± standard deviations (SD) and medians (ranges or interquartile ranges); categorical variables were reported using frequency counts and percentages. A conditional logistic regression model was used to analyze participant’s preference data. The dependent variable for the model was the participant’s preferred treatment choice, and the independent variables were the levels of each attribute coded as continuous variables. The regression coefficients represented the preference weights of the attributes, with the sign indicating participant’s positive or negative attitude toward an attribute, and the associated p-value indicating whether the estimated coefficient of the attribute was statistically significantly different from zero, that is, whether the attribute had a significant impact on participants’ preferences. Part-worth utility associated with each attribute level was calculated by multiplying the coefficients with each level of a given attribute. A higher part-worth utility indicates a greater impact on the overall utility. A negative part-worth utility indicates disutility (i.e., reduction in overall utility). Relative importance scores of attributes were calculated to allow for comparisons across attributes. The relative importance of each attribute was normalized to percentages based on the difference between the best and worst attribute levels, multiplied by the coefficient for each attribute. The value represents the importance of an attribute in a decision to choose a specific treatment profile [34]. The willingness to trade off was calculated using the formula \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$- \left( {\beta_{{{\mathrm{attribute}}}} /\beta_{{{\mathrm{itch}}}} } \right)$$\end{document}-βattribute/βitch, where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\beta_{{{\mathrm{attribute}}}}$$\end{document}βattribute represents the coefficient of a given attribute estimated from the conditional logistic regression. The value indicates how many percentage points in the probability of achieving meaningful itch improvement an average participant is willing to trade off to avoid a particular attribute. For AE attributes, the trade-off is expressed with respect to a 1-percentage-point reduction in the risk of having the AE. To evaluate the internal consistency of the preference data, a sensitivity analysis excluding participants who failed either of the validity tests (i.e., dominance and/or stability tests) was conducted. All analyses were carried out using SAS Enterprise Guide statistical software version 7.1 (SAS Institute, Cary, NC). Results Participant Characteristics A total of 300 eligible participants (68.3% from the eczema/AD panel) completed the DCE survey and were included in the main analysis (Table 2). The sample mean ± SD age was 44.9 ± 12.2 years and comprised equal numbers of male and female participants. The most commonly reported race was White or Caucasian (82.3%), followed by Black or African American (13.3%). Nearly 85% of participants were employed at the time of the survey, with 78.0% being employed full time and 6.7% part time; 5.3% were retired, 3.7% were unemployed, 3.7% were homemakers, and 2.7% were on disability.Table 2Participant characteristicsAll participantsN = 300Demographics Age (years) Mean ± SD44.9 ± 12.2 Median (range)42.0 (21.0, 85.0) Gender, n (%) Male150 (50.0%) Female150 (50.0%) Race, n (%)a White or Caucasian247 (82.3%) Black or African American40 (13.3%) Asian12 (4.0%) American Indian or Alaska Native8 (2.7%) Native Hawaiian or Other Pacific Islander3 (1.0%) Other3 (1.0%) Residence area, n (%) Urban170 (56.7%) Suburban102 (34.0%) Rural28 (9.3%) Employment status, n (%) Full time234 (78.0%) Part time20 (6.7%) Retired16 (5.3%) Unemployed11 (3.7%) Homemaker11 (3.7%) On disability8 (2.7%)Clinical characteristics CHE subtypea, n (%) Multiple subtypes211 (70.3%) Atopic181 (60.3%) Irritant157 (52.3%) Allergic116 (38.7%) Vesicular111 (37.0%) Hyperkeratotic105 (35.0%) Unknown6 (2.0%) Time since CHE diagnosis, n (%) 1–2 years ago73 (24.3%) 3–5 years ago103 (34.3%) More than 5 years ago124 (41.3%) Parts of hand(s) currently affected by CHEa,b, n (%) Back of hand(s)226 (75.3%) Finger(s)204 (68.0%) Wrist(s)188 (62.7%) Palm(s)166 (55.3%) Other parts of the hand(s)23 (7.7%) Most common body regions other than hands affected by eczema/dermatitisa, n (%) Elbow(s)146 (48.7%) Forearm(s)140 (46.7%) Face123 (41.0%) Neck118 (39.3%) Foot/feet103 (34.3%) Most common health conditions diagnosed at any timea, n (%) Anxiety96 (32.0%) Depression75 (25.0%) High blood pressure63 (21.0%) Diabetes43 (14.3%) Insomnia40 (13.3%) Obesity40 (13.3%) Food allergies31 (10.3%)CHE Chronic Hand Eczema, SD standard deviationaCategories are not mutually exclusivebAll participants were required to select “hands” or “wrists” to be eligible to participate in the study Table 2Participant characteristicsAll participantsN = 300Demographics Age (years) Mean ± SD44.9 ± 12.2 Median (range)42.0 (21.0, 85.0) Gender, n (%) Male150 (50.0%) Female150 (50.0%) Race, n (%)a White or Caucasian247 (82.3%) Black or African American40 (13.3%) Asian12 (4.0%) American Indian or Alaska Native8 (2.7%) Native Hawaiian or Other Pacific Islander3 (1.0%) Other3 (1.0%) Residence area, n (%) Urban170 (56.7%) Suburban102 (34.0%) Rural28 (9.3%) Employment status, n (%) Full time234 (78.0%) Part time20 (6.7%) Retired16 (5.3%) Unemployed11 (3.7%) Homemaker11 (3.7%) On disability8 (2.7%)Clinical characteristics CHE subtypea, n (%) Multiple subtypes211 (70.3%) Atopic181 (60.3%) Irritant157 (52.3%) Allergic116 (38.7%) Vesicular111 (37.0%) Hyperkeratotic105 (35.0%) Unknown6 (2.0%) Time since CHE diagnosis, n (%) 1–2 years ago73 (24.3%) 3–5 years ago103 (34.3%) More than 5 years ago124 (41.3%) Parts of hand(s) currently affected by CHEa,b, n (%) Back of hand(s)226 (75.3%) Finger(s)204 (68.0%) Wrist(s)188 (62.7%) Palm(s)166 (55.3%) Other parts of the hand(s)23 (7.7%) Most common body regions other than hands affected by eczema/dermatitisa, n (%) Elbow(s)146 (48.7%) Forearm(s)140 (46.7%) Face123 (41.0%) Neck118 (39.3%) Foot/feet103 (34.3%) Most common health conditions diagnosed at any timea, n (%) Anxiety96 (32.0%) Depression75 (25.0%) High blood pressure63 (21.0%) Diabetes43 (14.3%) Insomnia40 (13.3%) Obesity40 (13.3%) Food allergies31 (10.3%)CHE Chronic Hand Eczema, SD standard deviationaCategories are not mutually exclusivebAll participants were required to select “hands” or “wrists” to be eligible to participate in the study Participant characteristics CHE Chronic Hand Eczema, SD standard deviation aCategories are not mutually exclusive bAll participants were required to select “hands” or “wrists” to be eligible to participate in the study Among all participants, 70.3% reported multiple CHE subtypes, and the most common subtypes were atopic (60.3%), irritant (52.3%), and allergic (38.7%); other subtypes included vesicular (37.0%) and hyperkeratotic (35.0%) (Table 2). Participants most commonly reported having been diagnosed with CHE more than 5 years prior to the survey (41.3%), followed by 3–5 years prior (34.3%) and 1–2 years prior (24.3%). More than three-quarters (75.3%) of participants reported that the CHE affected the back of the hand(s) at the time of the survey; the proportions reporting involvement of the finger(s), wrist(s), and palm(s) were 68.0%, 62.7%, and 55.3%, respectively. Other body regions most commonly affected by eczema/dermatitis included the elbow(s) (48.7%) and forearm(s) (46.7%). Approximately 30% of participants were diagnosed with anxiety (32.0%) and/or depression (25.0%) at any time prior to the survey. Regarding treatment for CHE (Table 3), most participants had experience with systemic therapy (73.3%), most commonly with oral therapy only (42.7%). At the time of the survey, the majority of participants were using prescription topical corticosteroids (85.7%), half were taking oral pills (49.3%), and a quarter were being treated with injections (23.0%); 52.0% were using unmedicated moisturizers/emollients. When selecting their current treatment, nearly 1 in 4 participants (22.7%) reported that their healthcare provider presented only one treatment option. Only 75.7% reported that their healthcare provider discussed treatment efficacy and 59.7% reported discussing risk of side effects for their current treatment.Table 3CHE treatment historyAll participantsN = 300Systemic therapy experience (ever), n (%) Systemic experience220 (73.3%) Experience with oral therapy only128 (42.7%) Experience with injection therapy only10 (3.3%) Experience with both oral and injection therapy82 (27.3%) Without systemic experience80 (26.7%)Current treatment typea, n (%) Unmedicated moisturizer/emollient156 (52.0%) Mean time on treatment ± SD (months)21.3 ± 34.5 Median time on treatment (IQR) (months)8.5 (4.0, 24.0) Prescription topical corticosteroid257 (85.7%) Mean time on treatment ± SD (months)13.4 ± 18.3 Median time on treatment (IQR) (months)6.0 (4.0, 12.0) Oral pill148 (49.3%) Mean time on treatment ± SD (months)9.6 ± 12.3 Median time on treatment (IQR) (months)6.0 (3.0, 12.0) Injection69 (23.0%) Mean time on treatment ± SD (months)8.6 ± 15.7 Median time on treatment (IQR) (months)4.0 (3.0, 8.0) Phototherapy or light therapy43 (14.3%) Mean time on treatment ± SD (months)8.3 ± 11.7 Median time on treatment (IQR) (months)5.0 (2.0, 8.0)How current CHE treatment was selected, n (%) Healthcare provider presented several treatment options but particularly recommended this option.116 (38.7%) Healthcare provider presented several treatment options and decided together with the participant this would be the best option.96 (32.0%) Healthcare provider presented only that treatment option.68 (22.7%) Participant suggested treatment option to healthcare providers.17 (5.7%) Unknown2 (0.7%) Prefer not to answer1 (0.3%)Type of information discussed with healthcare provider regarding selected treatment, n (%) Treatment efficacy227 (75.7%) Mode and frequency of administration190 (63.3%) Risk of side effects179 (59.7%) How commonly treatment is used among patients with hand eczema/dermatitis159 (53.0%) Previous positive patient experiences with treatment134 (44.7%) Cost or insurance coverage of treatment108 (36.0%) Previous negative patient experiences with treatment53 (17.7%) Otherb1 (0.3%)CHE Chronic Hand Eczema, IQR interquartile range, SD standard deviationaCategories are not mutually exclusiveb“Other” information discussed included “nothing specific” Table 3CHE treatment historyAll participantsN = 300Systemic therapy experience (ever), n (%) Systemic experience220 (73.3%) Experience with oral therapy only128 (42.7%) Experience with injection therapy only10 (3.3%) Experience with both oral and injection therapy82 (27.3%) Without systemic experience80 (26.7%)Current treatment typea, n (%) Unmedicated moisturizer/emollient156 (52.0%) Mean time on treatment ± SD (months)21.3 ± 34.5 Median time on treatment (IQR) (months)8.5 (4.0, 24.0) Prescription topical corticosteroid257 (85.7%) Mean time on treatment ± SD (months)13.4 ± 18.3 Median time on treatment (IQR) (months)6.0 (4.0, 12.0) Oral pill148 (49.3%) Mean time on treatment ± SD (months)9.6 ± 12.3 Median time on treatment (IQR) (months)6.0 (3.0, 12.0) Injection69 (23.0%) Mean time on treatment ± SD (months)8.6 ± 15.7 Median time on treatment (IQR) (months)4.0 (3.0, 8.0) Phototherapy or light therapy43 (14.3%) Mean time on treatment ± SD (months)8.3 ± 11.7 Median time on treatment (IQR) (months)5.0 (2.0, 8.0)How current CHE treatment was selected, n (%) Healthcare provider presented several treatment options but particularly recommended this option.116 (38.7%) Healthcare provider presented several treatment options and decided together with the participant this would be the best option.96 (32.0%) Healthcare provider presented only that treatment option.68 (22.7%) Participant suggested treatment option to healthcare providers.17 (5.7%) Unknown2 (0.7%) Prefer not to answer1 (0.3%)Type of information discussed with healthcare provider regarding selected treatment, n (%) Treatment efficacy227 (75.7%) Mode and frequency of administration190 (63.3%) Risk of side effects179 (59.7%) How commonly treatment is used among patients with hand eczema/dermatitis159 (53.0%) Previous positive patient experiences with treatment134 (44.7%) Cost or insurance coverage of treatment108 (36.0%) Previous negative patient experiences with treatment53 (17.7%) Otherb1 (0.3%)CHE Chronic Hand Eczema, IQR interquartile range, SD standard deviationaCategories are not mutually exclusiveb“Other” information discussed included “nothing specific” CHE treatment history CHE Chronic Hand Eczema, IQR interquartile range, SD standard deviation aCategories are not mutually exclusive b“Other” information discussed included “nothing specific” Impact of CHE on Work Productivity and Daily Activities Based on the WPAI:SHP, the mean activity impairment due to CHE was 52.8% among all 300 participants, with the vast majority of participants (96.3%) experiencing at least some impairment in activity as a result of their CHE. The most common daily activities impacted by CHE symptoms were household chores (70.0%) and handwashing (69.3%), followed by cooking (49.3%), personal grooming (49.3%), and showering/bathing (47.7%) (Table 4).Table 4Impact on work productivity and daily activitiesAll participantsN = 300Activities impacted by CHE symptomsa, n (%) Household chores210 (70.0%) Handwashing208 (69.3%) Cooking148 (49.3%) Personal grooming148 (49.3%) Showering/bathing143 (47.7%) Writing129 (43.0%) Typing/using a keyboard or touchscreen104 (34.7%) Gardening99 (33.0%) Dressing88 (29.3%) Driving62 (20.7%) Childcare39 (13.0%) None of the above21 (7.0%)WPAI:SHPb,c Percent activity impairment due to CHEd Mean ± SD52.8 ± 26.3 Median (range)60.0 (0.0, 100.0) Any activity impairment due to CHE, n (%)289 (96.3%) Number of people employed at the time of the survey, n (%)254 (84.7%) Work environment among people employed, n (%) Fully in-person148 (58.3%) Mostly in-person71 (28.0%) Mostly remote18 (7.1%) Fully remote17 (6.7%) Number of responsese, n (%)254 (84.7%) Percent work time missed due to CHE (absenteeism)f Mean ± SD12.6 ± 19.0 Median (range)4.4 (0.0, 100.0) Percent impairment while working due to CHE (presenteeism)g Mean ± SD51.9 ± 28.4 Median (range)60.0 (0.0, 100.0) Percent overall work impairment due to CHEg Mean ± SD55.7 ± 29.2 Median (range)61.5 (0.0, 100.0) Any overall work impairment due to CHE, n (%)236 (92.9%)CHE Chronic Hand Eczema, SD standard deviation, WPAI:SHP Work Productivity and Activity Impairment Questionnaire: Specific Health ProblemaCategories are not mutually exclusivebA higher WPAI:SHP outcome indicates greater impairment and less productivitycThe recall period was the last 7 daysdPercent activity impairment due to CHE was assessed among the total sample, regardless of employment statuseThe number of responses reported those who were employed and worked > 0 h in the past 7 days of the surveyfPercent work time missed was assessed among all participants who were employed at the time of the surveygImpairment while working and overall work impairment were only assessed among participants who were employed and worked > 0 h in the past 7 days of the survey Table 4Impact on work productivity and daily activitiesAll participantsN = 300Activities impacted by CHE symptomsa, n (%) Household chores210 (70.0%) Handwashing208 (69.3%) Cooking148 (49.3%) Personal grooming148 (49.3%) Showering/bathing143 (47.7%) Writing129 (43.0%) Typing/using a keyboard or touchscreen104 (34.7%) Gardening99 (33.0%) Dressing88 (29.3%) Driving62 (20.7%) Childcare39 (13.0%) None of the above21 (7.0%)WPAI:SHPb,c Percent activity impairment due to CHEd Mean ± SD52.8 ± 26.3 Median (range)60.0 (0.0, 100.0) Any activity impairment due to CHE, n (%)289 (96.3%) Number of people employed at the time of the survey, n (%)254 (84.7%) Work environment among people employed, n (%) Fully in-person148 (58.3%) Mostly in-person71 (28.0%) Mostly remote18 (7.1%) Fully remote17 (6.7%) Number of responsese, n (%)254 (84.7%) Percent work time missed due to CHE (absenteeism)f Mean ± SD12.6 ± 19.0 Median (range)4.4 (0.0, 100.0) Percent impairment while working due to CHE (presenteeism)g Mean ± SD51.9 ± 28.4 Median (range)60.0 (0.0, 100.0) Percent overall work impairment due to CHEg Mean ± SD55.7 ± 29.2 Median (range)61.5 (0.0, 100.0) Any overall work impairment due to CHE, n (%)236 (92.9%)CHE Chronic Hand Eczema, SD standard deviation, WPAI:SHP Work Productivity and Activity Impairment Questionnaire: Specific Health ProblemaCategories are not mutually exclusivebA higher WPAI:SHP outcome indicates greater impairment and less productivitycThe recall period was the last 7 daysdPercent activity impairment due to CHE was assessed among the total sample, regardless of employment statuseThe number of responses reported those who were employed and worked > 0 h in the past 7 days of the surveyfPercent work time missed was assessed among all participants who were employed at the time of the surveygImpairment while working and overall work impairment were only assessed among participants who were employed and worked > 0 h in the past 7 days of the survey Impact on work productivity and daily activities CHE Chronic Hand Eczema, SD standard deviation, WPAI:SHP Work Productivity and Activity Impairment Questionnaire: Specific Health Problem aCategories are not mutually exclusive bA higher WPAI:SHP outcome indicates greater impairment and less productivity cThe recall period was the last 7 days dPercent activity impairment due to CHE was assessed among the total sample, regardless of employment status eThe number of responses reported those who were employed and worked > 0 h in the past 7 days of the survey fPercent work time missed was assessed among all participants who were employed at the time of the survey gImpairment while working and overall work impairment were only assessed among participants who were employed and worked > 0 h in the past 7 days of the survey Among the 254 participants who were employed at the time of the survey, 58.3% were working fully in-person, 28.0% were working mostly in-person, 7.1% were working mostly remote, and 6.7% were working fully remote. On average, employed participants reported having missed 12.6% of their scheduled work time (absenteeism) in the past 7 days of the survey due to their CHE. Among employed participants who worked > 0 h in the past 7 days, participants on average experienced 51.9% impairment while working (presenteeism) and an average overall work impairment of 55.7% attributable to their CHE (Table 4). Patient Preference for Treatment Attributes All attributes included in the DCE had a highly significant impact on participant preferences (p < 0.001; Table 5). Specifically, all efficacy attributes were associated with positive preference weights, indicating that participants preferred CHE treatments with a higher chance of achieving itch improvement, skin appearance improvement, and pain improvement. AE attributes, including risk of infections, risk of cancer, and risk of major heart problems, were all associated with negative preference weights, indicating that participants preferred CHE treatments with lower risks of these AEs. Furthermore, one injection every 2 weeks was associated with a negative preference weight when compared with topical application twice per day, indicating that between the two administration options, participants had a preference toward topical application twice per day.Table 5Preference weights and willingness to trade off—full sampleAll participants (N = 300)Coefficienta (95% CI)p-ValueWillingness to trade offbRelative importancecRelative importance rankingdEfficacy Itch improvement (percentage points)2.186(1.908, 2.463)< 0.001*Reference22%1 Skin appearance improvement (percentage points)2.050(1.781, 2.320)< 0.001*21%2 Pain improvement (percentage points)1.306(1.035, 1.578)< 0.001*13%5AEs Risk of infections (percentage points)−2.804(−3.335, −2.273)< 0.001*1.314%4 Risk of cancer (percentage points)−61.316(−72.675, −49.957)< 0.001*28.116%3 Risk of major heart problems (percentage points)−24.934(−36.080, −13.788)< 0.001*11.46%7Mode and frequency of administration Topical application twice per day (reference)————— One injection every 2 weeks−0.263(−0.362, −0.163)< 0.001*0.17%6AE adverse events, CI confidence intervalaRegression coefficients are from a conditional logistic model that regresses participants’ choice on the attribute’s levels of the treatment selected from each choice card. Coefficients may only be interpreted in terms of significancebWillingness to trade off was calculated using the formula −(βattribute/βitch). The number indicates how many percentage points in the probability of achieving meaningful itch improvement an average participant is willing to trade off to avoid a particular attribute. For AE attributes, the trade-off is expressed with respect to a 1-percentage-point reduction in the risk of having the AE. For example, an average participant is willing to trade off 1.3-percentage-point chance in achieving meaningful improvement in itch to avoid a 1-percentage-point risk of infectionscRelative importance scores were normalized to percentages based on the difference between the best and worst attribute level, multiplied by the coefficient for each attribute, to allow for comparisons across attributes. Relative importance scores could be interpreted as the proportional value of a change from a treatment with the worst levels to a treatment with the best levels, captured by a specific attributedThis ranking shows the relative importance of each attribute, with 1 being the highest ranked and 7 being the lowest ranked importance Table 5Preference weights and willingness to trade off—full sampleAll participants (N = 300)Coefficienta (95% CI)p-ValueWillingness to trade offbRelative importancecRelative importance rankingdEfficacy Itch improvement (percentage points)2.186(1.908, 2.463)< 0.001*Reference22%1 Skin appearance improvement (percentage points)2.050(1.781, 2.320)< 0.001*21%2 Pain improvement (percentage points)1.306(1.035, 1.578)< 0.001*13%5AEs Risk of infections (percentage points)−2.804(−3.335, −2.273)< 0.001*1.314%4 Risk of cancer (percentage points)−61.316(−72.675, −49.957)< 0.001*28.116%3 Risk of major heart problems (percentage points)−24.934(−36.080, −13.788)< 0.001*11.46%7Mode and frequency of administration Topical application twice per day (reference)————— One injection every 2 weeks−0.263(−0.362, −0.163)< 0.001*0.17%6AE adverse events, CI confidence intervalaRegression coefficients are from a conditional logistic model that regresses participants’ choice on the attribute’s levels of the treatment selected from each choice card. Coefficients may only be interpreted in terms of significancebWillingness to trade off was calculated using the formula −(βattribute/βitch). The number indicates how many percentage points in the probability of achieving meaningful itch improvement an average participant is willing to trade off to avoid a particular attribute. For AE attributes, the trade-off is expressed with respect to a 1-percentage-point reduction in the risk of having the AE. For example, an average participant is willing to trade off 1.3-percentage-point chance in achieving meaningful improvement in itch to avoid a 1-percentage-point risk of infectionscRelative importance scores were normalized to percentages based on the difference between the best and worst attribute level, multiplied by the coefficient for each attribute, to allow for comparisons across attributes. Relative importance scores could be interpreted as the proportional value of a change from a treatment with the worst levels to a treatment with the best levels, captured by a specific attributedThis ranking shows the relative importance of each attribute, with 1 being the highest ranked and 7 being the lowest ranked importance Preference weights and willingness to trade off—full sample AE adverse events, CI confidence interval aRegression coefficients are from a conditional logistic model that regresses participants’ choice on the attribute’s levels of the treatment selected from each choice card. Coefficients may only be interpreted in terms of significance bWillingness to trade off was calculated using the formula −(βattribute/βitch). The number indicates how many percentage points in the probability of achieving meaningful itch improvement an average participant is willing to trade off to avoid a particular attribute. For AE attributes, the trade-off is expressed with respect to a 1-percentage-point reduction in the risk of having the AE. For example, an average participant is willing to trade off 1.3-percentage-point chance in achieving meaningful improvement in itch to avoid a 1-percentage-point risk of infections cRelative importance scores were normalized to percentages based on the difference between the best and worst attribute level, multiplied by the coefficient for each attribute, to allow for comparisons across attributes. Relative importance scores could be interpreted as the proportional value of a change from a treatment with the worst levels to a treatment with the best levels, captured by a specific attribute dThis ranking shows the relative importance of each attribute, with 1 being the highest ranked and 7 being the lowest ranked importance Part-worth utility associated with each attribute level is presented in Fig. 2. Overall, part-worth utilities increased with greater levels (chance) of improvements in itch, skin appearance, and pain, respectively. Part-worth utilities were negative for any risk of AEs, including infections, cancer, and major heart problems, with higher risks associated with greater disutilities. When evaluating topical application twice per day as a reference, one injection every 2 weeks was associated with disutility.Fig. 2Part-worth utilities. aRefers to topical application twice per day. bRefers to one injection every 2 weeks Part-worth utilities. aRefers to topical application twice per day. bRefers to one injection every 2 weeks Across attributes included in the DCE, the most important attributes influencing participant preferences for CHE treatment were itch improvement and skin appearance improvement (relative importance: 22% and 21%, respectively), followed by risk of cancer (16%), risk of infections (14%), pain improvement (13%), mode and frequency of administration (7%), and risk of major heart problems (6%), which was the least important (Fig. 3; Table 5).Fig. 3Relative importance of the attributes Relative importance of the attributes On average, a participant was willing to accept a reduction of 28.1, 11.4, and 1.3 percentage points in the probability of achieving itch improvement to receive a treatment with 1-percentage-point reduction in the risk of cancer, major heart problems, and infections, respectively (Table 5). Sensitivity Analysis Of all 300 participants, 95 (31.7%) failed one of the validity tests, and 19 (6.3%) failed both tests. A sensitivity analysis was conducted in the remaining 186 participants (62.0%) who passed both validity tests. The estimated preference weights of the attributes were of similar magnitudes as those from the full sample, and the attributes’ relative importance rankings were also similar (Table 6), demonstrating the robustness of the main analysis.Table 6Preference weights and willingness to trade off—sensitivity analysisParticipants excluding those who failed at least one of the validity tests (N = 186)Coefficienta (95% CI)p-ValueWillingness to trade offbRelative importancecRelative importance rankingdEfficacy Itch improvement (percentage points)3.158(2.783, 3.533)< 0.001*Reference23%1 Skin appearance improvement (percentage points)2.774(2.410, 3.138)< 0.001*20%2 Pain improvement (percentage points)1.744(1.383, 2.105)< 0.001*13%5AEs Risk of infections (percentage points)−3.702(−4.417, −2.987)< 0.001*1.214%4 Risk of cancer (percentage points)−97.312(−112.691, −81.934)< 0.001*30.818%3 Risk of major heart problems (percentage points)−34.706(−49.584, −19.828)< 0.001*11.06%6Mode and frequency of administration Topical application twice per day (reference)—— One injection every 2 weeks−0.344(−0.479, −0.210)< 0.001*0.16%7AE adverse events, CI confidence intervalaRegression coefficients are from a conditional logistic model that regresses participants’ choice on the attribute’s levels of the treatment selected from each choice card. Coefficients may only be interpreted in terms of significancebWillingness to trade off was calculated using the formula −(βattribute/βitch). The number indicates how many percentage points in the probability of achieving meaningful itch improvement an average participant is willing to trade off to avoid a particular attribute. For AE attributes, the trade-off is expressed with respect to a 1-percentage-point reduction in the risk of having the AE. For example, an average participant is willing to trade off 1.3-percentage-point chance in achieving meaningful improvement in itch to avoid a 1-percentage-point risk of infectionscRelative importance scores were normalized to percentages based on the difference between the best and worst attribute level, multiplied by the coefficient for each attribute, to allow for comparisons across attributes. Relative importance scores could be interpreted as the proportional value of a change from a treatment with the worst levels to a treatment with the best levels, captured by a specific attributedThis ranking shows the relative importance of each attribute, with 1 being the highest ranked and 7 being the lowest ranked importance Table 6Preference weights and willingness to trade off—sensitivity analysisParticipants excluding those who failed at least one of the validity tests (N = 186)Coefficienta (95% CI)p-ValueWillingness to trade offbRelative importancecRelative importance rankingdEfficacy Itch improvement (percentage points)3.158(2.783, 3.533)< 0.001*Reference23%1 Skin appearance improvement (percentage points)2.774(2.410, 3.138)< 0.001*20%2 Pain improvement (percentage points)1.744(1.383, 2.105)< 0.001*13%5AEs Risk of infections (percentage points)−3.702(−4.417, −2.987)< 0.001*1.214%4 Risk of cancer (percentage points)−97.312(−112.691, −81.934)< 0.001*30.818%3 Risk of major heart problems (percentage points)−34.706(−49.584, −19.828)< 0.001*11.06%6Mode and frequency of administration Topical application twice per day (reference)—— One injection every 2 weeks−0.344(−0.479, −0.210)< 0.001*0.16%7AE adverse events, CI confidence intervalaRegression coefficients are from a conditional logistic model that regresses participants’ choice on the attribute’s levels of the treatment selected from each choice card. Coefficients may only be interpreted in terms of significancebWillingness to trade off was calculated using the formula −(βattribute/βitch). The number indicates how many percentage points in the probability of achieving meaningful itch improvement an average participant is willing to trade off to avoid a particular attribute. For AE attributes, the trade-off is expressed with respect to a 1-percentage-point reduction in the risk of having the AE. For example, an average participant is willing to trade off 1.3-percentage-point chance in achieving meaningful improvement in itch to avoid a 1-percentage-point risk of infectionscRelative importance scores were normalized to percentages based on the difference between the best and worst attribute level, multiplied by the coefficient for each attribute, to allow for comparisons across attributes. Relative importance scores could be interpreted as the proportional value of a change from a treatment with the worst levels to a treatment with the best levels, captured by a specific attributedThis ranking shows the relative importance of each attribute, with 1 being the highest ranked and 7 being the lowest ranked importance Preference weights and willingness to trade off—sensitivity analysis AE adverse events, CI confidence interval aRegression coefficients are from a conditional logistic model that regresses participants’ choice on the attribute’s levels of the treatment selected from each choice card. Coefficients may only be interpreted in terms of significance bWillingness to trade off was calculated using the formula −(βattribute/βitch). The number indicates how many percentage points in the probability of achieving meaningful itch improvement an average participant is willing to trade off to avoid a particular attribute. For AE attributes, the trade-off is expressed with respect to a 1-percentage-point reduction in the risk of having the AE. For example, an average participant is willing to trade off 1.3-percentage-point chance in achieving meaningful improvement in itch to avoid a 1-percentage-point risk of infections cRelative importance scores were normalized to percentages based on the difference between the best and worst attribute level, multiplied by the coefficient for each attribute, to allow for comparisons across attributes. Relative importance scores could be interpreted as the proportional value of a change from a treatment with the worst levels to a treatment with the best levels, captured by a specific attribute dThis ranking shows the relative importance of each attribute, with 1 being the highest ranked and 7 being the lowest ranked importance Discussion This DCE provides insights on the preferences of patients with moderate to severe CHE for treatment attributes spanning across efficacy, safety, and mode and frequency of administration. Patients preferred CHE treatments that maximize improvements in itch, skin appearance, and pain while minimizing risks of AEs, and favored topical application twice per day over one injection every 2 weeks. Of all the attributes assessed, itch improvement and skin appearance improvement were considered the most important attributes when considering CHE treatment options, followed by risk of cancer, risk of infections, and pain improvement. Meanwhile, mode and frequency of administration and risk of major heart problems ranked lowest in importance, while still having a significant impact on preferences. The importance of avoiding cancer risk was also evidenced in the willingness-to-trade-off analysis, in which an average patient was willing to accept a reduction in the probability of achieving itch improvement of more than 28 percentage points in exchange for a reduction of one percentage point in the risk of cancer, the highest willingness to trade off among all AEs considered, indicating how much participants value avoiding any cancer risk. Preference studies specific to the CHE population have been limited. One prior US DCE study conducted by Hauber and colleagues has assessed treatment preferences among adult patients with CHE [21]. In the study, patients rated efficacy improvements as more important than reducing the risks of specific AEs [21], which was consistent with the present findings that patients placed the most importance on efficacy attributes and that symptom relief, particularly itch and skin appearance, was central to patients’ perceived value of CHE treatments. Several distinctions in the study population and design between the Hauber et al. and our study are worth noting. Hauber et al. was conducted among patients with severe, topical-refractory CHE and focused solely on skin clearing as the efficacy measure along with a limited set of less serious AEs (e.g., headaches, hearing problems, vision problems, and bone problems) [21]. The present study encompasses a broader scope by incorporating multiple efficacy dimensions (itch, skin appearance, pain), different set of AEs (cancer, infections, heart problems), and treatment logistics (mode and frequency of administration), reflecting the evolving treatment landscape for moderate to severe CHE with topical and systemic JAK inhibitors and biologics [11]. Despite the limited number of existing DCE studies in CHE available for comparisons, our findings are in concordance with prior DCEs in AD, albeit most of the studies were not specific to atopic hand eczema. Across multiple DCEs among patients with AD [35–39], itch reduction, alongside chance of clear/almost clear skin and faster onset of action, consistently emerged as the top driver of treatment choice, in line with the present results that patients with CHE prioritized itch and skin appearance improvements. Patients from prior AD DCE studies showed aversion toward AEs; for instance, in a US DCE study among patients with moderate to severe AD (59% reported hands were affected), patients preferred treatments with a lower risk of all AEs evaluated, including cancer, infection, and heart problems [37], reinforcing our CHE finding that the risks of these AEs meaningfully shaped patient preferences. Findings on treatment logistics in prior AD DCEs were mixed—some, including the abovementioned US DCE study, showed low relative importance or nonsignificant impact on preferences [36, 37, 39], while others indicated a clear preference for oral once per day over injection every 2 weeks [35, 38], and one study found a statistically significant preference for topical cream twice per day over oral once per day in patients with mild to moderate AD [38]. In our CHE study, mode and frequency of administration ranked low in importance relative to other attributes but yet remained statistically significant. Specifically, participants preferred a twice-per-day topical regimen over an injection every 2 weeks, which may potentially reflect patients’ familiarity with topical treatments, as over 85% of participants were receiving topical treatment at the time of the survey. Together, these findings demonstrate the shared core priorities of symptom relief and safety in treatment preferences across chronic skin diseases while highlighting disease-specific nuances. As CHE has a multifactorial pathogenesis and distinct occupational and non-occupational exposures [3], consideration of individual- and disease-specific attributes when evaluating new therapies for patients with CHE is needed to ensure treatment decisions align with the outcomes most meaningful to this population. Prior studies have reported missed work and hindrances to daily activities due to CHE in Europe [4, 5, 7], whereas our study further quantifies the work and activity burden on patients due to their moderate to severe CHE in the USA. Despite the widespread use of current treatments, nearly all participants in this study experienced reduced activity levels, including common activities of daily living such as doing household chores and handwashing. In addition, over 90% of employed participants who worked in the past week reported at least some work impairment. These findings reaffirm that CHE extends beyond clinical symptoms, and its functional impact on work productivity and daily living underscores the need for more effective, durable, and convenient treatment options specifically approved for CHE to mitigate patient burden [4, 5, 7, 11, 40]. Given the impact of CHE on patients’ lives and potential variation in individuals’ risk tolerance, it is critical to incorporate patients’ preferences to guide treatment selection. Our study found that nearly one in four participants reported to be presented with only one treatment option by their healthcare providers when initiating their latest treatment, and that efficacy and side effect risks were discussed with only 76% and 60% of participants, respectively, suggesting opportunities to strengthen physician–patient communication. For instance, in a systematic review of evidence on ways to integrate patient preferences in health care, findings suggest physicians may ask open-ended questions to clarify patients’ values, show empathy to build trust and psychological safety, and share control of care in ways that validate patient autonomy and choices [41]. Tailored patient education and shared review of clinical evidence, such as implications of potential AEs associated with various treatment options, may also support engagement and informed shared decision-making [41]. Ultimately, these approaches may help improve treatment satisfaction and foster better adherence to treatment, which may in turn lead to better outcomes. Interpretation of the results should be made in light of study limitations. This online-panel-based survey study included only individuals accessible through the vendor’s panel who wished to participate in the study. As such, selection bias may exist, and the resulting sample may not be fully representative of the US population with moderate to severe CHE; nonetheless, the final sample included a diverse group of participants with a mix of CHE subtypes from all US regions. The survey also relied on self-reported diagnosis, which has potential for diagnostic misclassification and limiting generalizability to all patients with moderate to severe CHE. Nonetheless, the operational definition of CHE is objective and widely used in the literature [42, 43], and the severity criteria align with clinical guidelines [3] and dermatology expert input, which may mitigate the risk of substantial misclassification, supporting the overall validity of the findings. A limitation inherent to survey studies is the reliance on participants’ recollection of past events for data. Recall bias or errors in the accuracy or completeness of the recalled experiences may exist, particularly if past memories are influenced by more recent events. To minimize recall bias, participants in this study were asked to recall events that occurred in the recent past (e.g., in the last 7 days of the survey). Another study limitation is the limited number of key attributes included in the DCE questions, which aimed to minimize participants’ response burden. Although additional attributes may have been important for participants’ preferences, it should be noted that the attributes included in the present DCE were identified using a three-pronged data-driven approach via literature review, qualitative patient interviews, and clinical expert input, in an effort to include attributes most important to patients with CHE. Lastly, the use of a conditional logistic model is subject to limitations such as the assumption of preference homogeneity; thus, exploration of preference heterogeneity warrants future research. For instance, patients with prior systemic therapy exposure may have different risk tolerance compared with patients who were systemic-naïve; however, the present study was not designed or powered to support robust subgroup analyses, particularly given the imbalance in subgroup sizes. Future studies with larger and more balanced samples would be valuable to further explore this heterogeneity. Similarly, it should be cautioned that overall means may not reflect the preferences of individual patients. Conclusions In this quantitative DCE survey, patients with moderate to severe CHE prioritized efficacy attributes, including improvements in itch and skin appearance, when making treatment decisions, followed by avoidance of cancer and infection risks. The mode and frequency of administration also influenced preferences, with patients favoring topical application twice per day over an injection every 2 weeks, albeit to a lesser extent. All studied attributes had a significant effect on participants’ treatment choice, with all AE attributes assessed (risks of cancer, infection, and major heart problems) demonstrating negative preference weights, indicating a preference for CHE treatments with lower risks of these AEs. The study also found that nearly a quarter of participants reported only to be presented with one treatment option by their healthcare providers, and efficacy and side effects risks were not always discussed, suggesting a potential communication gap. This study provides insights to better understand patient preferences in the treatment for moderate to severe CHE that could be used as a basis for physicians to engage in shared decision-making with patients, including discussions of treatment trade-offs, which may help improve treatment satisfaction and foster better adherence to treatment.