Development and content validity of new patient-reported outcome questionnaires to assess the signs and symptoms and impact of atopic dermatitis: the Atopic Dermatitis Symptom Scale (ADerm-SS) and the Atopic Dermatitis Impact Scale (ADerm-IS)

Abstract

Objectives: Atopic dermatitis (AD) is a chronic, relapsing skin condition, with signs and symptoms that impact patients’ lives and are best measured from the patient perspective. Therefore, there is a need for AD-specific questionnaires that are consistent with Food and Drug Administration guidance and best measurement practices, assessing sign and symptom severity and associated impacts, to support treatment efficacy in regulated trials. The objectives were to develop patient-reported outcome (PRO) questionnaires assessing sign and symptom severity, as well as impacts of moderate-to-severe adult AD.

Methods: A targeted literature review and meetings with clinical experts (dermatologists) were conducted to identify AD-related sign, symptom, and impact concepts. Results were harmonized and used to construct two draft PRO questionnaires: the Atopic Dermatitis Symptom Scale (ADerm-SS; 11 items) and the Atopic Dermatitis Impact Scale (ADerm-IS; 10 items). The content validity and questionnaire content were evaluated via qualitative concept elicitation/cognitive debriefing interviews with adult patients with moderate-to-severe AD.

Results: From the literature (n = 13 articles), 13 sign and symptom and 43 impact concepts were identified, while 21 sign and symptom and 48 impacts were elicited from experts (n = 3). During the patient interviews (n = 15), 19 sign and symptom and 41 impact concepts were reported, the majority of which were evaluated by the ADerm-SS and ADerm-IS, thus substantiating the content of both questionnaires. Additionally, patients interpreted both questionnaires as intended by the developers.

Conclusions: The ADerm-SS and ADerm-IS can be regarded as content-valid PRO questionnaires for moderate-to-severe AD.

Keywords:

• Eczema
• Itch
• Sleep
• Symptom assessment
• Impact assessment

Introduction

Atopic dermatitis (AD) is a chronic, relapsing skin condition¹ characterized by pruritus, redness, and pain, typically on the face, the back of the knees, and inside the elbows, but which may also be widespread². These signs and symptoms have a considerable impact on quality-of-life (QoL) for those living with the condition¹. Although AD is less common in adults compared to children (up to 3% and 15–20%, respectively, globally)^1,3,4, adults with AD usually experience more intense signs and symptoms (e.g. pruritus) that can negatively impact daily function (e.g. sleep disruptions due to itchiness and scratching, decreased work productivity, social impact⁴) and overall QoL¹. A recently published report by the National Eczema Association (NEA) summarized the burden of the disease, including how pain and itch impact emotional and social functioning, QoL, work, academics, and economics⁵.

The authors sought to determine whether any of the existing patient-reported outcome (PRO) questionnaires were developed in a manner consistent with the US Food and Drug Administration (FDA) PRO Guidance⁶ and best measurement practices^7,8, and whether any would be appropriate to use as a daily and weekly diary for the assessment of the daily signs and symptoms, and weekly impacts of moderate-to-severe AD in adults to support efficacy and label claims in regulated clinical trials⁹. Because some of the most salient symptoms (e.g. pruritus, pain) and impacts (e.g. sleep disruption) of AD cannot easily be observed and evaluated by a clinician, the need for assessments from the patients themselves via PRO questionnaires is critical.

While AD-specific PRO questionnaires exist (e.g. Atopic Dermatitis Burden Scale Questionnaire for Adults [ABS-A], Dermatitis Family Impact [DFI] Questionnaire, Patient-Oriented Eczema Measure [POEM], Psychosomatic Scale for Atopic Dermatitis [PSS-AD]), and Quality of Life Index for Atopic Dermatitis [QoLIAD]), and have been evaluated by other groups (specifically, the consensus group to harmonize core outcome measures for atopic eczema/dermatitis; HOME)¹⁰, only the POEM focuses on symptoms of AD. The POEM assesses the frequency of the experience of symptoms and was developed prior to the release of the FDA PRO Guidance. With respect to the impact questionnaires, the DFI is completed by an observer (i.e. a parent or caregiver reports on a child’s experience on behalf of the child), the PSS-AD focuses on stress-induced flares of symptoms and treatment effects, the QoLIAD¹¹ assesses the impact of treatment on QoL, and, while the ABS-A was developed to assess burden in a broad sense (economic, work, daily life), it was not developed to be administered daily¹². To our knowledge, no disease-specific PRO questionnaires have been developed following the release of the FDA PRO guidance that assess the daily/weekly signs, symptoms, and impacts of moderate-to-severe AD.

Thus, the goal of the current research was to develop and establish the content validity of two new AD-specific PRO questionnaires to assess the signs, symptoms, and impacts of moderate-to-severe AD in adults. The AD Symptom Scale (ADerm-SS) and AD Impact Scale (ADerm-IS) were developed to be an electronic diary. The ADerm-SS includes three items to be completed daily and seven items to be completed weekly, all assessing signs and symptoms over the previous 24 h. The ADerm-IS includes three items to be completed daily, assessing impact over the previous 24 h, and eight items completed weekly to assess impacts over the past 7 days. To inform the development of these PRO questionnaires, the following qualitative research activities were conducted: a concept-focused literature review, consultative advice meetings with clinicians, and interviews with moderate-to-severe AD patients.

Materials and methods

This was a qualitative study, and the research activities described herein were conducted in three stages. In concept elicitation (CE), AD-related sign, symptom, and impact concepts were identified through a review of the literature and consultation with dermatologists. During questionnaire construction, results from the literature review and expert meetings were used to construct two draft PRO questionnaires to assess signs, symptoms, and impacts. During concept confirmation and content evaluation, qualitative hybrid interviews (i.e. interviews that comprised both CE and cognitive debriefing [CD] open-ended questions)^13,14 with adult patients of age ≥18 years with moderate-to-severe AD were conducted to evaluate the signs, symptoms, and impacts relevant and important to patients and to assess the content validity of the draft questionnaires (Figure 1).

Figure 1. Study flow diagram.

Concept elicitation

The goal of the CE stage of research was to document and gain a deeper understanding of AD-related signs, symptoms, and impacts, as reported by the literature and dermatologists.

Concept-focused literature review

A targeted search of the literature was conducted in MEDLINE, Embase, and PsycINFO in June 2016. The search terms included “atopic dermatitis OR eczema” AND “adult” AND “symptom OR sign OR impact”, and the search was limited to English-language studies in humans published in the previous 10 years. Supplemental grey literature searches were conducted in Google and Google Scholar. Abstracts returned from these searches were reviewed using a pre-defined set of selection criteria, and, through this process, publications were selected for full-text review. From these publications, primary signs and symptoms of adult AD, and associated impacts, were extracted and collated.

Qualitative meetings with clinical atopic dermatitis experts

Three dermatologists with expertise in AD, working in academic medical centers in the US, were invited to participate in 60-min telephone meetings. Trained facilitators utilized a semi-structured, open-ended meeting guide designed to elicit each clinician’s perspective on the signs, symptoms, and impacts of moderate-to-severe adult AD (i.e. concepts) experienced by patients they treat. More specifically, experts were asked to discuss in detail the common signs and symptoms experienced by adults with moderate-to-severe AD and the ways in which their patients’ lives are affected by these signs and symptoms (i.e. disease-related impacts). The experts were also asked to describe what they considered to be important treatment outcomes; which signs, symptoms, and impacts their patients reported being most bothersome; and which signs, symptoms, and impacts they considered to be the top five most important to their patients.

Audio-recordings from the meetings were transcribed verbatim, anonymized, and transcripts were subsequently uploaded into ATLAS.ti and coded¹⁵. Qualitative data analysis was conducted using a directed content analysis approach¹⁶, which consisted of a detailed reading of each transcript to identify and code expert discussion of sign, symptom, and impact concepts. The first transcript was coded independently by all coding team members, who then met to review the first coded transcript and discuss any discrepancies between coders until harmonization of all codes within the transcript was reached.

Questionnaire construction

Drawing upon data from the literature review and meetings with experts, the questionnaire development team met to select the sign, symptom, and impact concepts to be included in and assessed by the PRO questionnaires being developed. Concept selection was largely driven by (1) the frequency each sign, symptom, or impact was reported during the meetings with dermatologists; (2) whether the concept was reported by dermatologists as bothersome to patients with AD; and (3) whether the concept would be important to track in the context of a clinical trial or real world setting. During this process, two PRO questionnaires were constructed: the ADerm-SS (Table 1) and the ADerm-IS (Table 2). Once the concepts were selected, the formatting of the questionnaires was defined, including the recall period, instructions, phrasing of the items, response options, and mode of administration.

Table 1. Structure of the atopic dermatitis symptom scale (ADerm-SS).

Section	Concepts	Overall concepts	Recall period	Response options
Daily diary (Completed every night)	Itch during sleep hours	Signs/symptoms of AD at their worst	Past 24 h	11-point numeric rating scale (NRS) 0 = ‘No [sign/symptom]’ 10 = ‘Worst imaginable [sign/symptom]’
	Itch during awake hours
	Skin pain
Weekly diary (Completed once per week)	Skin cracking
	Pain caused by skin cracking
	Dry skin
	Skin flaking
	Rash (redness, blisters, bumpy skin)
	Skin thickening
	Bleeding
	Skin oozing

Table 2. Structure of the atopic dermatitis impact scale (ADerm-IS).

Section	Concepts	Overall concepts	Recall period	Response options
Daily diary (Completed every night)	Difficulty falling asleep	Sleep impacts of AD	Past 24 h	11-point NRS 0 = ‘Not difficult’ 10 = ‘Extremely difficult’
	Level of impact on sleep			11-point NRS 0 = ‘Not at all’ 10 = ‘Extremely’
	Bothersomeness of waking up at night			11-point NRS 0 = ‘Not bothersome’ 10 = ‘Extremely bothersome
Weekly diary (Completed once per week)	Limitation in household activities	Impacts of AD	Past 7 days	11-point NRS 0 = ‘Not limited’ 10 = ‘Extremely limited’
	Limitation in physical activities
	Limitation in social activities
	Difficulty concentrating			11-point NRS 0 = ‘Not difficult’ 10 = ‘Extremely difficult’
	Feeling self-conscious			11-point NRS 0 = ‘Not self-conscious’ 10 = ‘Extremely self-conscious’
	Feeling embarrassed			11-point NRS 0 = ‘Not embarrassed’ 10 = ‘Extremely embarrassed’
	Feeling sad			11-point NRS 0 = ‘Not sad’ 10 = ‘Extremely sad’

Concept confirmation and questionnaire content evaluation

Confirmation of the concepts included in the draft PRO questionnaires and content evaluation of the ADerm-SS and ADerm-IS were assessed through hybrid CE and CD interviews with adult patients with moderate-to-severe AD (n = 15). Sample size estimates were based on predictions about the number of patients needed to achieve saturation, which is the point at which no new relevant information is gained from additional interviews^7,17,18.

Ethics committee approval and recruitment

Ethics review board approval from the Quorum Review Independent Review Board (https://quorumreview.com/) was granted on 27 November 2016 and 9 December 2016 (QR#32117/1), and all study participants provided written consent. Participants were recruited from three clinical sites in the US (New Orleans, LA; St. Louis, MO; and Chicago, IL) and were included if they met the Eczema Area and Severity Index (EASI) score criteria (≥16) and body surface area (BSA) criteria (≥10%) for moderate-to-severe AD, and had a documented history (within 1 year prior to screening) of inadequate response to treatment with topical corticosteroids (TCS) or topical calcineurin inhibitors (TCI). Individuals were excluded if they had been previously treated with a Janus kinase (JAK) inhibitor or dupilumab; had been recently treated with TCS, TCI, omalizumab, H1-antihistamine treatments, or any investigational systemic treatment; or if they had other active skin diseases or infections requiring systemic treatment. These criteria were chosen to ensure that participants were similar to patients who will be enrolled in future clinical trials for moderate-to-severe AD.

Interview conduct and data analysis

During CE, participants were asked a series of open-ended questions about the signs, symptoms, and impacts they experienced due to AD to confirm the content of the questionnaires. During CD, participants were asked to complete the preliminary versions of the ADerm-SS and ADerm-IS questionnaires via screenshots of a handheld electronic devise using a “think-aloud” method^19,20, and to provide feedback on the instructions, items, response options, and recall period of each.

Each interview was audio-recorded, transcribed verbatim, de-identified, imported into ATLAS.ti¹⁵, and then qualitatively coded and analyzed in a manner similar to what was described earlier for the meetings with experts. Based on the content evaluation process, suggestions for questionnaire modifications were made.

The CE data were utilized to develop a patient-centered conceptual model (Figure 2)²¹, and the sign-, symptom-, and impact-related concepts described by participants were analyzed and tabulated with regard to (1) the frequency of participant report, (2) the words used by participants to describe the concept, (3) the bothersomeness of the concept to participants, and (4) the importance participants attached to improvement in the concept as an outcome of effective treatment. Additionally, the CE data were analyzed in regard to saturation, providing evidence that the study sample was large enough that additional interviews were unlikely to produce new relevant and important concepts^7,17,18.

Figure 2. Patient-centered atopic dermatitis conceptual model. *Napolitano M, Megna M, Patruno C, et al. Adult atopic dermatitis: a review. G Ital Dermatol Venereol 2016;151:403–411²⁷.

The CD data were analyzed and tabulated with regard to (1) any difficulties experienced by participants in reading and understanding the questionnaires; (2) whether participants felt the questionnaires comprehensively assessed relevant and important sign-, symptom-, and impact-related concepts; and (3) whether participants were able to interpret the questionnaires’ instructions, items, and response options in a manner that enabled them to provide meaningful responses to the items.

Results

Concept elicitation

Concept-focused literature review

Thirteen full-text articles were included in the analysis. From these, 13 sign and symptom concepts were identified: bleeding, blisters, burning, dry skin, fissures, inflammation, itching, pain, rash, redness, scaling, skin thickening, and swelling. Additionally, 43 impact concepts were identified and organized into eight domains: activities of daily living (ADLs), cognitive, emotional, financial, physical, sleep, social, and work/school. The most frequently reported impacts were sleep disturbances, followed by work/school activities, social withdrawal, anxiety, depressed feelings, embarrassment, and the inability to participate in ADLs.

Qualitative meetings with clinical experts

The expert advice meetings were conducted with three dermatologists who had 3–25 years of experience treating adult patients with AD and were working in US academic medical centers. A total of 21 signs and symptoms associated with moderate-to-severe AD were identified from the experts: bleeding, blisters, burning, dermatographism, discoloration/scars, excoriations, fissures, flaking, hair loss, inflammation, itching, nail changes, oozing/crusting, prurigo, rash, redness, skin thickening, pain, scaling, skin dryness, and swelling. All experts agreed that itching and discoloration/scars were the most bothersome to patients, and all experts ranked itching among what they considered to be the top five signs or symptoms associated with moderate-to-severe AD.

Experts also reported 48 disease-related impacts across 12 domains: ADLs, cognitive, emotional, leisure, physical, relationships, school, sexual, sleep, social, treatment, and work. The most frequently reported impact concepts were concentration, damage by alternative/non-evidence-based treatment, depressed mood, embarrassment, feeling helpless, frustration, scratching, self-esteem, sleep disruption, steroid phobia, stigmatization, work absenteeism (missed work), and work performance. All experts (n = 3, 100.0%) ranked sleep disruption among the top five impacts most important to their patients.

Experts were asked to discuss the treatment outcomes that they and their patients considered important, which included reduction in dermatitis, reduction in itch, patient happiness, and improvement in sleep.

Questionnaire construction

Drawing on data from the literature review and meetings with experts, the sign, symptom, and impact concepts were selected for measurement by the ADerm-SS and ADerm-IS. Both questionnaires are designed for self-administration as a patient diary completed on a handheld electronic device. Tables 1 and 2 summarize the structure of the ADerm-SS and ADerm-IS, respectively.

Concept confirmation and questionnaire content evaluation

The hybrid CE/CD interviews included 15 participants, interviewed in three waves (n = 5 each) across three clinic sites in the US. Participants’ ages ranged from 21.2–75.4 (mean = 47.3, standard deviation [SD] = 20.3) years. Approximately half were male (n = 8, 53.3%), and the majority were White/Caucasian (n = 11, 73.3%).

All (n = 15, 100.0%) met the EASI score criteria (≥16) and BSA criteria (≥10%) for moderate-to-severe AD, and had a documented history (within 1 year prior to screening) of inadequate response to treatment with TCS or TCI (Table 3).

Table 3. Participant demographic and health characteristics.

Characteristic	Total (n = 15) n (%)^a
Age (in years)
Range	21.2–75.4
Mean (standard deviation)	47.3 (20.3)
Gender
Male	8 (53.3%)
Female	7 (46.7%)
Ethnicity
No, not Spanish/Hispanic/Latino	14 (93.3%)
Yes, Puerto Rican	1 (6.7%)
Race
Black or African American	3 (20.0%)
White/Caucasian	11 (73.3%)
Other	1 (6.7%)
Education level
High school diploma (or GED) or less	1 (6.7%)
Some college or certificate program	7 (46.7%)
College or university degree (2- or 4-year)	6 (40.0%)
Graduate degree	1 (6.7%)
Work status
Working full-time	8 (53.3%)
Retired	4 (26.7%)
Working part-time	2 (13.3%)
Homemaker	1 (6.7%)
Participant-reported general health
Excellent	1 (6.7%)
Very good	7 (46.7%)
Good	5 (33.3%)
Fair	1 (6.7%)
Poor	1 (6.7%)
Criteria for severity^a
Eczema area and severity index ≥16	15 (100.0%)
Body surface area ≥10%	15 (100.0%)
Investigator opinion	11 (73.3%)

^a Patients were deemed eligible for the study if they were diagnosed with moderate-to-severe AD using one or more of the criteria listed. Counts are not mutually exclusive.

A total of 19 signs and symptoms of moderate-to-severe adult AD were reported by participants during the CE portion of the interviews (Table 4). All symptoms assessed by the ADerm-SS were reported by participants (as well as by the literature and experts): itching, dry skin, flaking, skin cracking, bleeding, pain, oozing, pain from skin cracking, rash, and skin thickening. Additionally, participants reported experiencing redness, burning, bumpy skin, scaly skin, blisters, inflammation, discoloration, stinging, and tenderness. Participants also indicated symptom bothersomeness and which symptoms should improve with treatment.

Table 4. Participant-reported signs and symptoms of moderate-to-severe AD.

Sign/symptom	Frequency of participant report; (n = 15) n (%)	Participants indicated sign/symptom was bothersome or rated it among the top five most bothersome signs/symptoms^a (n = 15) n (%)	Participant indicated that he/she would like to see sign/symptom improved with treatment^Table Footnoteb	Exemplary participant quotations	Reportedby experts	Identified inliterature
Itching	15 (100.0%)	15 (100.0%)	✓	‘It itches when and once you start itching it does not want to stop …’ 30 year old female	✓	✓
Redness	10 (66.7%)	7 (46.7%)		‘… it’s red, blotchy … The color more reddish than your normal skin color.’ 53 year old male	✓	✓
Skin dryness	10 (66.7%)	7 (46.7%)	✓	‘My skin just gets really, really dry like … Impossibly dry.’ 24 year old female	✓	✓
Skin flaking	8 (53.3%)	6 (40.0%)		‘It flakes . It’s almost like if you have dandruff …’ 71 year old female	✓	✓
Burning	7 (46.7%)	4 (26.7%)	✓	‘… it burned . it was kind of like that when you put … medication on a cut or something.’ 56 year old male	✓	✓
Skin cracking	7 (46.7%)	4 (26.7%)	✓	‘… it looks like surfaces of an outer planet you know there are things here and what those are those are what the, what I call, what I’m calling fissures … So you’ll see a different color at the top than you will mid-way … And then at the bottom it’s red.’ 68 year old male	✓	✓
Bleeding	5 (33.3%)	1 (6.7%)		‘The red blood is where it’s, where the crack has cracked … you see the red blood there … And if you dab it, it will … It will come up, yeah.’ 68 year old male	✓	✓
Bumpy skin	5 (33.3%)	4 (26.7%)		‘… the bumpiness, you know, start breaking out like little tiny bumps all over …’ Subject 30 year old female	✓	✓
Pain	5 (33.3%)	3 (20.0%)		‘Oh, eczema is painful all the time … All the time like if I break out on my feet it, it hurts to put on and take off shoes.’ 25 year old male	✓	✓
Scaly skin	5 (33.3%)	2 (13.3%)		‘… as a kid my, my brother probably said it best as alligator skin … And it’s terrible but you know it’s, it’s rough, it’s dry, it’s, Uh—it’s, it’s just, it’s just a terrible feeling.’ 25 year old male	✓	✓
Blisters	4 (26.7%)	3 (20.0%)		‘… tiny little blisters sometimes like little bubbles … Tiny little bubbles.’ 53 year old male	✓	✓
Rash	4 (26.7%)	1 (6.7%)		‘The breakout, it was like, it almost looked like a, some kind of a rash.’ 71 year old female	✓	✓
Inflammation	4 (26.7%)	2 (13.3%)		‘… so the inflammation is like when they start it is red and swellingness … I believe the inflammation is like when it is swollen to me …’ 30 year old female	✓	✓
Oozing	4 (26.7%)	1 (6.7%)		‘… and it starts to ooze … Oozing is like, it almost looks like pus.’ 71 year old female	✓
Pain from skin cracking	4 (26.7%)	2 (13.3%)	✓	‘… the shirt would get caught into my skin … Because uh, the skin was breaking so bad … And that uh, yeah, that was very painful.’ 21 year old male
Skin thickening	4 (26.7%)	1 (6.7%)		‘Because the skin gets thick … if you look the skin looks real thick there … it feels rougher and it looks … thicker.’ 68 year old male	✓	✓
Discoloration	2 (13.3%)	1 (6.7%)	✓	‘It will leave with a darkening of my complexion where it was …’ 30 year old female	✓
Stinging	2 (13.3%)	1 (6.7%)		‘… it’s not as bad as a bee bite. If you ever get bitten by a bee … It stings, it just stings.’ 56 year old male
Tenderness	1 (6.7%)	1 (6.7%)		‘… tenderness … it’s really not painful but it’s just, um, like sensitive kind of like to my touch …’ 68 year old male

^a Participants either reported sign/symptom as bothersome during the open-ended discussion about sign/symptoms, or ranked sign/symptom as among their top five most bothersome signs/symptoms of atopic dermatitis when specifically asked “of the symptoms you currently experience, which are the top five symptoms that are most bothersome to you?”

^b During open-ended discussion about signs/symptoms, a sub-set of participants responded to the following question: “Please think about all the signs and symptoms that you experience or have experienced that we have discussed today. Which signs or symptoms would you like to see improve if an effective treatment was available to you? Why?”

Participants reported a total of 41 disease-related impacts across 11 domains: ADLs, cognitive, emotional, family, financial, physical, recreation, relationships, school/work, sleep, and social (Table 5). All impact concepts assessed by the ADerm-IS were reported by participants.

Table 5. Participant-reported impacts of moderate-to-severe AD.

Impact concept	Frequency of participant report; n = 15 n (%)	Concept description	Reportedby experts	Identifiedin literature
Activities of daily living
Clothing choice	14 (93.3%)	An impact on one’s choice of clothing	✓
Avoiding substances or products that irritate skin	7 (46.7%)	Avoiding substance or products that may irritate the skin (e.g. soaps, perfumes, cleaning products, water)	✓	✓
Diet	3 (20.0%)	Monitoring diet to avoid certain foods that trigger AD	✓
Managing/treating condition	3 (20.0%)	Additional steps in daily routine taken to manage and/or treat condition (e.g. applying moisturizer or medication)
Cleaning	1 (6.7%)	Additional cleaning required as a result of skin flaking
Appearance	1 (6.7%)	Additional steps taken to manage appearance
Writing	1 (6.7%)	An impact on one’s ability to use writing utensils
Cognitive
Concentration	2 (13.3%)	An impact on one’s ability to focus or concentrate on tasks	✓	✓
Drained	1 (6.7%)	Feeling drained of energy due to preoccupation with AD
Emotional
Annoyed	6 (40.0%)	Feeling irritated or aggravated by AD
Self-conscious	6 (40.0%)	Feeling uncomfortable in one’s own skin	✓
Embarrassment	5 (33.3%)	Feeling embarrassed by the visibility of AD	✓	✓
Angry	4 (26.7%)	Feeling frustrated or agitated due to AD
Discomfort	2 (13.3%)	Feeling uncomfortable (emotionally) due to AD
Feeling different	2 (13.3%)	Feeling abnormal, or different from others, due to AD
Sad	2 (13.3%)	Feeling sadness or depressed feelings due to AD	✓	✓
Alienated	1 (6.7%)	Feeling alienated by others due to AD
Fear of condition spreading	1 (6.7%)	Fear that AD may spread to other parts of one’s body
Reduced self-confidence	1 (6.7%)	Feeling less confident in oneself due to AD
Stress	1 (6.7%)	Feeling stressed by flaring of AD	✓	✓
Family
Not wanting to have children	1 (6.7%)	Not wanting to have children, due to fear that AD will be genetically passed down to children
Financial
Treatment cost	2 (13.3%)	Additional costs of treatments used to manage AD		✓
Physical
Scratching	14 (93.3%)	Scraping or rubbing the skin with hands or foreign objects in order to relieve the symptoms of AD	✓	✓
Infection	3 (20.0%)	Increased risk of infection at sites of AD	✓	✓
Mobility	3 (20.0%)	An impact on one’s ability to move freely due to AD
Tiredness	2 (13.3%)	Feeling physically drained or lethargic due to AD	✓	✓
Recreation
Swimming	5 (33.3%)	An impact on one’s ability to go swimming due to AD
Sports	4 (26.7%)	An impact on one’s ability to play sports due to AD
Avoid sun	3 (20.0%)	An impact on one’s ability to tolerate sun exposure due to AD
Outdoor activities	2 (13.3%)	An impact on one’s ability to participate in outdoor activities due to AD
Relationships
Dating	2 (13.3%)	An impact on one’s ability to date (e.g. difficulty opening up to potential partners)	✓
Intimacy	2 (13.3%)	An impact on one’s ability to experience intimacy with one’s partner	✓
Acquaintances	1 (6.7%)	An impact on one’s interactions with acquaintances (e.g. acquaintances perceiving AD as infectious)
Sexual activity	1 (6.7%)	An impact on one’s ability to engage in sexual activity	✓	✓
School/work
Missed work	2 (13.3%)	Absences from work caused by AD	✓	✓
Fatigue	1 (6.7%)	Experiencing fatigue at work due to AD
Job choice	1 (6.7%)	An impact on one’s choice of job or career due to AD (e.g. changing job due to impact of AD)	✓	✓
Sleep
Waking from sleep	9 (60.0%)	Waking from sleep due to itching related to AD	✓
Difficulty falling asleep	8 (53.3%)	Difficulty falling asleep due to itching related to AD	✓	✓
Social
Avoid social activities	7 (46.7%)	Avoiding social activities due to symptoms of AD	✓	✓
Negative attention	7 (46.7%)	Experiencing negative attention from others in response to AD (e.g. staring)	✓	✓

Participant-reported signs, symptoms, and impacts were organized into a patient-centered conceptual model (Figure 2).

Saturation analysis found that all 19 (100.0%) sign and symptom concepts, and 95.1% (n = 39) of all impact concepts reported by participants emerged during the first 75% of interviews. The two concepts that emerged during the final 25% of interviews either overlapped with concepts that were previously elicited or were unique to a single participant. Taken together, these results provide evidence for the adequacy of the sample size.

During CD, participants were asked to complete the ADerm-SS and ADerm-IS and provide feedback on the recall period, instructions, items, and response options. Following each wave, results were reviewed to determine whether revision of either questionnaire was warranted. No revisions were made between interview waves, but, following completion of the interviews, two modifications were suggested: revise instructions asking participants to complete the diary “before you go to bed at night” to “each night”, and the instructions asking participants to complete the diary “once a week before you go to bed at night” to “once a week at night”. A decision about the recommended revision will be made following evaluation of data on the draft questionnaires after future quantitative analyses are conducted.

In the debriefing of the ADerm-SS and ADerm-IS daily and weekly items, generally participants interpreted the instructions, items, and response options as intended by the developers.

The CD results demonstrate that participants were able to complete and comprehend both the ADerm-SS and ADerm-IS as intended. All of the sign and symptom concepts evaluated by the ADerm-SS were found to be relevant to the AD experience, as described during CE. In addition, the majority of participants reported experiencing the impact concepts evaluated by the ADerm-IS. These results substantiate the content validity of the ADerm-SS and ADerm-IS PRO questionnaires.

Discussion

The results of this research support the content validity and use of the ADerm-SS and ADerm-IS as endpoints in clinical trials of patients with moderate-to-severe AD. This qualitative development work (including the conduct of conceptual research across multiple data sources: literature, experts, and patients) addresses the need for PRO questionnaires specific for moderate-to-severe AD and developed in accordance with regulatory guidelines⁶ and scientific best practices⁸. While AD-specific tools are already in existence, many were developed prior to the FDA PRO Guidance. While some include some of the key concepts that emerged from the current qualitative work, none assess all of the key signs, symptoms, and impacts of moderate-to-severe AD. In addition, none of the existing tools were designed to assess the daily and weekly symptoms and impacts of the condition.

The ADerm-SS and ADerm-IS questionnaires include daily items that are assessed using a 24-h recall period and weekly items that are assessed every 7 days. Minimizing the recall period was essential to reduce the potential for bias in the assessments, which aligns with regulatory and scientific recommendations^6,22. Previous literature has documented that concepts such as pain are susceptible to recall bias, as people tend to remember and attend to the most salient times, specifically the times when the concept is experienced the most (e.g. highest pain), and the most recently experienced time (e.g. the pain experienced when the person is answering the questionnaire)^23–25. Therefore, to evaluate the symptoms of itch and skin pain, and the impact on sleep, the ADerm-SS and ADerm-IS use a 24-h recall period for these daily items, administered via an electronic device (which includes time/date stamps to track the compliance with the diary assessment schedule). In addition, to reduce patient burden for the daily diary, additional symptoms of AD that have less daily variability (e.g. bleeding, dry skin, flaking, oozing, rash, skin cracking and pain, and skin thickening) were designed to be administered by the ADerm-SS on the electronic diaries every 7 days, but still use a 24-h recall period. Finally, the ADerm-IS includes physical and emotional impacts that were also assessed each week, but using a 7-day recall period. In this way, the tools were designed to track the key signs, symptoms, and impacts of moderate-to-severe AD, minimizing recall bias and patient burden. Too, the fact that the ADerm-SS and ADerm-IS questionnaires both have items assessed daily and weekly may be a potential source of confusion for respondents, especially those with a low education level. To mitigate this potential, the questionnaires are intended to be completed on an electronic device which would prompt respondents to complete the items necessary for that particular day.

The signs and symptoms assessed by the ADerm-SS are in line with recently published articles from the NEA⁵ and the HOME initiative¹⁰, including pain, which was noted by both groups as a concept that should be explored more fully in future research. From the summary of the international consensus meeting of the HOME initiative, while QoL questionnaires were reviewed by the group, no impacts/QoL questionnaire was recommended for inclusion in the core set of PRO questionnaires at that meeting¹⁰. Given this lack of consensus in the field, the ADerm-IS was developed to be appropriate for patients with moderate-to-severe AD, and to assess the sleep impacts daily and physical and emotional functioning impacts weekly.

A limitation of this research is that the ADerm-SS and ADerm-IS were not developed to be used with pediatric patients or those with milder AD. Further research would be necessary to evaluate the use of these questionnaires in those broader samples. Another limitation of the present study is that the hybrid CE/CD interviews included a single participant with an education level of high school or lower. As noted above, given the different recall periods and administration schedules for each questionnaire, care should be taken to train all participants, included those with low education, on how to complete the questionnaires electronically before implementation in future studies. Finally, it is important to note that the first version of the questionnaires was developed prior to the patient interviews, due to timeline constraints of the clinical development program. However, during meetings with experts, they were queried on the signs, symptoms, and impacts that their patients report as more important or bothersome. In addition, the methodological approach of developing a draft questionnaire based on a targeted literature review and expert input, and then subsequently conducting hybrid CE/CD interviews to confirm the key concepts of assessment and evaluate the comprehensibility and comprehensiveness of the questionnaires does align with best measurement practices and has been used in other PRO development studies^13,14,26.

Conclusions

The ADerm-SS and ADerm-IS can be regarded as content-valid PRO questionnaires for the assessment of the signs, symptoms, and impacts of adult moderate-to-severe AD. Next steps in the development of the ADerm-SS and ADerm-IS involve an evaluation of their psychometric properties and development of score interpretation guidelines for use in the target patient population, the results of which are intended to be reported separately.

Transparency

Declaration of funding

Study design, conduct, and financial support were provided by AbbVie Inc., and AbbVie Inc. participated in the interpretation of data, review, and approval of the manuscript.

Declaration of financial/ other interests

ES reports grants, personal fees, and non-financial support from Eli Lilly and Company, grants and personal fees from Anacor Pharma, Glaxo Smith Kline, Regeneron Pharmaceuticals, Sanofi Genzyme, Pfizer, Leo, and Eli Lilly and Valeant Pharmaceuticals, personal fees from AbbVie, Celgene Corporation, Dermira, Galderma, Genentech, Leo Pharma, Menlo Therapeutics, and grants from MedImmune, Novartis, Roivant Sciences, Tioga Pharmaceuticals, Vanda Pharmaceuticals. AB, HDT, and NT are employees of AbbVie Inc. and may own AbbVie stock or stock options. LLK and CF are employed by Adelphi Values LLC, which received payment from AbbVie Inc. to support the research activities presented in this manuscript. Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Acknowledgments

The authors would like to acknowledge Roger E. Lamoureux and Masami Kelly of Adelphi Values, and Clark Jackson, formerly of Adelphi Values, all of whom contributed to this research and provided scientific writing assistance. Data availability statement: The authors confirm that the data supporting the findings of this study are available within the article.