Atopic dermatitis (AD) is a chronic skin disease affecting more than 20% of children [
We aimed to determine whether a memory button and an app would affect the Quality of Life (QoL) and subjective and objective severity measures amongst AD patients over one month during the patient’s normal schedules of treatment.
The user survey was carried out in accordance with the Helsinki II Declaration and according to the national regulations, which stated it was not considered necessary to report the study to the Scientific Ethics Committee. The purpose of the survey was not to measure safety or effect of the CE-marked device [
This randomized, investigator-blinded, prospective observational feasibility study was conducted in Copenhagen and Aarhus, Denmark. The primary endpoints measurements were Dermatology Life Quality Index (DLQI), Patient Oriented Eczema Measure (POEM), SCORing Atopic Dermatitis (SCORAD) and Eczema Area and Severity Index (EASI).
All patients were recruited online using advertisements on Facebook and transferred to the website
Patients who did not fulfill the inclusion criteria or who were not able to attend in the time period or never answered phone calls were excluded.
Patients were informed verbally and in writing that we, in the process, would follow their treatment; however, they were not informed how the monitoring would be accomplished.
The SCORAD and the EASI index are validated to record the objective severity of AD [
The memory button is CE-marked [
In the study, patients were assigned to one out of three groups, 1: control group, no interference (
The patients were randomized into three groups based on their POEM score at the first screening consultation in order to distribute the patients equally in groups based on their eczema severity. No intervention was done until the first visit. All patients participated in two consultations with the same doctor, preferably 28 ± 3 days in between. SCORAD, EASI, POEM, and DLQI were assessed at both consultations. Depending on the randomization, they were also introduced to the button and/or app at the first consultation.
The group allocations were: group 1: control group, no interference; group 2: intervention group, participants only received the memory buttons without the app; group 3: intervention group, participants received the memory buttons and the app.
For statistical computing and graphical representations, the R-language was used in Bioconducter (R foundation). Average mean, SD, CI 95%, and
Nineteen and 99 patients were invited for the first screening consultation in Aarhus and Copenhagen, respectively. Of those, 102 showed up where 5 had absence without notice, and 11 cancelled on the day or prior to the consultation without being interested in booking a new consultation after a reminder email. At the first screening consultation, 6 patients were excluded from the first visit because they did not bring moisturizers for weighing or did not have the symptoms of AD according to Hanifin and Rajka criteria; hence, the total number of included patients was 96. A total of 84 patients also showed up for the second consultation and were included in the final results. All patients were verified by a blinded dermatologist via telemedicine. This resulted in the exclusion of one patient, leaving a total of 83 patients on which the analysis was performed. For all demographic’s parameters, no statistically significant differences were found between the three groups, except for the EASI scores where group 2 (
Patient demographics.
Variables | Group 1 ( |
Group 2 ( |
Group 3 ( |
|
---|---|---|---|---|
Age | 31.1 (26.4–35.8) | 32.8 (29.0–36.6) | 35.6 (29.0–42.1) | 0.505 |
Sex (%) | 0.907 |
|||
Female | 72.4 | 70.6 | 70.0 | |
Male | 27.6 | 29.4 | 30.0 | |
Weight | 71.5 (65.8–77.2) | 71.9 (70.6–73.3) | 71.0 (64.8–77.1) | 0.970 |
Height | 172.6 (169.2–176.1) | 170.5 (165.5–175.4) | 174.4 (170.3–178.5) | 0.263 |
BMI | 23.9 (22.4–25.4) | 24.7 (20.9–28.5) | 23.2 (21.6–24.8) | 0.411 |
Asthma (%) | 0.994 |
|||
Yes | 31.0 | 29.4 | 30.0 | |
No | 69.0 | 70.6 | 70.0 | |
Allergy (%) | 0.320 |
|||
Yes | 65.5 | 79.4 | 80.0 | x |
No | 34.5 | 20.6 | 20.0 | |
Smoker (%) | 0.932 |
|||
Yes | 13.8 | 11.8 | 15.0 | |
No | 86.2 | 88.2 | 85.0 | |
Education (%) | 0.856 |
|||
Primary school | 6.9 | 5.9 | 5.0 | |
General upper secondary school | 27.6 | 23.5 | 10.0 | |
Higher education (less than 3 years) | 13.8 | 11.8 | 10.0 | |
Higher education (3–5 years) | 27.6 | 38.2 | 35.0 | |
Higher education (more than 5 years) | 24.1 | 20.6 | 40.0 | |
No. of medical products | 3.1 (2.3–3.8) | 3.5 (2.8–4.2) | 3.6 (2.2–4.9) | 0.645 |
SCORAD | 28.0 (21.5–34.6) | 33.3 (26.5–40.2) | 27.8 (21.0–34.0) | 0.287 |
EASI | 4.7 (2.7–6.8) | 10.1 (6.2–14.0) | 5.7 (2.7–8.8) | 0.028∧ |
DLQI | 5.6 (3.7–7.5) | 7.0 (5.6–8.4) | 5.3 (3.3–7.2) | 0.289 |
POEM | 12.0 (9.9–14.1) | 13.2 (10.9–15.4) | 10.4 (7.1–13.6) | 0.283 |
Objective severity measures were statistically significantly decreased in all three groups at the second consultation (Figures
Doctors’ evaluation of objective AD severity by (a) EASI and (b) SCORAD. Patients’ own assessment of AD severity by (c) POEM (subjective) and (d) perception of Quality of Life (DLQI). Depicted is the amount of change between the two visits (time points) for each of the three groups (1: control group, no interference (
Absolute values of EASI, SCORAD, POEM, and DLQI at visits 1 and 2.
EASI |
EASI |
SCORAD |
SCORAD |
POEM |
POEM |
DLQI |
DLQI |
---|---|---|---|---|---|---|---|
Group 1 | |||||||
2.6 ± 5.4 | 2 ± 5.1 |
26.6 ± 17.3 | 17.6 ± 12.3 |
11 ± 5.5 | 12 ± 5.8 | 5.0 ± 5.0 | 6.0 ± 3.2 |
|
|||||||
Group 2 | |||||||
6.5 ± 11.1 | 3.7 ± 7.2 |
32.2 ± 19.6 | 25.8 ± 14.5 |
13.0 ± 6.4 | 10.0 ± 5.9 | 5.5 ± 4.0 | 6.5 ± 2.6 |
|
|||||||
Group 3 | |||||||
2.8 ± 6.4 | 2.0 ± 5.8 |
32.4 ± 14.3 | 18.5 ± 13.8 |
13.0 ± 7.1 | 9.0 ± 6.5 |
5.0 ± 4.4 | 6.8 ± 2.5 |
Within each group, significant change is depicted as follows at visit 2 compared to visit 1.
The study was designed to observe if a memory button and/or app would introduce a behavioral change for patients leading to increased effects of treatment plans interpreted as improved adherence. Patient adherence has, in general, been shown to decrease over time [
According to “the health belief model” [
“The health belief model” has previously been used successfully as a tool itself to create change in health behavior leading to improved adherence [
In this feasibility study, patients chose their preferred topical treatment and were provided with patient care and education on how to apply medications effectively. This resulted in reduction of severity following objective assessments of the AD and highly significant for patients who were also offered a memory button and corresponding app. Furthermore, patients reported a significant subjective effect if they did use the memory button and app. This indicates that digital solutions may have a place in clinical practice and may reduce nonadherence.
The data used to support the findings of this study are available from the corresponding author upon request.
KM Joergensen (MD), A Eiken (MD), M Malmstedt-Miller (MD), M Joergensen (MSc), A Schultz (MMSc), and JR Zibert (MSc, PhD) are employees of LEO Innovation Lab. All other authors have no conflicts of interest.
This work was supported by LEO Innovation Lab, a separate unit of LEO Pharma A/S. The authors would like to thank Ari Pall Isberg (MSc) and Azita Khorshad (MSc), LEO Innovation Lab, for statistical support.