Smartphone apps have recently become a growing field in the health care sector, garnering increasing attention and importance in health research. The use of such technologies can help optimize broad and location-independent health care in real time and support the prevention of chronic diseases and health promotion to improve adherence to therapy, quality of life, and clinical outcomes in the long term [1,2]. Smartphone apps not only help monitor health data [3,4] but also help change health behavior and develop new habits. For this purpose, health behavior apps offer the potential to technically integrate a variety of behavior change techniques (BCTs) [1,5,6]. BCTs are based on health psychology knowledge about the mechanisms of behavior change processes [7]. The most common BCTs in behavior change apps are, for example, self-monitoring by tracking the duration and frequency of a behavior and feedback on the behavior or outcomes by visualizing the progress graphically [1,5,6,8]. Feedback and the visualization of success also increase motivation in the users, which plays an important role in the long-term success of behavior change [5]. Behavior change apps can also be used to set goals and monitor their achievement. Automatic reminders provide prompts and support the development of routines in everyday life. Written, visual, and audio information can show users how to perform the behavior [1,5,6,8].

A known and widely discussed problem in research on behavior change apps is a low adherence and engagement when using these technologies [2,9-11]. To address these problems, user experience (UX) is becoming increasingly important in mobile health research [12-15]. UX is an overarching holistic concept and covers all emotional, cognitive, and physical experiences that users undergo when interacting with software products such as apps [12,16-18]. As shown in Figure 1, a high UX when using an app for the first time is associated with the willingness to continue using it and recommend it to others in the short term. A high UX leads to increased adherence and app use in the long term and, thus, can also improve clinical outcomes and the effectiveness of health-related interventions [14,19-21]. Therefore, the investigation of UX in the context of digital technologies in patient care is of great importance and should be considered early during the development process [14,15,22]. Thus, a thorough formative evaluation of the UX should precede a summative evaluation to optimize the app from the end user’s perspective [14].

As a multidimensional construct, UX comprises various dimensions that can differ depending on the underlying model, understanding, and context [15,17,19]. According to Thielsch and Salaschek [21], the dimensions content, usability, and aesthetics are particularly important in a research-related context. In addition, feelings and emotions that arise through interaction with the technical medium are described as an important aspect of UX [15,16]. As personal data are collected, passed on, and stored in digital care structures, trust in the technology is another predictor of future app use [21].

Apps to support behavior change and disease-specific self-management have also been developed and evaluated in the context of chronic skin diseases. These are usually aimed at specific groups, such as parents of children with atopic dermatitis (AD) [23,24]; patients with AD [25,26]; or people with specific areas of the skin affected, such as hand and foot eczema [27]. Weigandt et al [27] developed the first smartphone app in Germany specifically for hand and foot eczema, with which patients can monitor and manage their disease following a patient educational intervention by photographing their skin, tracking their quality of life and symptoms, and chatting with their dermatologist.

However, thus far, such technology does not exist for work-related skin diseases (WRSDs) even though they have been among the most common work-related diseases in Germany for decades [28]. Up to 90% of all diagnosed cases affect the hands in the form of irritant or allergic contact eczema, which often occurs in combination with a genetic predisposition to AD [29-32]. Work-related hand eczema is usually caused by hazardous activities in the workplace (eg, skin contact with allergens and irritants, wet work, high handwashing frequencies, and mechanical stress). Workers in the health care, metalworking, hairdressing, and construction sectors are at a particularly high risk, which results in high prevalence rates [30,31]. Due to the high relevance of WRSDs, the German Social Accident Insurance has developed a complex and hierarchical multistep procedure for patients with WRSDs, with different outpatient and inpatient prevention measures depending on the severity of the skin disease. In the case of severe or recurrent skin diseases, patients are offered to participate in an inpatient interprofessional rehabilitation program with a duration of 3 weeks. In German occupational dermatology, this program is also known as tertiary individual prevention (TIP) [29,33]. An important element of TIP are health educational interventions [34] with the aim of gaining disease-specific knowledge and increasing the motivation to implement and optimize individual skin protection behavior. In addition, individual strategies for skin protection behavior are developed with occupational therapists and practiced as part of a workplace simulation in occupational therapy. To restore the skin barrier and reduce the risk of recurrence of hand eczema, it is important to implement skin protection behavior (eg, regular use of skin protection and skin care products as well as reducing the frequency of handwashing by using mild detergents) in the long term [34-36]. However, changing skin protection behavior is complex and requires a high level of self-management after participating in the TIP as the new behavior must be successfully transferred to and implemented in the professional and private contexts. A structured maintenance program that supports patients in this subsequent implementation of skin protection behavior does not yet exist [34]. To fill this gap, we developed the Mein Hautschutz im Alltag (MiA) app (MiA translates to “My skin protection in everyday life”) to support the self-management of patients with WRSDs after the TIP [37].

The aim of this study was to evaluate the UX of and adherence to the MiA app and pilot the feasibility of an app-based maintenance program in our clinical setting. In this publication, we report the following results: (1) description of the intervention in a standardized manner; (2) UX, subjective quality, and perceived impact of the MiA app; and (3) adherence to the MiA app.

The occupational dermatology maintenance program is a complex intervention consisting of several interacting components (Multimedia Appendix 1). According to the Medical Research Council’s framework, the development of complex interventions is based on a 4-stage process consisting of development [37], feasibility, evaluation, and implementation [38]. This study focuses on feasibility (phase 2).

The maintenance program is offered by the Institute for Interdisciplinary Dermatological Prevention and Rehabilitation (iDerm), Osnabrück, Germany, which is a specialized center in Germany for inpatient and outpatient interprofessional treatment of patients with WRSDs.

The maintenance program consists of 2 main elements: individual goal-setting interview and the MiA app. The systematic development of these elements is described in detail elsewhere [37].

The individual goal-setting interview is conducted during the inpatient stay by a health educator face-to-face 5 or 6 days before discharge. The goals, which are defined by the patients, are entered into the app on the coaching platform by the health educator and can be viewed by patients via the app. During the subsequent onboarding, patients receive an introduction and explanations of the various app functions. Patients are also given access to a video via a QR code that shows the app with screenshots and explains the app’s functions.

MiA is a fully automated app with free access for study participants and consists of 6 components focusing either on interaction or information. The smartphone-based components with interactive nature are named My Skin Protection Goals (German: Meine Hautschutz-Ziele), My Skin Protection Behavior (German: Mein Hautschutz-Verhalten), and My Skin Documentation (German: Meine Hautdokumentation). The smartphone-based components To Listen (German: Hörenswert), Skin Protection 101 (German: Das Hautschutz 1x1), and My Accountabilities (German: Meine Zuständigkeiten) provide different information about the skin disease in different modes of delivery (podcasts, textboxes, and videos; Figure 2). In addition, users automatically receive a reminder message if no information has been entered into the app for a period of 14 days.

Multimedia Appendix 1 provides a more detailed description of each component according to the Template for Intervention Description and Replication checklist [39,40], including what, why, how often, and when components are unlocked during the maintenance program and where and which tailoring options are available. Table 1 describes the contents of the active components of the intervention in a standardized manner using the Behavior Change Technique Taxonomy version 1 [7,41].

This study was approved by the Ethics Committee of Osnabrück University (Ethics-50/2022 and Ethics-15/2023). All patients assessed for eligibility were informed verbally and in writing about the study and its voluntary nature. All participants had to sign a declaration of consent to take part in the study. The participants were informed that data collection would be pseudonymized and that the evaluation and publication of the results would be anonymous. They received no compensation for their participation.

Recruitment took place at iDerm in Osnabrück on the fourth day during the TIP. Study information was provided in patient groups as 7 to 9 new patients are usually admitted to the TIP each week. After all patients had been fully informed about the study, they were given sufficient time to decide whether to participate. Subsequently, interested patients received the written study information and declaration of consent. The inclusion criteria were (1) signed declaration of consent for study participation; (2) legal age (>18 years); (3) sufficient German language skills to understand the app content, participate in focus groups, and complete questionnaires; and (4) access to an internet-enabled smartphone (for step 2).

This study followed a user-centered and multistage iterative approach, which is a well-known and widespread procedure in the development of technologies in health care and UX research [42,43]. This approach allows for the consideration of the opinions and feedback of the target group as future users of the app in a participatory manner [23,44-46]. As illustrated in Figure 3, the entire process consisted of 2 steps with both qualitative and quantitative data collection.

Various instruments have been developed to assess the UX of mobile health apps. Among them, the user version of the Mobile Application Rating Scale (uMARS) is a well-known and reliable instrument that measures the UX of health apps from the user’s perspective [47]. The uMARS is based on the Mobile Application Rating Scale, which is used to assess the quality of apps by clinical or technical experts [48,49]. The uMARS is made up of 16 items and comprises the UX dimensions of engagement, functionality, aesthetics, and information quality. The scales of subjective quality and perceived impact assess additional information about recommendation; willingness to use and pay for the app; and information regarding awareness and knowledge of and intention for behavior change. For this study, an instrument for step 1 and step 2 was developed based on the uMARS and the model shown in Figure 1.

Statistical analyses of the quantitative data were carried out using SPSS (version 28; IBM Corp). We calculated a score for all 5 UX dimensions as well as a score for all UX items. Adherence to the My Skin Protection Behavior, My Skin Documentation, and My Goal Achievement functions was determined based on the adherence values in Multimedia Appendix 1 and by dichotomizing the data into adherent and nonadherent. For continuous data (eg, age and UX), we calculated the mean, median, IQR, and SD. We present categorical data (eg, gender, subjective quality, and perceived impact) in frequencies and percentages.

The aim of this study was to evaluate the UX of and adherence to an app-based maintenance program. The MiA app is a self-management tool for patients with WRSDs after participating in an inpatient rehabilitation program (TIP) to support change in and maintenance of skin protection behavior. To the best of our knowledge, the MiA app is the first app for people with WRSDs. Our study was conducted by following a multistage and iterative development and pilot-testing process, with this paper focusing on the results of phase 2.

UX research is an important part in the development of apps to identify and address potential barriers to the use of apps as early as possible. It should precede the summative evaluation [38]. However, many studies investigating the influence of app-based interventions in the dermatological setting only report on clinical outcomes such as quality of life, eczema severity, or itching [24,27] or app use behavior [25,26]. Zvulunov et al [24] point out that there are missing data on the usability of and adherence to apps in the field of chronic skin diseases.

We evaluated the UX using the dimensions of feelings and emotions, usability, aesthetics, content, and trust. Our study generally showed high to very high ratings for all dimensions with the exception of feelings and emotions, which was rated comparatively lower. Studies show that gamified apps can improve the interaction, motivation, and adherence [23,52]. However, the benefit of gamification differs between genders and decreases with age, making it difficult to successfully implement those functions in our heterogeneous study population [53]. According to Tolks et al [53], gamified apps are particularly useful for people with low motivation to implement a behavior, and Matterne et al [54] were able to show that patients with WRSDs have a high motivation to implement skin protection, which increased significantly during the TIP. Furthermore, in a focus group discussion in our study, patients expressed that the app should not contain any playful elements and should remain factual as it dealt with a disease. The extent to which the app is used also depends on the app’s BCTs. Milne-Ives et al [55] were able to demonstrate that some BCTs are associated with higher use of and engagement with an app. These included goal setting and feedback, self-monitoring, and social support, which are also part of the MiA app. However, in the future, the app could be expanded to include further interactive functions that allow users, for example, to assess their skin condition more accurately. One possibility might be to analyze the severity of hand eczema in photo documentation using artificial intelligence [24]. Weigandt et al [27] were able to show that the eczema severity determined via photos significantly correlates with live assessment by dermatologists. Therefore, photographs appear to allow for reliable results regarding the actual severity of hand eczema.

The presentation of information on the app was rated as particularly interesting by patients. We attribute this to the different ways in which the information was presented (podcasts, videos, and short questions and answers).

The descriptive results of the UX also show single exceptions regarding usability. Due to the small study cohort at T2, we were unable to test any associations between UX and the sociodemographic characteristics of the participants. However, digital health literacy plays an important role in usability. It is known that, in addition to app-specific factors, the use of an app may also depend on personal factors such as age, sex, educational level, and socioeconomic status [15,27,56]. Nationally and internationally, it has been shown that older people with a lower educational level and of a lower socioeconomic status tend to have a lower digital health literacy [57,58]. With an average age of 49.5 (SD 13.1) years and a high proportion of people with elementary and secondary school–leaving certification (24/42, 57%), our study cohort had characteristics that could indicate a lower level of digital health literacy. Therefore, a potential influence of digital health literacy on the use and handling of the MiA app is conceivable. Further research should be carried out to determine which socioeconomic characteristics in our field of application are particularly affected by low digital health literacy. This is important to provide patients with individual and tailored support from health care staff during the onboarding process [59]. Nevertheless, onboarding via personal contact in face-to-face meetings is a particular strength of our study because we were able to respond to the individual digital competence of the patients.

We found differences in adherence between the interactive functions of the app. The daily tracking of skin protection behavior was used more often than the weekly skin documentation and goal assessment. Furthermore, skin care and skin cleansing were tracked more frequently than skin protection. In addition, adherence to skin cleansing and skin care tracking decreased over the 12 weeks. The lower tracking of skin protection at the beginning might be explained by the fact that patients are at home for a period of 3 weeks after discharge and do not start working in their profession straight away. That means that the use of skin protection creams at work only becomes relevant from day 21 onward. The digital documentation of skin protection creams may not have increased from day 21 onward as the practical implementation of digital tracking during work is not feasible. However, in contrast to this, we observed a clearly higher proportion of formulated individual goals in the skin protection category than in the skin cleansing, skin care, support, and other categories. This emphasizes that the topic of skin protection is nevertheless an important aspect for patients after the TIP even if this cannot be reflected in skin protection tracking for the aforementioned reasons. Against this background, linking both functions appears to be useful to adequately cover all dimensions of skin protection behavior. There are various possible reasons for the less frequent use of weekly photo documentation and target assessment. First, both components are technically more complex and more challenging functions and require more time. One possible explanation for the low level of photo documentation could be that photos are not taken routinely by the users but, rather, when acute skin changes occur. The weekly rhythm we set may not correlate with the progression of the individual skin diseases. WRSDs occur in over half of cases with a comorbidity of AD and progress in intervals of varying length [33]. The goal assessment is also cognitively more complex because it requires reflection on the last behavior performed by the users and the decision of the appropriate response option. The reasons for the less frequent use of both functions should be further investigated, for example, by users and nonusers reporting on their experiences, motivation, and the usefulness of these functions (eg, in qualitative interviews [56]). Nevertheless, it should be kept in mind that high adherence to the app functions (eg, skin protection tracking) is not necessarily associated with behavior changes and better clinical outcomes. In everyday life, skin diseases are associated with a high level of adherence to therapy. Daily documentation may be too time-consuming to implement in practice, meaning that regular app use does not necessarily lead to better outcomes in terms of eczema severity or quality of life [27]. The average willingness to use the app approximately once a week supports this assumption. The use of the MiA app after the TIP can also be individual depending on the extent to which patients benefited from patient education during the TIP. Another influencing factor on app use could be the duration of the disease. TIP patients have usually had the skin disease for several years and have developed their own self-management strategies in the past. In the future, it could be investigated what are the use and benefits for patients with milder forms of WRSDs (eg, as part of outpatient care). The results on the perceived impact show that the MiA app can also be used as a tool for disseminating knowledge about skin protection–related content.

Overall, our results are consistent with those of other studies on dermatological apps and, thus, show expected results on adherence, which decreases over time [25]. Shah et al [26] showed that the average duration of use of an app for people with AD was approximately 6 weeks. The frequent short duration of use of app-based interventions and the associated lack of evidence of long-term effects is a general and well-known problem in this research context [2,26].

We were able to demonstrate a high quality of the app. Despite the low adherence, the results on the perceived effect also show promising indications that the app might influence relevant determinants such as motivation, knowledge, or awareness regarding the change in health behavior according to the subjective statements of the participants. However, the effect of the MiA app on skin protection behavior, quality of life, and eczema severity remains unclear thus far and should be further investigated in the future in comparison with a control group.

In addition to the limitations already mentioned, this study has further limitations that should be considered when interpreting the results. First, UX is a multifactorial and complex construct. There are different methods for assessing UX that have their own strengths and limitations [13]. While questionnaires are cost-efficient and provide quick results, dishonest answers and individual interpretations of the results can be disadvantageous. Focus group discussions are easy to conduct but dependent on the interviewing skills of those conducting them [14]. To measure UX, we used a validated instrument as a basis and adapted it to the context and target group of the app. We concede that the validity of the instrument and a direct comparison of the results with those of other studies is possible to a limited extent. The standardization of the definitions of the UX dimensions and the development of validated instruments for UX research is an important future research focus [19].

The participants in our study matched the sociodemographic characteristics of those of other studies on TIP [60]. Nevertheless, due to the small number of participants, the low use behavior, and a remaining cohort of 57% (24/42) at T2, it was not possible to calculate differences in UX and adherence among age, sex, and educational level.

The target group participated during the development and testing of the MiA app. Nevertheless, their involvement could have been even stronger in the form of agile co-design (eg, for prototyping the app before programming and technical implementation and for identifying realistic adherence [23,61-63]). However, such an approach was not possible due to the organizational processes and duration of this study.

We were also unable to determine the extent to which the informative functions (podcasts, questions and answers, and videos) were used. On the basis of other studies, it can be assumed that the use of educational elements also decreases after a few weeks [25]. However, podcasts activated over the course of the maintenance program could be a way to maintain use over time.

Our study shows a high UX of the MiA app in almost all dimensions. To increase interaction, functions can be added to the app (eg, use of artificial intelligence with photo documentation to allow for a clearer assessment of the skin condition). We refrained from using gamification due to the heterogeneous study population of patients with WRSDs. We also observed single exceptions in usability. Usability depends, among other things, on the digital health literacy of the user. Future research should investigate in this context which sociodemographic characteristics show lower usability to provide these groups with further support when using the app.

We obtained expected results in terms of adherence, with decreasing use over time, which corresponds to the current state of research in this field. Particularly after resuming work after 3 weeks, daily documentation of skin protection behavior appears to be too time-consuming. The effectiveness of the MiA app in terms of skin protection behavior, quality of life, and eczema severity should be further investigated with a control group.