Healthy lifestyle and behavior changes are difficult to initiate but even more challenging to sustain over time [1]. There has been an increase in lifestyle-related diseases such as obesity, cardiovascular diseases, and diabetes worldwide [2-4]. The estimated number of people with obesity nearly tripled between 1975 and 2016, with >650 million adults with obesity (BMI≥30 kg/m²) worldwide in 2016 [5]. Similarly, the number of people with cardiovascular diseases increased by >50% in the period from 1990 to 2019 [6], and people living with diabetes increased by >60% in the period from 2009 to 2019 [7].

For people with obesity, weight loss may be difficult; however, maintaining weight after weight loss appears to be even more of a challenge [8], and several factors (eg, environmental, biological, behavioral, and cognitive) [9-12] contribute to the complexity of this health problem. The impact and burden of obesity on an individual are substantial and multifaceted, as it can affect both health and well-being, with an increased risk of medical conditions, premature mortality, and reduced quality of life [13,14].

Lifestyle interventions focusing on diet and physical activity, as well as behavioral and cognitive strategies, are widely recommended for obesity management [12,15-20]. However, even when weight loss is achieved, mechanisms such as increased hunger, reduced energy expenditure, and reduced satiety frequently contribute weight regain [10,17]. As much as 30% to 50% of the initial weight that is lost during lifestyle interventions is often regained during the subsequent 2 to 3 years [21], and few manage to maintain their lost weight in the long term [22]. With several factors contributing to weight regain [9,10,23], solving the weight loss maintenance challenge appears to be a complicated endeavor.

Health behaviors and self-regulation play central roles in weight loss and weight loss maintenance [24,25]. Implementing sustainable and feasible behavior change and self-regulation strategies into daily life takes time and effort, and finding ways of initiating and maintaining behavior change over a long period is a complex undertaking [26].

Health behaviors also change over time and are often person and context related (eg, individual motives, habits, and social and environmental factors) [23,26]. With many people failing to maintain weight after initial weight loss, innovative approaches for long-term behavior change and weight loss maintenance are needed [10,27].

Digital interventions are increasingly being used to promote healthy lifestyles and improve health outcomes [28-31] and can be an accessible and feasible way of supporting behavior change through its availability and scalability. A digital health (ie, eHealth) intervention can be defined as a digital technology focusing on intervening in an existing situation aiming to change health behavior [29]. The potential of evidence-informed digital technologies supporting weight loss maintenance may be significant, as digital interventions can overcome the time and place barrier and adapt to a person’s context, needs, and preferences [32], consequently offering people the support needed for sustained behavior change [29,33].

Research exploring technologies for weight loss maintenance support is still at an early stage, and little is known about their long-term effectiveness, as few evidence-based eHealth interventions are available [34,35]. In addition, little is known about potential changes in user needs over time and how digital technologies can support end users in maintaining weight in an optimal manner for long-term behavior change [32]. In fact, knowledge is lacking on how to translate persuasive system design (PSD) principles and behavior change techniques (BCTs) into design features when developing digital interventions aimed at facilitating continued health behavior change and weight loss maintenance [35].

An important factor for the success of digital health research, development, and implementation is the early involvement of end users and other key stakeholders in the design and formative evaluation of a product or technology [29,36-39]. Human-centered design approaches such as design thinking [40,41] can be combined with principles from Agile software development [42], stimulating the collective creativity of end users and other key stakeholders (eg, designers, developers, researchers, and experts) for the rapid development and evaluation of digital health interventions [31,43-45].

Design thinking is an approach that emphasizes understanding and empathy with end users, multidisciplinary collaboration, and iterative involvement of stakeholders through generation of creative ideas and action-oriented rapid prototyping to create desirable, feasible, and viable innovative solutions [40,46]. Agile software development is a flexible approach that emphasizes active stakeholder involvement through rapid iterations to test assumptions and validate possible solutions to quickly learn and adapt to changes in needs [42]. Combining these design and development methods could therefore be a time- and cost-effective approach to explore and validate whether an innovative solution is desirable (ie, what users and stakeholders want) [40,42,43,47] and whether it solves the right problem (ie, meet user and stakeholder needs) for a problem-solution fit [48].

To support long-lasting behavior change through digital technologies, the design needs to support the user in adopting sustainable behaviors related to their individual goals and values [31,32]. Although goals relate to something a person would like to achieve, values can refer to what a person considers important in life [49], reflecting end users’ ideals or interests [50]. As such, values can be defined as the main drivers of behaviors or high-level needs [32]. Therefore, finding a solution that can help address end users’ goals by taking their key values and needs into account is important to create meaningful and effective eHealth technologies in support of continued health behavior change.

Complex interventions, such as digital behavior change interventions, usually consist of many active ingredients or interactive components [51-54] and can be designed to facilitate motivation and adherence to healthy behaviors [35]. PSD principles and BCTs can be such active ingredients or building blocks of digital behavior change interventions. To design motivating and effective digital technologies, integrated PSD principles and BCTs should match end user values and needs [32]. PSD principles are designed to influence users’ attitudes and behaviors [55] and can be applied to match user profiles to motivate and trigger health behavior change in the design of technologies [35,55,56]. By contrast, BCTs are designed to alter or redirect causal processes that regulate behavior [51,52], can be applied to any intervention focusing on behavior change to improve the health and well-being of people [29], and can be applied to daily life without technology [32]. As such, the PSD principles and BCTs overlap and complement each other [29,35]. Theoretical principles from the PSD model by Oinas-Kukkonen [55] and BCTs from the Behavior Change Taxonomy by Michie et al [52] can be translated into design features during the design and development of digital behavior change interventions to meet end user values and needs [29,32]. Such combined PSD [55] and BCT [52] features can be embedded in a digital application with the specific aim of forming, altering, or reinforcing healthy attitudes and behaviors [57].

However, research involving digital behavior change interventions often fails to clearly show how theories and techniques of behavior change have been combined and applied to design and practical delivery forms [31,35,57-61]. There appears to be a lack of theoretical frameworks and specifications of design features when reporting on digital interventions. It is often unclear which design features, PSD principles, and behavior change strategies are most effective in meeting end user values and needs and how they influence health-related outcomes, including weight loss maintenance [29,35,52,62-64].

In response to these gaps, the current research group performed a scoping review and a qualitative study aimed at identifying PSD principles and BCTs from eHealth interventions applied in existing weight loss maintenance research [35], as well as key values and needs related to what people want for maintaining weight and why [32]. The scoping review [35] identified PSD principles [55] (eg, self-monitoring, reminders, rewards, tailoring, personalization, and praise) and BCT clusters [52] (eg, feedback and monitoring, goals and planning, repetition and substitutions, social support, associations, and shaping knowledge) applied in eHealth interventions to stimulate adherence, motivation, and weight loss maintenance. The technology characteristics of existing eHealth weight loss maintenance interventions were usually supported by mobile phone technology, sometimes in combination with an activity tracker and wireless scale [35]. The subsequent qualitative study (ie, individual and focus group interviews) identified key end user values of people with the aim of maintaining weight loss in the long term (ie, autonomy, self-management, motivation, personalized care, happiness, health, feel supported, and positive self-image), as well as PSD principles and BCTs that might be essential to include in eHealth interventions to meet end user values and needs [32]. The findings indicated that the most successful and promising eHealth weight loss maintenance interventions entailed a combination of both PSD principles and BCTs [32,35]. The studies overlapped in findings [32,35]; however, some less frequently applied PSD principles (eg, rehearsal) and BCTs (eg, identity, self-belief, and natural consequences) were identified in the qualitative study, which might be of importance to support end user values and needs to prevent weight regain in the long term [32]. The qualitative study also highlighted the tailoring and personalization of digital interventions to address the often multifaceted and dynamic changes in individual needs over time (eg, related to behaviors, thoughts, and emotions) for continued health behavior change [32].

The results from these recent studies strongly suggest that the process of translating end user values into design features, as well as the exploration of how PSD principles and BCTs can be combined and implemented, are important parts of the design and development processes when aiming to create motivating and engaging digital interventions to support sustained behavior change and weight loss maintenance.

The overall aim of this study was to investigate how digital technology can meet end users’ values and needs by exploring and validating design features (ie, the combination and implementation of PSD principles and BCTs) through iterative design, development, and formative evaluation of a digital intervention called eCHANGE.

The following describes how 2 theoretical frameworks [52,55] targeting motivation and behavior change can be combined with applied innovation methodologies, such as design thinking [41,65] and Agile development [42], during the design of digital behavior change interventions. The findings show the development of digital technology aimed at facilitating sustainable behavior change to maintain weight after weight loss. The main research question of this study was as follows: how can an evidence-informed digital behavior change intervention, combining and implementing PSD principles and BCTs into design features, be designed and developed to support end user values and needs for long-term weight loss maintenance?

For this study, design features for (sustainable) behavior change were defined as the combination of PSD principles and BCTs embedded in a digital intervention, with the specific aim of supporting end users’ (ie, target users) values and needs to facilitate sustainable health behavior change (Figure 1).

To develop and deliver a digital intervention aimed at supporting long-term weight maintenance following weight loss (ie, eCHANGE; ClinicalTrials.gov NCT04537988), the process was guided by the Double Diamond (ie, design thinking process) [65,66] and the Center for eHealth Research and Disease Management (CeHRes) roadmap [29,36] (Figure 2 and Multimedia Appendix 1 [36,66]). In addition, refer to the study by Asbjørnsen et al [32] for additional background and details. To improve the uptake and impact of digital technology [29], multidisciplinary collaboration, human-centered design approaches such as design thinking and service design [41,46], Agile software development principles [42], and formative evaluation were integrated into a systematic and iterative development process to optimize the fit between values and needs, technology, and context [36].

This study focused mainly on the second diamond of the Double Diamond process (ie, develop and deliver; Figure 2) [66], particularly the Design and Operationalization phases of the CeHRes Roadmap, including formative evaluation to gather input to improve the intervention [29,36].

During the Design phase [36], the aim was to explore how digital technology can be developed and to validate which design features meet end users’ values and needs to support long-term weight loss maintenance. Knowledge, insights, and ideas from previous studies [32,35] were translated into prototypes through co-design workshops [41], rapid prototyping [46], and user testing [29] with end users and other key stakeholders. PSD principles from the PSD model [55] and BCTs from the Cross-Domaine Taxonomy (version 1) by Michie et al [52] were combined and implemented into design features during several iterative cycles, providing ongoing information on how to address end user values and needs [32] (Figure 1).

The Operationalization phase [36] included Agile software development (ie, Scrum development sprints) [42] and testing of the digital intervention and focused on how the technology could be delivered and further improved to prepare for a feasibility pilot trial [29]. The PSD model [55] includes 28 individual PSD principles, which are categorized into 4 categories: primary task support, dialogue support, system credibility support, and social support. In comparison, the Cross-Domaine Taxonomy by Michie et al [52] comprises 93 distinct BCTs divided into 16 theory-independent clusters. The 2 frameworks applied to target behavior change [52,55] can facilitate standardized reporting of the PSD principles and BCTs embedded in a digital behavior change intervention.

In line with the iterative approach to eHealth development [36] and the Medical Research Council guidance for developing and evaluating complex interventions [54], the data collection, analysis, and results were intertwined throughout the design and development processes (Figure 2).

Digital intervention development involved a multidisciplinary research and development team and was led by the study principal investigator (PI; LSN), a clinical psychologist with health psychology specialization and long-standing experience with digital behavioral interventions. The multidisciplinary team (14/45, 31%) entailed key stakeholders identified through stakeholder analysis [32] with diverse professional backgrounds and expertise, including researchers and clinicians in obesity and weight management, behavioral science, and eHealth; content editors; a digital designer; and software developers. In addition, a service designer was involved in facilitating service design workshops. Table 1 provides an overview of the multidisciplinary research and development team members, including their expertise.

End users were defined as people aged ≥18 years with a BMI of ≥30 kg/m² [67] before weight loss (ie, who had lost ≥8% of their body weight through a low-calorie diet or behavior change program) who were in need of support to prevent weight regain. People who met these criteria and were able to speak and read Norwegian were invited to participate in this study. Recruitment was conducted at 3 secondary or tertiary obesity research and treatment centers (ie, hospitals) in Norway through convenience sampling. In addition, end users who participated in a prior formative study [32] and a group of healthy volunteers were invited to participate in usability testing. To compensate for time spent and potential costs (eg, parking and transport), the study participants (ie, end users and healthy volunteers) received a gift certificate (ie, approximately US $25 and US $50 for individual testing and workshops, respectively).

Representatives of other key stakeholders identified during stakeholder analysis [32] (eg, health care providers and behavior change and obesity experts) were recruited based on convenience sampling through the collaborating obesity research and treatment centers.

This study was approved by the Hospital Privacy and Security Protection Committee (ie, institutional review board equivalent; approval number 2017/12702) at the Oslo University Hospital in Norway. All study participants (ie, end users and other key stakeholders) received written and oral study information and signed an informed consent form before participation.

The development process started with 2 predesign activities to prepare for the Design and Operationalization phases to develop and deliver the digital behavior change intervention. An overview of the design and development cycles during the eCHANGE intervention, including activity participation, is shown in Figure 3.

During the iterative intervention development (Figure 3), data collection and analysis provided ongoing information on how to improve the technology (eg, intervention content and operationalization of the design features). The prototypes evolved during co-design, rapid prototyping, and usability testing, where findings from one design cycle served as input to the next and set the direction for creation and validation of design features.

Rapid analysis [86] was applied during the design and development process to ensure that the collected data provided quick and thorough input to optimize the digital intervention (eg, actionable suggestions and specification of requirements through user stories). This included the structuring and summarizing of notes, including illustrative quotes and voice/video recordings from usability testing, into themes related to the design features to elicit input on how to combine and implement the PSD principles [55] and BCTs [52] into design features to support end user needs and preferences. In addition, a more in-depth directed content analysis [87] was performed to secure evidence-informed development and ensure that no themes were missed. This included the coding of study participants’ feedback and suggestions into predefined categories with respect to design, content, and functionality. The feedback and findings, including inconsistencies and conflicting needs, were first reviewed and discussed by the PI (LSN), 2 researchers (RAA and MLS), the product owner (MO), the digital designer, and/or a content editor to validate that the intervention content and design features (ie, PSD principles and BCTs) matched the identified end user values and needs. Thereafter, the findings were discussed (until consensus was reached) with participants from the multidisciplinary research and development team for continuous evaluation, prioritization, and improvement of features before the next iteration was conducted.

The product backlog reflected the prioritizations and decisions made by the research and development team. This was done to ensure that the selected design features and development of the intervention (ie, MVP) were in line with end user needs and preferences and that findings from previous research [32,35], evidence-informed knowledge, and feasibility considerations were taken into account before technical adaptations and updates were executed.

The smartphone-based eCHANGE intervention app was designed and distributed as a native app through official app stores for iOS and Android. Web technology in a Cordova container was used, and when in use, all information is stored locally and encrypted with the Advanced Encryption Standard algorithm in Galois/counter mode before being written to the local device file system.

The first time the app runs, a 256-bit encryption key is generated. Between app invocations, an Advanced Encryption Standard key-wrapped algorithm with a (wrapping) key acquired from the user’s personal identification number is used and stored on the device keychain (ie, based on an existing technology platform) [88]. The keychain ensures that the mobile device is protected with a 4-digit personal identification number code or optional biometric authentication (ie, face recognition or fingerprint).

When in use, log data (ie, system use, including navigation, frequency of use, and use of functionalities) and self-monitoring data are sent through an encrypted channel to a secure server (ie, Services for Sensitive Data, University of Oslo) for future analysis and summative evaluation. If interested, the user may elect to import relevant personal data (eg, number of steps taken per day and weight measurements) from Apple Health or Google Fit on their device; however, this integration is neither required by the research team nor necessary to realize the full potential of the app.

The eCHANGE intervention was developed by the Department of Digital Health Research at Oslo University Hospital in Norway in line with the national and international privacy and security standards and regulations (eg, Norwegian Digitalization Agency and European General Data Protection Regulations of 2018) [81-83]. A legal and privacy declaration was included as part of the information about the app in the settings functions in accordance with the existing requirements. The procedures applied in relation to the data protection impact assessment and risk assessment analysis of the technical solution were approved by the institution’s Department of Information Safety (approved in June 2020).

This section focuses on the design operationalization of the incorporated PSD principles and BCTs into design features and main intervention components to address the following research question: how can an evidence-informed digital behavioral change intervention, combining and implementing PSD principles and BCTs into design features, be designed and developed to support end user values and needs for long-term weight loss maintenance?

To meet the diverse and dynamically changing needs of end users (ie, people aiming to maintain weight after weight loss) [32], the eCHANGE intervention was developed as a personalized self-management intervention and delivered as a smartphone-based app for flexible and easily available weight loss maintenance support.

A total of 45 end users and other key stakeholders participated in intervention development, including 13 (29%) external stakeholders (ie, n=9, health care providers, including 1 health care manager and 4 healthy volunteers), 17 (38%) prospective end users, 14 (31%) stakeholders from the multidisciplinary research and development team, and 1 (2%) service designer. As presented in Multimedia Appendix 4, most end users were female (12/17, 71%), and the median age was 48 (range 30-63) years.

This study provides insights into the design and development of a digital behavior change intervention called eCHANGE, which aims to combine and implement PSD principles and BCTs into design features to support end user values and needs for long-term weight loss maintenance. The results revealed specific design features for sustainable health behavior change to prevent weight regain, combining PSD principles and BCTs, as well as how these design features could be operationalized into core components during the design and development of the digital intervention.

On the basis of participant feedback, 17 design features were identified in this study (Table 3) to support 8 previously identified key end user values for weight loss maintenance [32]. During the co-design and prototype sessions, the values of self-management, personalized care, and motivation received the most attention and feedback from end users. The identified design features were implemented in the digital intervention through 4 interconnected main components: Week Plan; My Overview; Knowledge and Skills; and a Virtual Coach and Smart, Tailored Feedback System.

The findings indicate that to support the identified, interconnected end user values and needs [32], digital weight loss maintenance interventions should include design features that focus on health and well-being (ie, not only weight); facilitate the generation of habitual behavior, self-regulation, autonomous motivation, knowledge, and skills; and provide positive, tailored support to maintain weight after weight loss in the long term. Existing research has pointed toward a need for such an approach [32,35] and that a combination of PSD, BCTs, and behavior change theories might facilitate the design of effective technology-based tools and strategies for behavioral obesity interventions [35,58]. However, to the best of our knowledge, this is the first study to show how PSD principles and BCTs can be translated into design features to support end user values and needs to maintain weight in the long term. In line with existing research [33,97], the findings indicate a need for a shift in goal focus from weight to health and well-being. This may suggest that technologies for digital weight loss maintenance should aim to support healthy lifestyles and sustained motivation in line with the self-determination theory, supporting psychological needs such as competence, relatedness, and autonomy [26,92,95]. Digital interventions incorporating features supporting habit formation in line with self-determined goals, individual values, and identity, as well as focusing on meaningful areas such as improved health, might also enhance a sense of purpose and facilitate long-term health behavior changes and weight outcomes [17,26,32,98-101]. The results also indicate that the application of PSD principles from the primary support, dialogue support, social support, and credibility support categories might be required to aid weight loss maintenance and that the application of PSD principles such as personalization and tailoring of design features is important to match individual challenges, goals, and key values.

The eCHANGE intervention was developed as a personalized (adaptive) digital intervention aimed at providing self-management support for long-term weight maintenance. The results illustrate how PSD principles can be combined with BCTs that have been identified [32,35] as effective and promising in supporting weight loss maintenance and how they can be integrated into a digital intervention based on multidisciplinary stakeholder feedback. Existing research has indicated that BCT combinations targeting motivation and persistence in health-promoting interventions might increase the chances of successful health behavior change [102]. A systematic review identifying active ingredients in complex behavioral interventions for adults with obesity also indicated that the inclusion of BCTs in interventions could be beneficial, facilitating assorted phases of the behavior change process [53]. As such, the adaptive intervention format of eCHANGE can potentially support the behavior change required to maintain weight, allowing the user to choose between a number of BCTs depending on individual needs.

Development of healthy habits and adjustment or breaking of less healthy habits are required to optimize and maintain the health benefits of weight loss, as well as maintain new weight, over time [103,104]. Therefore, self-regulation strategies and skills are essential to enhance when new physically active lifestyles or healthy eating patterns are not fully automated and are likely vital to maintain healthy behaviors and not relapse into previous habits [26,105]. The findings from this study show that targeting holistic aspects (eg, cognitive, emotional, social, and behavioral) of behavior change [103] during the design of digital technologies may be essential when aiming to deal with the often multifaceted challenges and needs of weight loss maintenance [23,32,98]. Informed by the self-determination and self-regulation theories [91,92,95], some of the features identified in this study focus on the generation of habitual behaviors through recognized self-regulation and habit formation techniques. The application of habit-forming techniques could free cognitive capacity to facilitate engagement in desired behaviors that may help adopt health-related behaviors to maintain weight loss in the long term [104,106,107]. The incorporation of BCTs in digital interventions, such as self-regulatory strategies to bridge the intention-behavior gap (eg, if-then plans/problem solving), habit-related techniques (eg, graded tasks), and self-belief (eg, focus on past success, self-talk), could also be important for targeting specific values (eg, self-management, motivation, and positive self-image) or mechanisms of action (eg, behavioral regulation and beliefs about capability) that may facilitate continued behavior change and weight gain prevention [15,25,32,104,108-112]. A literature review aimed at identifying links between BCTs and mechanisms of action [112] indicated that some of the core BCTs incorporated in the eCHANGE intervention could affect behavior change mechanisms (eg, beliefs about capabilities, behavioral regulation, and motivation), which is in line with identified weight loss maintenance values and needs [32]. This underlines the importance of keeping in mind the links among intervention content, design features, and the values and needs of the target group when designing and developing digital interventions for weight loss maintenance.

Human behavior is affected and shaped by a range of individual factors, including cognitive, psychological, biological, physical, and emotional factors; habits; values; motivational and demographic factors; and external factors such as environmental, cultural, social, and physical contexts in which behaviors occur [113]. Therefore, when designing digital technologies for sustainable behavior change, a fit between technological, human, and contextual factors is required [36,114]. In addition, the success or failure of digital interventions subsequently depends on whether individual end user needs are met [32,115]. The development of technology-supported programs for weight loss maintenance based on theory, evidence, and person-based approaches is gradually receiving increased attention [116-119]. The eCHANGE intervention’s value-based approach to eHealth development, in which PSD principles and BCTs identified to match end user values and needs [32] were gradually embedded into the digital technology through the design features, represents a major novelty. However, how persuasive and behavior change strategies are operationalized into design features may affect the usefulness and effectiveness of technology in terms of health outcome improvements. This points to potential issues, as the use of digital interventions that do not fit end user values and needs could affect intervention acceptance, adoption, and diffusion, indicating that a person-centered and iterative approach is needed [29,31,32,36].

A recent scoping review examining human-centered eHealth development indicated that because of the complexity of eHealth development, multiple strategies and methods should be combined in line with the research objectives when conducting such studies [115]. As such, a combination of theory, research-based, creative, and innovative methods, guided by the Double Diamond (ie, design thinking process) [65,66] and the CeHRes Roadmap [29,36] (Figure 2), can be applied to translate values and needs from ideation to the operationalization of design features. The application of co-design and Agile development methods during this process may increase the chance that the values and needs of (future) users are met [43,45,110]. The tailoring and personalization of digital technologies in line with the values of individuals might be a promising motivational strategy for continued health behavior change [32]. Therefore, the human-centered and value-driven approach can be useful in strengthening the self-regulatory capacity and autonomous motivation required to achieve sustainable behavior change [26,32,120]. The application of theory-based approaches [52,55] when developing complex digital behavior change interventions such as eCHANGE can also facilitate specification and transparency of the internal structure of the technology and may enable the development and evaluation of high-quality and effective digital behavior change interventions [29,35,121].

Future research should aim to investigate how digital technologies can be effective in facilitating sustainable behavior change and successful weight gain prevention and whether the identified design features, PSD principles, and BCTs contribute to long-term behavior change and sustainable health effects. In addition, while sustained technology engagement is not necessarily needed for a digital intervention to be useful, long-term weight loss maintenance supported by digital tools requires some degree of use. The degree of use needed and the design features that best support engagement and motivation for continued health behavior change and long-term weight maintenance are yet unknown. Therefore, future research should aim to evaluate technology engagement related to individual differences and the design features or intervention components that contribute to the creation of engaging technologies [122,123]. Hence, to optimize the impact of digital health interventions through personalized design, future research should also aim to provide more knowledge on how to tailor and personalize digital weight loss maintenance interventions (eg, design preferences) to support identified values and needs for long-term weight maintenance [32].

Finally, studies testing the actual use of the technology (eg, through log data analysis) may also provide knowledge related to end user preferences and needs (eg, latent, contextual, and future needs) and help identify which features may work best for whom and why to maintain weight. As such, a feasibility pilot trial may be a cost-effective way of acquiring more knowledge related to feasibility, desirability, and preliminary efficacy to support weight loss maintenance [54]. A feasibility pilot trial can also provide insights into user experiences and the actual use of the different components of technology to optimize and tailor digital interventions. Future research should also aim to evaluate the (long-term) effectiveness of weight loss maintenance and health-related outcomes through randomized controlled trials. To find the most effective combination of design features to support behavior change and weight loss maintenance over time, experimental designs, such as fractional factorial designs, could also be performed to test which features or combinations of features might work best for whom and when [29].

This study had some limitations. First, a few participating end users were still aiming to reduce their weight, although their initial goal for weight loss had been achieved. This could be a limitation for this study as end user needs during weight loss may differ from needs when aiming for weight maintenance [32,35]. However, the balance between weight loss and weight regain likely reflects real-life aspects and challenges related to weight loss maintenance, and this balance was also reflected in the technology design (eg, availability of back-on-track features).

Second, with respect to end user participation, more women than men participated in the study. There may be gender differences in weight loss maintenance strategies; therefore, this gender imbalance may limit the generalizability of the intervention to men, although several other key stakeholders participating were men, which might strengthen a more representative perspective.

Third, the participants in this study reported having a high school education or higher, which could indicate that this was a fairly well-educated sample. As such, the sample may not fully represent the diversity of the future end user population (ie, people with obesity aiming for weight maintenance after weight loss). However, existing research demonstrates inconsistent patterns of engagement and nonuse/attrition with respect to user characteristics (eg, education, age, and gender) [124,125]. In addition, the reported expectations and perceptions of a certain technology might not predict its actual use in practice [29]. To address these possible limitations, data from this study, as well as a previous study [32], were used to create user profiles (eg, need-based personas and user stories) representing potential end users and to guide the co-design and implementation process of adaptive, persuasive features to tailor the technology to individual preferences.

Fourth, because of limited time, resources, and privacy and security restrictions, some end user/stakeholder wishes (eg, personalized virtual coach, context-aware suggestions, social learning, and collaboration with other users) were not reflected in the current MVP. On the basis of the nature of Agile development, the selected design features were prioritized with respect to practical (eg, privacy and security issues) and cost-effectiveness considerations (ie, value-based prioritization based on stakeholder input).

This study has numerous strengths. The focus on high multidisciplinary stakeholder involvement and end user values and needs are such strengths. Formative design results from co-design workshops, prototype validation, and usability testing with end users and other key stakeholders provided input with respect to what to deliver, as well as how, to support weight loss maintenance through design features (eg, content, function, and design). This highlights the importance of multidisciplinary collaboration between end users and other key stakeholders during the design and development of digital interventions to meet end user values and needs. Another strength is the focus on, and novelty of, how PSD principles and BCTs can be combined and implemented into design features of digital weight loss maintenance interventions. The application of the design features identified in this study could also be relevant for other areas requiring continued health behavior change, as the identified needs might reflect universal values or drivers for sustained behaviors [32].

To the best of our knowledge, this study is the first to combine PSD principles and BCTs into design features during the design and development of an evidence-informed digital behavior change intervention to support end user values and needs for long-term weight loss maintenance. The findings indicate that personalized digital weight maintenance interventions should aim to support health and well-being by including design features and strategies supporting the self-regulation of behaviors, thoughts and emotions, habit formation, autonomous motivation, competence, skills, and tailored support. The design and development of the eCHANGE intervention can provide valuable input for future design and tailoring of evidence-informed digital interventions, even beyond digital interventions in support of health behavior change and long-term weight loss maintenance.