Skip to main content

Theory-driven development of a medication adherence intervention delivered by eHealth and transplant team in allogeneic stem cell transplantation: the SMILe implementation science project

Abstract

Background

Medication adherence to immunosuppressants in allogeneic stem cell transplantation (alloSCT) is essential to achieve favorable clinical outcomes (e.g. control of Graft-versus-Host Disease). Over 600 apps supporting medication adherence exist, yet they lack successful implementation and sustainable use likely because of lack of end-user involvement and theoretical underpinnings in their development and insufficient attention to implementation methods to support their use in real-life settings. Medication adherence has three phases: initiation, implementation and persistence. We report the theory-driven development of an intervention module to support medication adherence (implementation and persistence phase) in alloSCT outpatients as a first step for future digitization and implementation in clinical setting within the SMILe project (Development, implementation and testing of an integrated care model in allogeneic SteM cell transplantatIon faciLitated by eHealth).

Methods

We applied Michie’s Behavior Change Wheel (BCW) and the Capability-Opportunity-Motivation and Behavior (COM-B) model using three suggested stages followed by one stage added by our team regarding preparation for digitization of the intervention: (I) Defining the problem in behavioral terms; (II) Identifying intervention options; (III) Identifying content and implementation options; (IV) SMILe Care Model Prototype Development. Scientific evidence, data from a contextual analysis and patients’/caregivers’ and clinical experts’ inputs were compiled to work through these steps.

Results

(I) Correct immunosuppressant taking and timing were defined as target behaviors. The intervention’s focus was determined within the COM-B dimensions Capability (lack of knowledge, lack of routine), Opportunity (lack of cues, interruptions in daily routine) and Motivation (lack of problem solving, trivialization). (II) Five intervention functions were chosen, i.e. education, training, modelling, persuasion and enablement. (III) Twenty-four behavior change techniques were selected, e.g., goal setting, action planning and problem solving. (IV) Finally, seventeen user stories were developed to guide the SMILeApp’s software development process.

Conclusion

Our example on the theory-driven development of an intervention module in alloSCT delivered by eHealth and transplant team using a rigorous 3 + 1-stage approach based on BCW, COM-B and agile software development techniques, can be used as methodological guidance for other eHealth intervention developers. Our approach has the potential to enhance successful implementation and sustained use of eHealth solutions in real-life settings.

Peer Review reports

Background

Allogeneic stem cell transplant (alloSCT) is a well-established, potentially curative treatment modality for various malignant and non-malignant hematological diseases in which a patient’s diseased blood-producing system is replaced by that of a healthy person [1]. These patients’ complex care needs demand innovation in care delivery [2]. As information technology widely available through smartphones has the potential to support patients in adopting and developing essential behaviors such as those surrounding medication adherence [2, 3], the National Institutes of Health Hematopoietic Cell Transplantation Late Effects Initiative [2] recommend that related interventions include the support of eHealth solutions.

Medication adherence is defined as “The process by which patients take their medications as prescribed, composed of initiation, implementation and discontinuation” (Fig. 1) ( [5], p. 697). Initiation means that the patient starts taking a prescribed medication. Implementation means the correspondence of a patient’s actual dosing compared to the prescription. ImplementationFootnote 1 errors can take the form of late, skipped, extra, or reduced doses or ‘drug holidays’ (skipping several doses in a row). If the patient quits the prescribed medication for any reason(s), this is called discontinuation. The time span between the first and last dose is called persistence [5].

Fig. 1
figure1

Concept of medication adherence with initiation, implementation and persistence [4], based on [5]. From Annals of Internal Medicine, De Geest S, Zullig LL, Dunbar-Jacob J, Helmy R, Hughes DA, Wilson IB, Vrijens B, ESPACOMP Medication Adherence Reporting Guideline (EMERGE), 169, 1, 30–35. Copyright© [2018] American College of Physicians. All Rights Reserved. Reprinted with the permission of American College of Physicians, Inc

Medication adherence is vital regarding clinical outcomes in patients following alloSCT [6, 7]: as many as 80% of alloSCT patients develop an acute or chronic Graft-versus-Host Disease (GvHD) [8, 9], a serious complication in which the donor’s immune cells attack the recipient’s body [1] leading to increased morbidity and mortality [8,9,10,11]. Although there is limited evidence and the few existing studies did not distinguish between the three medication adherence phases (initiation, implementation, persistence), we found medication adherence to immunosuppressants in alloSCT essential to prevent or treat complications such as acute and chronic GvHD [12, 13]. The correlation is extremely compelling: In patients without chronic GvHD only 15.6% showed medication non-adherence to immunosuppressants while in patients with mild GvHD already 66.7% were non-adherent. In patients with moderate GvHD 74.1% showed medication non-adherence and in patients with severe GvHD even 88.9% were non-adherent [12]. As alloSCT receive their first doses of immunosuppressants during their hospitalization, the initiation phase of adherence is not an issue [6, 7]; adherence enhancing interventions need to focus on post-discharge implementation and persistence of medication intake [12]. While available evidence on medication adherence in alloSCT populations is limited, we know that the prevalence of overall non-adherence to immunosuppressants in adult alloSCT patients is 64.6% with 33.3% taking, 61.2% timing and 4.1% dosing non-adherence, 3.2% drug holidays and 3.1% discontinuation [12].

Findings from randomized controlled trials (RCT) in other chronic disease areas, including solid organ transplantation, positively link behavioral interventions using eHealth with improvements in medication adherence [14,15,16], unplanned inpatient acute care admissions [16], costs [16], general health outcomes (e.g., blood glucose or blood pressure) [17] and overall quality of life [16, 17]. The most successful behavior change techniques (BCTs) (i.e., active intervention elements) used in these trials were self-monitoring of and feedback on medication intake [15,16,17,18,19], reminder cues [15, 18, 19], goal setting, action planning and problem solving [14,15,16,17, 20]. To date, though, no interventions focusing on these elements have been developed and tested in alloSCT.

Moreover, compared to the wealth of findings emerging from trials, comparatively little information is available on successful implementation and subsequent evaluation of medication adherence interventions in general and eHealth supported behavioral interventions in particular in real-world settings. A 2018 review on high quality medication adherence intervention trials in different patient populations found that most of them did not report on essential implementation research elements, hindering the adaptation and implementation of trial findings into real-world setting [21]. Another 2018 review paper found 681 apps to support medication adherence in the Apple App and Google Play Stores [22]. However, only 84 (12.3%) of these were developed by or in collaboration with health care professionals, only eight (1.2%) were theory-based, and none reported patient involvement in their development process [22]. And regardless of whether they were developed based on a specific theory, roughly two-thirds contained zero to two BCTs. Of apps including BCTs, 96% included reminders – which are often unneeded or unwanted by patients [23]. Some higher-value BCTs such as self-monitoring or feedback on behavior were included in a minority (36–39%), but techniques such as social support or information about the health consequences were very rare (1–2%) [23]. Combined with developers’ general omission of end-user involvement (especially from patients) [22, 24], general shortages both of theoretical underpinnings [2, 25] and of contextual information to adapt eHealth interventions for implementation in specific settings [26,27,28] very likely contribute to two major concerns: 1) More than two-thirds of patients who use health-related apps abandon them within 3 months [24]; and 2) Very few health behavior support interventions are ever successfully implemented in real-world settings [15, 18, 19, 29,30,31,32].

In order to support post-alloSCT medication adherence effectively and sustainably, eHealth supported medication adherence interventions need to be planned from start to finish with a clear focus on one overarching outcome: implementation in daily clinical alloSCT practice. Accordingly, our major goal with this paper is to report the development of an eHealth facilitated care medication adherence module (implementation and persistence phase) for alloSCT. To do so, we have employed the Behavioral Change Wheel (BCW) and the Capability-Opportunity-Motivation and Behavior (COM-B) model – both of which, unlike other behavioral theories (e.g. Theory of Planned Behavior and Health Belief Model [33]), provide multilevel perspectives and include emotional factors [34]. In addition, we have developed user stories that facilitate the translation of research findings first to digitalization of the intervention, then to standard practice among both clinicians and patients.

Methods/results

The SMILe project

This work is part of the international, interdisciplinary, multi-phase, multicenter SMILe project, which aims at reengineering follow-up care to patients with alloSCT into an integrated care model first at the University Medical Center Freiburg, Germany and in a second step at the University Hospital of Basel, Switzerland (https://nursing.unibas.ch/de/forschungsprojekte/forschung/forschung/smile/). The SMILe integrated care model builds on the eHealth Enhanced Chronic Care Model (eCCM) [3]. Its major aim is to introduce follow-up based on a chronic care model by providing self-management support, optimizing delivery system design (e.g., increasing continuity of care), clinical decision support and clinical information systems.

As these functions involve considerable time and expertise, we recommend the introduction of a Care-coordinator (CC) to perform them, along with the use of a SMILe e-platform including the SMILeApp and the SMILeCare monitoring component for the CC. For the first year post-alloSCT, in addition to connecting patients to the CC in the transplant center, the app will allow patients to assess relevant biomedical and behavioral parameters on a continuous basis.

Four intervention modules are integrated to support self-management and promote alloSCT patients’ target health behaviors: Monitoring and Follow-Up, Infection Prevention, Physical Activity and Medication Adherence. This paper reports on the development of the medication adherence module. This patient-centered, theory-based development approach will increase the likelihood of adoption and implementation in practice.

Guiding methodology

Our intervention development process used an iterative approach applying the three stages of the BCW and one additional stage for the SMILe Care Model Prototype Development (Fig. 2), with each stage of execution reviewed by the entire SMILe research team [34, 35]. The various stages and steps will be further described in the next section (“The BCW”). Intervention development was informed by data from our previously performed multi-methods contextual analysis and evidence. Contextual analysis refers to the mapping of relevant multilevel contextual factors for the implementation of an intervention, e.g. local infrastructure, leadership, motivation of the stakeholders [27]. This means the inquiry of a specific context (in our case alloSCT follow-up care at the University Hospital of Freiburg, Germany), practice patterns as well as the attitudes and behaviors of all parties involved. Our preliminary work included such a contextual analysis using a cross-sectional quantitative survey of 60 alloSCT patients and five clinicians working in alloSCT. That analysis also included qualitative data, namely individual interviews with another ten alloSCT patients and three alloSCT clinician focus groups. The detailed description of the methods and results of this contextual analysis were recently published elsewhere [36]. It studied all relevant contextual aspects of the proposed integrated model of care, e.g., socio-cultural aspects at the micro (i.e., patient) and meso (i.e., transplant center) levels. Further, we examined the feedback of 21 stem cell transplant patients and eleven caregivers collected during three feedback rounds in Freiburg im Breisgau (Germany). Both, allogeneic and autologous stem cell transplant recipients were invited to give feedback as they participated a shared self-help group. We presented them our preliminary results and asked them to rank e.g. the importance of barriers to medication adherence or whether the proposed BCTs were feasible for them. The patients’ and caregivers’ feedback was included in the further development of the intervention. In addition, we reviewed existing evidence on 1) definition, prevalence and consequences of medication non-adherence in alloSCT (quantitative, appendix search diagram 1), 2) barriers and facilitators of medication adherence in alloSCT and solid organ transplantation (qualitative, appendix search diagram 2), and 3) existing interventions to support medication adherence in alloSCT and solid organ transplantation (quantitative, appendix search diagram 3). Because evidence in alloSCT is very limited, but medication adherence tasks are similar in alloSCT and solid organ transplant patients, we also included pertinent literature from solid organ transplantation.

Fig. 2
figure2

The 3 stages and 8 steps of the BCW [34] with our added stage 4 (self-developed figure)

The BCW

One particularly rigorous and widely useful system of developing and implementing behavior change interventions is the BCW [37, 38]. Combining 19 behavior change frameworks, the BCW is used to identify, understand and explain behaviors and their influencing factors. It consists of three stages: 1) understanding the behavior, 2) identifying intervention options and 3) identifying content and implementation options, which are further divided into eight steps (Fig. 2) [34].

BCW stage 1: understand the behavior

Step 1: define the problem in behavioral terms

Before attempting to modify a behavior it is necessary to understand it in detail and to identify possible starting points for change [34]. Therefore, the first step of the BCW is to define the problem in behavioral terms using three leading questions: What is the behavior?; Where does it occur?; and Who is involved?

In order to answer these questions, we first deepened our understanding of the problem by performing a scoping review of studies reporting definitions, prevalences and consequences of medication non-adherence in alloSCT (appendix, search diagram 1). From the twelve studies we identified, we extracted information on definition and relevance of medication non-adherence.

Results. Particularly regarding immunosuppressive regimes we found medication adherence essential in alloSCT to prevent or treat complications such as acute and chronic GvHD [12, 13]. However, almost 65% of participating patients struggled with medication intake, especially immunosuppressants taking and timing (implementation phase of medication adherence) [12] (Table 1).

Table 1 Definition of the problem in behavioral terms

The contextual analysis data of our preliminary work (see Guiding methodology section) [36] indicated that alloSCT patients were aware of the importance of following the medication regimen closely, especially when family, friends and caregivers were involved in medication management. According to clinicians, medication adherence was not systematically assessed at the studied transplant center. They also acknowledged a need for a person such as an Advanced Practice Nurse (APN) to coordinate follow-up, i.e., a CC [36]. The insights obtained from the literature review and contextual analysis enabled us to define the problem in behavioral terms (Table 1).

BCW step 2: select target behavior

The second step is to create a list of behavioral components known to influence the problem. From this list, behaviors are selected based on their expected impact (when improved) on target outcomes, the likelihood that they can be changed, the extent of expected spillover effects on other health behaviors and their measurability [34].

Results. In the literature search we identified 12 quantitative studies about prevalence and consequences of different behavioral components (appendix, search diagram 1). Based on the information from that literature review, we compiled a list of potential target behaviors that influence implementation and persistence adherence. For example, taking, timing and dosing immunosuppressants correctly (three behaviors), following food considerations concerning immunosuppressants and not taking drug holidays (implementation dimension). From this list, we selected correct taking and timing of immunosuppressants as the two most prevalent and important target behaviors. This selection was guided by a systematic rationale: changing these behaviors was expected to have the best overall combination of direct impact on clinical outcomes, spillover effects on other behaviors, e.g., taking no drug holidays, and ease of measurement (Table 2). We based our rating system (i.e., ++ very promising, + promising, ± not promising but worth considering, − unacceptable) on evidence from the literature and the research team’s clinical expertise.

Table 2 List of possible target behaviors

BCW step 3: specify the target behavior

As the third step is to examine the selected behaviors (in our case, two) both from the patient’s perspective and within the context of the surrounding system [34], we specified each in terms of who, what, when, where, how, how often and with whom they occurred. This choice of details was based on previously evaluated evidence from the literature [12, 13, 41], but mainly on our contextual analysis [36] and our research team’s clinical expertise.

Results. The resulting details (Table 3) allowed identification of target behaviors [34].

Table 3 Specification of target behaviors

BCW step 4: identify what needs to change

The fourth step is to identify what needs to change by analyzing risk factors not only at the individual but also higher in the system. For this step, the COM-B model (Fig. 3) – the center of the BCW (Fig. 4) – supports behavioral analysis very well, as it acknowledges that behaviors such as medication non-adherence are influenced by the physical and psychological capability, physical and social opportunity as well as reflective and automatic motivation. Capability is the capacity to engage in a certain health behavior (e.g., medication intake); opportunity refers to factors external to the individual that make behavior possible; and Motivation refers to the brain processes – whether reflective or unconscious – that direct behavior. Two common examples of capability are cognitive functionality (psychological capability) and the physical capability to swallow medication. Opportunities may lie in the accessibility of the medication (physical opportunity) or support from a partner (social opportunity); and motivational aspects would include attitudes about the medication, such as beliefs about its efficacy (reflective motivation) or treatment fatigue (automatic motivation).

Fig. 3
figure3

The COM-B framework to understand a behavior [34] (open access figure)

Fig. 4
figure4

The behavior change wheel [34] (open access figure)

Of course, capability and opportunity also affect motivation, and the behavior itself influences all three adherence components (Fig. 3). As an ecologic model (i.e., involving factors on the patient, health care provider, health care organization and health care system level), the BCW is helpful in understanding behaviors’ determinants and emotional drivers and provides the basis for the following steps.

In addition, the COM-B can be combined with the Theoretical Domains Framework (TDF) [34], which synthesizes multiple behavioral theories and includes 14 domains of behavioral influence (Fig. 4). Each domain represents theoretical constructs such as knowledge, skills or beliefs and relates to one COM-B component. Used together, the COM-B and the TDF allow a behavioral diagnosis, which facilitates the choice of effective behavior change interventions [34].

Where exactly to initiate behavior change was determined partly by the results of the contextual analysis and partly by qualitative evidence. We located the latter via a scoping review for qualitative findings on alloSCT patients’ and clinicians’ perspectives, barriers and facilitators to medication adherence. Due to limited evidence in the alloSCT population we also included literature from the solid organ transplant population (appendix, search diagram 2).

Results. According to the findings of the 16 identified qualitative studies and the contextual analysis, implementation of medication adherence is influenced by a broad set of multilevel factors, which we organized within the COM-B model (Table 4). For instance, lack of knowledge and lack of routine (Capability), lack of cues and interruptions in daily routine (Opportunity), lack of problem solving, trivialization and denial (Motivation) were identified as possible barriers to medication adherence implementation.

Table 4 Barriers of medication adherence sorted by COM-B

We crosschecked and compared our review’s results with those of two recent quantitative studies on multilevel determinants of medication non-adherence in heart transplantation [62]. In addition, we analyzed data from a long-term bio-psychosocial follow-up in solid organ transplantation [64] in which our research group was involved. The results supported our previous findings.

According to the findings of the 16 identified qualitative studies [48,49,50,51,52,53,54,55,56,57,58,59,60,61, 63, 65], the results of our contextual analysis and patient feedback, we selected the highest-priority barriers. The behavioral diagnosis findings for each domain and necessary changes are listed in Table 4.

BCW stage 2: identify intervention options

BCW step 5: identify intervention functions

After identifying the target needs, the relevant intervention functions (Fig. 4) must be identified. The BCW lists nine of these (education, persuasion, incentivization, coercion, training, restriction, environmental restructuring, modeling, and enablement), each of which is a set of effective intervention categories with the potential to mitigate barriers to behavior change identified by the COM-B and/or TDF models.

To guide our selection, we applied the APEASE criteria (test of Affordability, Practicability, Effectiveness and cost-effectiveness, Acceptability, Side effects/safety, and Equity) [66] on each intervention function (Table 5). The identified components were rated (i.e., ++ very promising, + promising, ± not promising but worth considering, − unacceptable) with consideration for available evidence, the research team’s clinical expertise, our contextual analysis and patient feedback. The APEASE criteria were also applied in further steps, always using the same data sources mentioned here. To search for existing interventions that support medication adherence in alloSCT, we performed a systematic literature search (appendix, search diagram 3). Due to limited evidence in the alloSCT setting, we also included findings from the solid organ transplant population.

Table 5 Applying the APEASE criteria to select useful intervention functions

Results. According to eleven quantitative studies we identified on medication adherence-enhancing interventions, combinations of education, training, enablement, modeling, environmental restructuring and persuasion were effective functions to support medication adherence [15, 18,19,20]. After considering the feasibility of each according to our contextual analysis, we excluded environmental restructuring, as this was not feasible within the SMILe project’s context. We then presented the selected functions to alloSCT patients and caregivers to discuss their feasibility and usefulness. The research team applied the APEASE criteria to the functions favored by the patient and caregiver group. They found that education, training, enablement, modeling and persuasion to have the greatest potential to improve medication adherence in post-alloSCT patients (Table 5).

BCW step 6: identify policy categories

In this step, policy categories are identified (Fig. 4). The BCW lists seven such categories (communication/marketing, guidelines, fiscal measures, regulation, legislation, environmental/social planning, service provision), from which researchers can choose those that will best support delivery of the intervention functions selected in step 5 [34]. We based our selection of policy categories on the results of our contextual analysis.

Results. After applying the APEASE criteria once again we selected two policy categories – regulation and service provision – as those most congruent with our chosen intervention functions (Table 6).

Table 6 Applying the APEASE criteria to select useful policy categories based on contextual analysis

BCW stage 3: identify content and implementation options

BCW step 7: identify behavioral change techniques

The seventh step is to identify BCTs, which are the active elements of an intervention. The BCT Taxonomy describes 93 BCTs, classified in 19 categories (e.g., goal setting, action planning, problem solving, information about health consequences, self−/monitoring of behavior, self-reward, punishment) to change a behavior. Thanks to the BCT Taxonomy, uniform terminology can also be used to describe intervention components, allowing replication of a study [34]. Based on the findings of the previous BCW steps and the data sources explained above, we selected the BCTs that we judged would best support our chosen intervention functions.

Results. In the literature search, we identified 11 quantitative studies on interventions (appendix, search diagram 3). According to the results, cognitive-educational interventions (e.g., information and instructions) were frequent, but showed inconsistent results in enhancing medication adherence [18, 20]. Behavioral interventions (e.g., counseling, reminders, self-monitoring and feedback on medication intake) led to significant improvements in medication adherence [15, 18,19,20]. While combinations of cognitive-educational, behavioral and psychological-affective interventions showed the best improvements, even these results were inconsistent [18, 19]. To select suitable BCTs for our intervention, we applied the APEASE criteria to those we judged most promising (Table 7).

Table 7 Applying the APEASE criteria to select useful BCTs in relation to COM-B and TDF

Based on these findings, we prepared a draft of the intervention, i.e. visualized the selected BCTs (Table 7) including a short description and a self-drawn representation of an app on a PowerPoint presentation. We showed the presentation to stem cell transplant patients and caregivers and discussed the various parts’ feasibility and usefulness. The participants had the possibility to vote for the different BCTs via a real-time smartphone-based survey (using mentmeter.com) as well as via the accompanying oral group discussion. The patients’ feedback indicated that they saw the intervention as both feasible and supportive. With consideration for all of the information thus far compiled, we selected 24 patient-level BCTs, including goal setting, action planning and self-monitoring of behavior. All selected BCTs are listed in Table 8.

Table 8 The selected intervention functions, policy categories, BCT and delivery mode relating to COM-B and TDF

BCW step 8: identify mode of delivery

The eighth and last of the BCW’s core processes is to identify the most suitable mode of delivery for the selected BCTs. This determines how the intervention should be delivered to the alloSCT patients and their caregivers, i.e., the end-users. A wide variety of delivery modes can support an intervention’s implementation and effectiveness, ranging face-to-face individual or group sessions to mobile phone apps [34]. The optimal delivery mode was determined using information generated from the previous steps, the contextual analysis and end-user feedback [34].

Results. Our contextual analysis [36] showed that alloSCT patients were open to technological assistance but emphasized that eHealth support should not replace personal contact with the health care team. Accordingly, of all available possibilities for face-to-face (delivered by the CC) or distance (technology assisted) interventions, we selected the most suitable mode of delivery for each of our intervention’s BCTs (Table 8).

The resulting draft of the SMILe Care Model Prototype integrates a CC (an APN with specialization in oncology) and the SMILeApp (“prototype” means the whole care model, not only the app component). The BCTs to support medication adherence will be delivered to the patients during two face-to-face visits with the CC, e.g. demonstration how to perform the behavior (e.g. prepare the medication) or habit formation. Depending on the patient’s condition, the care protocols allow for a step-up approach including more intensive tailored interventions. Between face-to-face visits, we planned to support patients with several BCTs delivered by the SMILeApp, e.g., self-monitoring of the behavior (Table 8).

The intervention draft was discussed with alloSCT patients and caregivers to prioritize functionalities. This process will later guide the order of the software and intervention development. Patients agreed unanimously that the SMILeApp should ideally include a current and complete medication plan that could be automatically updated after every change to the medication regimen. Patients considered this as a feasible way to confirm their medication intake after every intake. Most participants also considered it acceptable and helpful to receive a reminder for data entry once a day at a user-defined time, with graphical feedback for entered values.

All findings arising from steps 4 to 8 are presented in Table 8 which also indicates which COM-B and TDF domains are influenced by each intervention function and policy category, as well as the resulting BCTs and modes of delivery.

Additional stage 4: SMILe care model prototype development

The translation of the SMILe intervention into an eHealth solution will enable delivery of important BCTs, e.g., self-monitoring and feedback on target behaviors to patients anywhere (e.g., in their homes or workplaces) at any time [2, 3]. Therefore, after working through the three stages and eight core steps of the BCW, we developed a fourth stage – SMILe Care Model Prototype Development – which includes two steps, both of which are based on agile software development techniques (Fig. 2). Developed to facilitate early integration within the target setting, a highly iterative methodology and the inclusion of end-user feedback in the development process ensures the software’s usefulness to patients, caregivers and clinicians alike [35].

Additional step 9: prepare digitalization

To prepare for digitization, step 9 deals with the formulation of user stories [35]. User stories are a means to capture requirements that should be delivered by a software product. This makes their stories a valuable basis for discussion between software developers and intervention developers [35]. User stories are an opportunity for researchers to provide key information to software developers about important functionalities that are needed in the app. They will not directly appear in the app, only by means of the functionality implemented. The goal of user stories it to provide key information between researchers and software developers and, thus, are not communicated to patients.

While user stories can be structured in numerous ways, the role-feature-reason format is most popular [35]. This begins with a structured sentence (As a … (person, e.g., allo-SCT patient), I want … (action, e.g., keep track of my medication intake), so that … (expected outcome, e.g., I know if I take my medication correctly)) describing a possible software function. Structuring the essential elements in a standard form eases the translation of the corresponding BCTs into the SMILeApp [35]. The concrete realization of a user story is decided in close consultation with the software development team.

Results. To ensure the translation of the most effective BCTs into our eHealth component, we wrote 17 user stories to the software developers in the role-feature-reason format based on the previous findings (Table 9). Following the principles of agile software development, continuous end-user testing and user-centered design, the needs and priorities these stories suggest will help the designers first to develop mock-ups of new modules for early user testing, then to ensure that later versions of the app meet user needs as fully as possible [35]. A separate paper will describe the development of the technology aspects in detail (in preparation).

Table 9 User stories according to the BCT

Additional step 10: prepare CC intervention

Based on the previous steps of the BCW, we wrote a comprehensive intervention protocol for the intervention’s face-to-face CC visits. This protocol describes every face-to-face visit in terms of when, by whom and to whom it should be delivered (patient ± caregiver(s)), as well as which BCTs should be applied in what order and which methods should be used. To ensure standardized, reproducible intervention delivery, descriptions and terminology adhered closely to the BCT taxonomy. The intervention protocol was written in alignment with implementation science lens, meaning, that the intervention has been designed using the previously described implementation science methods (e.g. contextual analysis, stakeholder involvement) in such a way that it can be implemented and used in real world settings in the future.

Results. The complete protocol consisted of 70 pages for all modules with the detailed description of the content to be used for education, training and supervision of the CC. We also wrote a short version with 49 pages for all modules to be used by the CC during the face-to-face visits as a checklist.

We initially planned to implement and test the whole medication adherence module via both, the CC (face-to-face visits) and the SMILeApp in an RCT. However, due to lack of time resources, the current version of the SMILeApp does not include medication adherence support yet. For this reason, in the current RCT, which started at the beginning of 2020 at the University Hospital of Freiburg im Breisgau (FiB), only the face-to-face visits of the medication adherence module are implemented and tested. This downsized version of the care model is called the SMILe-V1 Care Model Prototype–FiB as it is tested at the University Hospital of FiB. Our prepared user stories will be prioritized and translated into the next version of the SMILeApp to be implemented and evaluated in combination with the CC’s face-to-face visits in an RCT at the University Hospital of Basel (USB). This full version of the care model will be called the SMILe-V2 Care Model Prototype–USB.

Discussion

To our knowledge, this paper is the first to provide an example of how to develop a theory-based medication adherence intervention for translation into an eHealth system. Our work, embedded in an implementation science approach, applied the principles recommended by the BCW, and particularly of the COM-B model at its center. This framework’s major strengths are its multilevel perspective and the explicit inclusion of emotional factors, which tend to be less prominent or absent in other behavioral theories [34]. While other authors have used the BCW to develop eHealth facilitated behavior change interventions, none have described their development process (including the translation of an intervention into an eHealth application) in this detail, and many have not included patients or incorporated them only at a late stage of intervention development [37, 38, 67].

Worldwide, health care services and providers are moving in the direction of digitization [68]. The number of newly released health related apps is currently growing by over 200 per day: between 2015 and 2018, the number available in the top app stores almost doubled to nearly 320′000 [24]. Many aim to support patients in medication adherence [22] and some have showed promising results in RCTs in various populations [14, 16, 17]. In the development of an eHealth facilitated intervention, it is crucial to rely on the theory, consider the most recent evidence, integrate information on the context where the intervention or app will be applied and involve all major stakeholders – especially patients – from the earliest stages [24, 28, 69, 70]. And while theory-based interventions are most likely to improve medication adherence [28, 69], extremely few apps are developed following these recommendations.

To exacerbate the problem, producers typically supply limited or no information on the processes either of their apps’ implementation or of their results in real-life settings [16, 29]. This paper addresses these shortcomings using the example of the SMILe integrated care model’s medication adherence intervention module. Unlike many existing eHealth tools, our intervention development incorporated extensive end-user involvement (i.e., of patients, caregivers, clinicians), enhancing its relevance regarding implementation in real-life settings. Our goal was not simple to develop the best possible intervention to provide the basis for digitalization and medication adherence interventional components, but also to maximize the likelihood of its successful implementation and sustainability in daily clinical alloSCT follow-up care.

This paper also presents a first step towards bridging the gap between trial and real-world contexts in the development and implementation of a medication adherence module: it describes how to combine empirical evidence with contextual data [36] as the foundation of an eHealth facilitated intervention. The involvement of all relevant stakeholders was indispensible for this module’s durable implementation into clinical practice [27, 34].

In addition, by developing user stories based on end-user needs and feedback, we combined the central principles of Implementation Science – sensitivity to context, building on an existing evidence base, extensive stakeholder involvement – with the principles of agile software development and user-centered design [35, 70]. With the result – an iterative, incremental, user-centered approach with early connection to the target context – we intend to speed the translation of cutting-edge research findings into routine use not only by clinicians but also by the target (alloSCT patient) population [24, 70]. By involving the various stakeholder groups at each appropriate stage, we ensured that the intervention would fit the needs of end-users and be feasible for sustainable use in clinical practice. This fact alone sets our eHealth tools apart from others, the vast majority of which are typically developed by software designers with little or no input from health care research teams [24].

The BCW is a relatively novel multilevel behavioral framework that both explains and provides a stable framework for intervention development. Building for our work upon the evaluation of correlates/determinants of medication adherence intervention development, not only does it include cognitive patient-level factors, it also explicitly includes emotional factors that are not prominent in currently prevalent health behavior models, e.g., the Integrated Model of Behavioral prediction [34, 71].

By assisting in the selection of appropriate behavior change interventions [34], the BCW provides invaluable guidance for intervention development. And by drawing from diverse information sources it allows the combination of quantitative and qualitative evidence, the results of contextual analysis and stakeholder involvement [34], and the researchers’ clinical expertise. The full range of these sources informed our intervention development.

Integral to the BCW, the BCT taxonomy [72] provides standardized language to label even the smallest units of behavioral change interventions. In addition to enhancing the reporting and communication of complex interventions’ content and enhancing their replicability, this level of standardization facilitates meta-analysis, especially as it is used to detect the best-performing intervention components.

Although the BCW is arguably an excellent foundation for intervention development, we added one stage to it at the end of our development process. To speed and simplify translation of the medication adherence module into an eHealth application, we recommend and describe the formulation of user stories as a bridge between intervention developers and software developers. Congruent with the BCW’s aims, agile software development focuses both on stakeholder involvement and on the adaption of the software according to the needs of its end-users [35].

One notable challenge regarding the use of the BCW is that, even while its developers provide a step-by-step process for intervention development, they tend to provide only brief descriptions of the links between those steps, which can make following their recommendations rather challenging especially for those who use the framework for the first time. Regarding applying COM-B to our specific research question, we faced the issue that there is very few evidence regarding medication adherence in alloSCT. Therefore, we had to expand our searches on similar populations. Additionally, following the steps left us with more potential options (i.e., BCTs) than could possibly be realized from a resource and logistical perspective. Future refinement of the BCW could help to overcome this challenge by providing more guidance on which of the possible BCTs would be most appropriate. In our example we based the selection and reduction to the most important and promising options on the feedback of the stakeholders to promote feasible implementation: We applied the APEASE criteria to each possible BCT and discussed which possible app functionalities would specifically help stem cell transplant patients and their caregivers with decision-making and prioritization. This approach ensured that the intervention would fit the needs and preferences of the end-users.

Application of theory is facilitated by examples. While our example focuses directly on the development of behavioral intervention and eHealth app, it and its underlying theory can also be applied to the adaptation of existing interventions and apps to new contexts. Nevertheless, a strong knowledge of the theory is crucial and needs to be firmly in place before starting the development process. To bring all the needed competencies and perspectives to the table, an interdisciplinary team and stakeholder involvement are essential.

Our international multidisciplinary research team consists of 31 researchers and clinicians, two of whom were members of the BCW development team for the medication adherence module. Several of our researchers also went to London to follow summer courses offered by the group led by Susan Michie, who developed the BCW. In all, development of the SMILe intervention modules required an investment of more than 1 year. Therefore, we advise other researchers that, to prepare adequately for this task, they should combine readings of papers and books with formal training. Another valuable option would be to collaborate with more experienced researchers to peer support this process.

It is still unknown, whether a theory-based developed intervention module such as ours is able to improve medication adherence in alloSCT. In addition, it is unclear whether an analogue or digital intervention, or even the combination of both is the most successful which needs further investigation. Our study’s results will support agile technology development of the SMILe medication adherence module as part of the SMILe integrated care model. The SMILe-V1 Care Model Prototype–FiB (without medication adherence in the SMILeApp) and SMILe-V2 Care Model Prototype–USB (with medication adherence in the SMILeApp) will subsequently be implemented and tested as part of an implementation science study to address these knowledge gaps (https://smile.nursing.unibas.ch/). Once the SMILe Care Model Prototype has proven its effectiveness, it can be sustainably implemented in clinical practice as it was developed under consideration of implementation science aspects (e.g. feasibility and acceptability).

Limitations

Our development of the eHealth component of our care model was subject to certain restrictions. For example, as contextual factors such as data security legislation are very strict in Switzerland and the EU, certain proposed app functions could not be included. Even if qualitative literature and the involved patients had certain priorities regarding the functionality of the SMILeApp, other factors such as the legal framework could force different sequences of the app development process. However, the whole medication adherence module is more than just the eHealth part, why a first testing of the intervention is still possible.

Conclusion

As intended, this paper describes the theory-driven development, based on the BCW, of a medication adherence intervention module as part of the SMILe integrated care model for alloSCT patients, which includes a patient-centered app and introduces one new care team role: a CC. The associated study’s results are currently in use for agile technology development, employing a user-centered design approach, of the innovative SMILe integrated care model’s medication adherence module. The overall care model will be implemented and tested as part of a planned implementation science study. While the methods described are applied to this particular medication adherence intervention, as they follow BCW recommendations, they can be adapted to the development of a wide range of behavior-targeted eHealth applications.

Availability of data and materials

The data of the previously performed contextual analysis included in the development of the intervention were recently published by LL and our research group [36]. As the original data is personal and cannot be completely anonymized, it cannot be made available for open access. Other data analyzed and synthesized are available from the corresponding author upon request.

Notes

  1. 1.

    Although implementation adherence uses the same word, this does not equate with implementation in the context of implementation science.

Abbreviations

AlloSCT:

Allogeneic stem cell transplantation

BCT:

Behavior Change Techniques

BCW:

Behavior Change Wheel

CC:

Care-coordinator

COM-B:

Capability-Opportunity-Motivation-Behavior Model

CReDECI 2:

Criteria for Reporting the Development and Evaluation of Complex Interventions in healthcare: revised guideline

EM:

Electronic monitoring

EMERGE:

ESPACOMP Medication Adherence Reporting Guideline

ESPACOMP:

European Society for Patient Adherence, Compliance, and Persistence

FiB:

Freiburg im Breisgau

GvHD:

Graft-versus-Host Disease

MNA:

Medication non-adherence

RCT:

Randomized controlled trial

TDF:

Theoretical Domains Framework

TIDieR:

Template for Intervention Description and Replication

USB:

University Hospital of Basel

References

  1. 1.

    Maedler-Kron C, Marcus VA, Michel RP. Hematopoietic Stem Cell Transplantation. In: Michel RP, Berry GJ, editors. Pathology of Transplantation: A Practical Diagnostic Approach. Cham: Springer International Publishing; 2016. p. 401–49.

    Google Scholar 

  2. 2.

    Bevans M, El-Jawahri A, Tierney DK, Wiener L, Wood WA, Hoodin F, et al. National Institutes of Health hematopoietic cell transplantation late effects initiative: the patient-centered outcomes working group report. Biol Blood Marrow Transplant. 2017;23(4):538–51.

    PubMed  Google Scholar 

  3. 3.

    Gee PM, Greenwood DA, Paterniti DA, Ward D, Miller LMS. The eHealth enhanced chronic care model: a theory derivation approach. J Med Internet Res. 2015;17(4):e86.

    PubMed  PubMed Central  Google Scholar 

  4. 4.

    De Geest S, Zullig LL, Dunbar-Jacob J, Helmy R, Hughes DA, Wilson IB, et al. ESPACOMP medication adherence reporting guideline (EMERGE). Ann Intern Med. 2018;169:30–5.

    PubMed  Google Scholar 

  5. 5.

    Vrijens B, De Geest S, Hughes DA, Przemyslaw K, Demonceau J, Ruppar T, et al. A new taxonomy for describing and defining adherence to medications. Br J Clin Pharmacol. 2012;73(5):691–705.

    PubMed  PubMed Central  Google Scholar 

  6. 6.

    Smith A, Wisloff F, Samson D. Guidelines on the diagnosis and management of multiple myeloma 2005. Br J Haematol. 2006;132(4):410–51.

    PubMed  Google Scholar 

  7. 7.

    Tomblyn M, Chiller T, Einsele H, Gress R, Sepkowitz K, Storek J, et al. Guidelines for preventing infectious complications among hematopoietic cell transplant recipients: a global perspective. Bone Marrow Transplant. 2009;44(8):453.

    CAS  PubMed  Google Scholar 

  8. 8.

    Tabbara IA, Zimmerman K, Morgan C, Nahleh Z. Allogeneic hematopoietic stem cell transplantation: complications and results. Arch Intern Med. 2002;162(14):1558–66.

    PubMed  Google Scholar 

  9. 9.

    Hilgendorf I, Greinix H, Halter JP, Lawitschka A, Bertz H, Wolff D. Long-term follow-up after allogeneic stem cell transplantation. Dtsch Arztebl Int. 2015;112(4):51.

    PubMed  PubMed Central  Google Scholar 

  10. 10.

    Wingard JR, Majhail NS, Brazauskas R, Wang Z, Sobocinski KA, Jacobsohn D, et al. Long-term survival and late deaths after allogeneic hem-atopoietic cell transplantation. J Clin Oncol. 2011;29:2230–9.

    PubMed  PubMed Central  Google Scholar 

  11. 11.

    Wood WA, Chai X, Weisdorf D, Martin PJ, Cutler C, Inamoto Y, et al. Comorbidity burden in patients with chronic GVHD. Bone Marrow Transplant. 2013;48:1429–36.

    CAS  PubMed  PubMed Central  Google Scholar 

  12. 12.

    Gresch B, Kirsch M, Fierz K, Halter JP, Nair G, Denhaerynck K, et al. Medication nonadherence to immunosuppressants after adult allogeneic haematopoietic stem cell transplantation: a multicentre cross-sectional study. Bone Marrow Transplant. 2017;52(2):304.

    CAS  PubMed  Google Scholar 

  13. 13.

    Kirsch M, Götz A, Halter JP, Schanz U, Stussi G, Dobbels F, et al. Differences in health behaviour between recipients of allogeneic haematopoietic SCT and the general population: a matched control study. Bone Marrow Transplant. 2014;49(9):1223.

    CAS  PubMed  Google Scholar 

  14. 14.

    Flodgren G, Rachas A, Farmer AJ, Inzitari M, Shepperd S. Interactive telemedicine: effects on professional practice and health care outcomes. Cochrane Database Syst Rev. 2015;9:CD002098.

    Google Scholar 

  15. 15.

    Dobbels F, De Bleser L, Berben L, Kristanto P, Dupont L, Nevens F, et al. Efficacy of a medication adherence enhancing intervention in transplantation: the MAESTRO-Tx trial. J Heart Lung Transplant. 2017;36(5):499–508.

    PubMed  Google Scholar 

  16. 16.

    Schmid A, Hils S, Kramer-Zucker A, Bogatyreva L, Hauschke D, De Geest S, et al. Telemedically supported case Management of Living-Donor Renal Transplant Recipients to optimize routine evidence-based aftercare: a single-center randomized controlled trial. Am J Transplant. 2017;17(6):1594–605.

    CAS  PubMed  Google Scholar 

  17. 17.

    Linn AJ, Vervloet M, van Dijk L, Smit EG, Van Weert JC. Effects of eHealth interventions on medication adherence: a systematic review of the literature. J Med Internet Res. 2011;13(4):e103.

    PubMed  PubMed Central  Google Scholar 

  18. 18.

    De Bleser L, Matteson M, Dobbels F, Russell C, De Geest S. Interventions to improve medication-adherence after transplantation: a systematic review. Transpl Int. 2009;22(8):780–97.

    PubMed  Google Scholar 

  19. 19.

    Low JK, Williams A, Manias E, Crawford K. Interventions to improve medication adherence in adult kidney transplant recipients: a systematic review. Nephrol Dial Transplant. 2014;30(5):752–61.

    PubMed  Google Scholar 

  20. 20.

    Welch MA, Kaled ES. Ensuring optimal adherence to BCR-ABL1 tyrosine kinase inhibitor therapy for chronic myeloid leukemia. Commun Oncol. 2013;10(5):138–46.

    Google Scholar 

  21. 21.

    Zullig LL, Deschodt M, Liska J, Bosworth H, De Geest S. Moving from the trial to the real world: improving medication adherence using insights of implementation science. Annu Rev Pharmacol Toxicol. 2019;59:423–45.

    CAS  PubMed  Google Scholar 

  22. 22.

    Ahmed I, Ahmad NS, Ali S, Ali S, George A, Danish HS, et al. Medication adherence apps: review and content analysis. JMIR mHealth and uHealth. 2018;6:e62.

    PubMed  PubMed Central  Google Scholar 

  23. 23.

    Morrissey EC, Corbett TK, Walsh JC, Molloy GJ. Behavior change techniques in apps for medication adherence: a content analysis. Am J Prev Med. 2016;50(5):e143–6.

    PubMed  Google Scholar 

  24. 24.

    Aitken M, Clancy B, Nass D. The growing value of digital health. IQVIA Institute for Human Data Science. 2017:1–76.

  25. 25.

    Webb T, Joseph J, Yardley L, Michie S. Using the internet to promote health behavior change: a systematic review and meta-analysis of the impact of theoretical basis, use of behavior change techniques, and mode of delivery on efficacy. J Med Internet Res. 2010;12:e4.

    PubMed  PubMed Central  Google Scholar 

  26. 26.

    Prendergast MB, Gaston RS. Optimizing medication adherence: an ongoing opportunity to improve outcomes after kidney transplantation. Clin J Am Soc Nephrol. 2010;5(7):1305–11.

    PubMed  PubMed Central  Google Scholar 

  27. 27.

    Damschroder LJ, Aron DC, Keith RE, Kirsh SR, Alexander JA, Lowery JC. Fostering implementation of health services research findings into practice: a consolidated framework for advancing implementation science. Implement Sci. 2009;4(1):50.

    PubMed  PubMed Central  Google Scholar 

  28. 28.

    Horne R, Weinman J, Barber N, Elliott R, Morgan M, Cribb A, et al. Concordance, adherence and compliance in medicine taking. London: NCCSDO; 2005:40–6.

  29. 29.

    Demonceau J, Ruppar T, Kristanto P, Hughes DA, Fargher E, Kardas P, et al. Identification and assessment of adherence-enhancing interventions in studies assessing medication adherence through electronically compiled drug dosing histories: a systematic literature review and meta-analysis. Drugs. 2013;73(6):545–62.

    CAS  PubMed  PubMed Central  Google Scholar 

  30. 30.

    McDonald HP, Garg AX, Haynes RB. Interventions to enhance patient adherence to medication prescriptions: scientific review. Jama. 2002;288(22):2868–79.

    PubMed  Google Scholar 

  31. 31.

    Nieuwlaat R, Wilczynski N, Navarro T, Hobson N, Jeffery R, Keepanasseril A, et al. Interventions for enhancing medication adherence. Cochrane Database Syst Rev. 2014;2014(11):CD000011.

    PubMed Central  Google Scholar 

  32. 32.

    Peterson AM, Takiya L, Finley R. Meta-analysis of trials of interventions to improve medication adherence. Am J Health Syst Pharm. 2003;60(7):657–65.

    PubMed  Google Scholar 

  33. 33.

    West R. Theory of addiction. Oxford: Blackwells; 2006.

    Google Scholar 

  34. 34.

    Michie S, Van Stralen MM, West R. The behaviour change wheel: a new method for characterising and designing behaviour change interventions. Implement Sci. 2011;6(1):42.

    PubMed  PubMed Central  Google Scholar 

  35. 35.

    Abrahamsson P, Salo O, Ronkainen J, Warsta J. Agile software development methods: Review and analysis. Espoo: VTT publication 478; 2017.

    Google Scholar 

  36. 36.

    Leppla L, Mielke J, Kunze M, Mauthner O, Teynor A, Valenta S, et al. Clinicians and patients perspectives on follow-up care and eHealth support after allogeneic hematopoietic stem cell transplantation: a mixed-methods contextual analysis as part of the SMILe study. Eur J Oncol Nurs. 2020;45:101723.

    PubMed  Google Scholar 

  37. 37.

    Murphy M, McCloughen A, Curtis K. Using theories of behaviour change to transition multidisciplinary trauma team training from the training environment to clinical practice. Implement Sci. 2019;14:43.

    PubMed  PubMed Central  Google Scholar 

  38. 38.

    Jackson C, Eliasson L, Barber N, Weinman J. Applying COM-B to medication adherence: a suggested framework for research and interventions. European Health Psychologist. 2014;16(1):7–17.

    Google Scholar 

  39. 39.

    Bixby DL. Managing inadequate responses to frontline treatment of chronic myeloid leukemia: a case-based review. Cancer Treat Rev. 2012;39(3):241–51.

    PubMed  Google Scholar 

  40. 40.

    Talati C, Ontiveros EP, Griffiths EA, Wang ES, Wetzler M. How we will treat chronic myeloid leukemia in 2016. Blood Rev. 2015;29(2):137–42.

    PubMed  Google Scholar 

  41. 41.

    Lehrer J, Brissot E, Ruggeri A, Dulery R, Vekhoff A, Battipaglia G, et al. Medication adherence among allogeneic hematopoietic stem cell transplant recipients: a pilot single-center study. Bone Marrow Transplant. 2018;53(2):231–3.

    CAS  PubMed  Google Scholar 

  42. 42.

    Bertz H, Finke J. Transplant manual: Hämatopoetische Zelltransplantation (HZT). Freiburg: University Hospital of Freiburg; 2016.

  43. 43.

    Hoodin F. Psychological and behavioral correlates of medical adherence among adult bone marrow transplant recipients. Detroit: Wayne State University; 1995.

    Google Scholar 

  44. 44.

    Montesinos P, Rodríguez-Veiga R, Boluda B, Martínez-Cuadrón D, Cano I, Lancharro A, et al. Incidence and risk factors of post-engraftment invasive fungal disease in adult allogeneic hematopoietic stem cell transplant recipients receiving oral azoles prophylaxis. Bone Marrow Transplant. 2015;50:1465–72.

    CAS  PubMed  Google Scholar 

  45. 45.

    Schäfer-Keller P, Steiger J, Bock A, Denhaerynck K, De Geest S. Diagnostic accuracy of measurement methods to assess nonadherence to immunosuppressive drugs in kidney transplant recipients. Am J Transplant. 2008;8:616–26.

    PubMed  Google Scholar 

  46. 46.

    De Geest S, Abraham I, Moons P, Vandeputte M, Van Cleemput J, Evers G, et al. Late acute rejection and subclinical noncompliance with cyclosporine therapy in heart transplant recipients. J Heart Lung Transplant. 1998;17:854–63.

    PubMed  Google Scholar 

  47. 47.

    Pinsky BW, Takemoto SK, Lentine KL, Burroughs TE, Schnitzler MA, Salvalaggio PR. Transplant outcomes and economic costs associated with patient noncompliance to immunosuppression. Am J Transplant. 2009;9:2597–606.

    CAS  PubMed  Google Scholar 

  48. 48.

    Muduma G, Shupo FC, Dam S, Hawken NA, Aballea S, Odeyemi I, et al. Patient survey to identify reasons for non-adherence and elicitation of quality of life concepts associated with immunosuppressant therapy in kidney transplant recipients. Patient Prefer Adherence. 2016;10:27.

    PubMed  PubMed Central  Google Scholar 

  49. 49.

    Tong A, Howell M, Wong G, Webster AC, Howard K, Craig JC. The perspectives of kidney transplant recipients on medicine taking: a systematic review of qualitative studies. Nephrol Dial Transplant. 2010;26(1):344–54.

    PubMed  Google Scholar 

  50. 50.

    Williams A, Low JK, Manias E, Crawford K. The transplant team’s support of kidney transplant recipients to take their prescribed medications: a collective responsibility. J Clin Nurs. 2016;25(15–16):2251–61.

    PubMed  Google Scholar 

  51. 51.

    Rebafka A. Medication adherence after renal transplantation—a review of the literature. J Ren Care. 2016;42(4):239–56.

    PubMed  Google Scholar 

  52. 52.

    Jamieson NJ, Hanson CS, Josephson MA, Gordon EJ, Craig JC, Halleck F, et al. Motivations, challenges, and attitudes to self-management in kidney transplant recipients: a systematic review of qualitative studies. Am J Kidney Dis. 2016;67(3):461–78.

    PubMed  Google Scholar 

  53. 53.

    O'Grady JG, Asderakis A, Bradley R, Burnapp L, McPake DM, Perrin M, et al. Multidisciplinary insights into optimizing adherence after solid organ transplantation. Transplantation. 2010;89(5):627–32.

    PubMed  Google Scholar 

  54. 54.

    Tong A, Morton RL, Webster AC. How qualitative research informs clinical and policy decision making in transplantation: a review. Transplantation. 2016;100(9):1997–2005.

    PubMed  Google Scholar 

  55. 55.

    Pinter J, Hanson CS, Craig JC, Chapman JR, Budde K, Halleck F, et al. ‘I feel stronger and younger all the time’—perspectives of elderly kidney transplant recipients: thematic synthesis of qualitative research. Nephrol Dial Transplant. 2016;31(9):1531–40.

    PubMed  Google Scholar 

  56. 56.

    Low JK, Crawford K, Manias E, Williams A. Stressors and coping resources of Australian kidney transplant recipients related to medication taking: a qualitative study. J Clin Nurs. 2017;26(11–12):1495–507.

    PubMed  Google Scholar 

  57. 57.

    Israni A, Dean C, Kasel B, Berndt L, Wildebush W, Wang CJ. Why do patients forget to take immunosuppression medications and miss appointments: can a mobile phone app help? JMIR Public Health Surveill. 2016;2(1):e15.

    PubMed  PubMed Central  Google Scholar 

  58. 58.

    Low JK, Crawford K, Manias E, Williams A. A compilation of consumers’ stories: the development of a video to enhance medication adherence in newly transplanted kidney recipients. J Adv Nurs. 2016;72(4):813–24.

    PubMed  Google Scholar 

  59. 59.

    Janelle C, O’Connor K, Dupuis G. Evaluating illness representations in heart transplant patients. J Health Psychol. 2016;21(9):1850–9.

    PubMed  Google Scholar 

  60. 60.

    Williams A, Crawford K, Manias E, Ellis C, Mullins K, Howe K, et al. Examining the preparation and ongoing support of adults to take their medications as prescribed in kidney transplantation. J Eval Clin Pract. 2015;21(2):180–6.

    PubMed  Google Scholar 

  61. 61.

    Ivarsson B, Ekmehag B, Sjöberg T. Patients’ experiences of information and support during the first six months after heart or lung transplantation. Eur J Cardiovasc Nurs. 2013;12(4):400–6.

    PubMed  Google Scholar 

  62. 62.

    Denhaerynck K, Berben L, Dobbels F, Russell CL, Crespo-Leiro MG, Poncelet AJ, et al. Multilevel factors are associated with immunosuppressant nonadherence in heart transplant recipients: the international BRIGHT study. Am J Transplant. 2018;18(6):1447–60.

    PubMed  PubMed Central  Google Scholar 

  63. 63.

    Boaz A, Morgan M. Working to establish ‘normality’post-transplant: a qualitative study of kidney transplant patients. Chronic illn. 2014;10(4):247–58.

    PubMed  Google Scholar 

  64. 64.

    De Geest S, Burkhalter H, Bogert L, Berben L, Glass TR, Denhaerynck K, et al. Describing the evolution of medication nonadherence from pretransplant until 3 years post-transplant nonadherence to immunosuppressives: the Swiss transplant cohort study. Transpl Int. 2014;27:657–66.

    PubMed  Google Scholar 

  65. 65.

    Cooke L, Chung C, Grant M. Psychosocial Care for Adolescent and Young Adult Hematopoietic Cell Transplant Patients. J Psychosoc Oncol. 2011;29(4):394–414.

    PubMed  PubMed Central  Google Scholar 

  66. 66.

    Michie S, Atkins L, West R. The behaviour change wheel : a guide to designing interventions. Great Britain: Silverback Publishing; 2014.

    Google Scholar 

  67. 67.

    Brito Félix I, Pereira Guerreiro M, Cavaco A, Paula Cláudio A, Mendes A, Balsa J, et al. Development of a complex intervention to improve adherence to antidiabetic medication in older people using an anthropomorphic virtual assistant software. Front Pharmacol. 2019;10:680.

    Google Scholar 

  68. 68.

    Lucas B, Brammen D, Schirrmeister W, Aleyt J, Kulla M, Roehrig R, et al. Requirements for a sustainable standardization and digitalization in clinical emergency and acute medicine. Unfallchirurg. 2019;122:243–6.

    PubMed  Google Scholar 

  69. 69.

    Conn VS, Enriquez M, Ruppar TM, Chan KC. Meta-analyses of theory use in medication adherence intervention research. Am J Health Behav. 2016;40:155–71.

    PubMed  PubMed Central  Google Scholar 

  70. 70.

    Lyon AR, Bruns EJ. User-centered redesign of evidence-based psycho-social interventions to enhance implementation—hospitable soil or better seeds? JAMA Psychiatry. 2019;76(1):–4.

  71. 71.

    Yzer M. The integrative model of behavioral prediction as a tool for designing health messages. In: Hyuniyi C, editor. Health communication message design: Theory and practice; 2012. p. 21–40.

    Google Scholar 

  72. 72.

    Michie S, Richardson M, Johnston M, Abraham C, Francis J, Hardeman W, et al. The behavior change technique taxonomy (v1) of 93 hierarchically clustered techniques: building an international consensus for the reporting of beh. Ann Behav Med. 2013;46:81–95.

    PubMed  PubMed Central  Google Scholar 

  73. 73.

    Möhler R, Köpke S, Meyer G. Criteria for reporting the development and evaluation of complex interventions in healthcare: revised guideline (CReDECI 2). Trials. 2015;16(1):204.

    PubMed  PubMed Central  Google Scholar 

  74. 74.

    Hoffmann TC, Glasziou PP, Boutron I, Milne R, Perera R, Moher D, et al. Better reporting of interventions: template for intervention description and replication (TIDieR) checklist and guide. Bmj. 2014;348:g1687.

    PubMed  Google Scholar 

Download references

Acknowledgements

We acknowledge the SMILe study team, especially Sonja Beckmann, Juliane Mielke and Anja Schmid of the Institute of Nursing Science, University of Basel, Switzerland; Nathalie Duerinckx of the Academic Centre for Nursing and Midwifery, University of Leuven, Belgium, as well as Phillip Heidegger, Margarita Fürmann, Daniela Neupert, Dennis Rockstein, Viktor Werlitz, Michael Fürmann, Tobias Schulz, Marina Lemcke and Vanessa Schumacher of the University of Applied Sciences, Augsburg, Germany. On the clinical side, we acknowledge Robert Zeiser, Monika Engelhardt, Monika Hasemann and Klaus Kaier from the University Medical Center Freiburg, Germany, as well as Sabine Gerull, Jakob Passweg, Anja Ulrich, Florian Grossmann, Dora Bolliger, Sigrun Reitwiessner, Sabine Degen, Sandra Schönfeld, Yuliya Senft, and Birgit Maier from the University Hospital of Basel, Switzerland. We also thank Chris Shultis for the the editing of this paper.

Reporting guideline checklist

As we were developing an intervention we followed the Criteria for Reporting the Development and Evaluation of Complex Interventions in healthcare: revised guideline (CReDECI 2) [73], the Template for Intervention Description and Replication (TIDieR) where applicable and to the extent appropriate to the objectives of this paper [74]; and because the intervention concerns medication adherence we also followed the ESPACOMP Medication Adherence Reporting Guideline (EMERGE) [4] where applicable.

Funding

The overall SMILe research project is funded by the B. Braun Foundation, the Jose Carreras Foundation, the Federal Ministry of Education and Research (Germany) and the Swiss Cancer League. The funders had no role in study design, data collection, analysis, interpretation and writing the report.

Author information

Affiliations

Authors

Consortia

Contributions

The study was designed by SDG (corresponding author: sabina.degeest@unibas.ch) and JR. Intervention development, following the steps of the BCW, including data collection and interpretation, was performed by SDG and JR with involvement of LL, AT, SV and FD. The manuscript was written by JR with regular revision and feedback from SDG and LL. Before publication the final revision was reviewed and approved by all co-authors.

Corresponding author

Correspondence to Sabina De Geest.

Ethics declarations

Ethics approval and consent to participate

The prototype development of the SMILe-V1 Care Model Prototype–FiB including multi-methods contextual analysis and user tests was approved by the Ethics Committee of Freiburg, Germany (EK67/17). All participants signed informed consent.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Ribaut, J., Leppla, L., Teynor, A. et al. Theory-driven development of a medication adherence intervention delivered by eHealth and transplant team in allogeneic stem cell transplantation: the SMILe implementation science project. BMC Health Serv Res 20, 827 (2020). https://doi.org/10.1186/s12913-020-05636-1

Download citation

Keywords

  • Allogeneic hematopoietic stem cell transplantation
  • Medication adherence
  • Intervention development
  • Behavior change wheel
  • Theory-driven
  • eHealth intervention
  • Implementation science