Key Elements and Theoretical Foundations for the Design and Delivery of Text Messages to Boost Medication Adherence in Patients With Diabetes, Hypertension, and Hyperlipidemia: Scoping Review

Introduction

Chronic diseases are a major global health burden [1], profoundly affecting quality of life, health care systems, and economic stability [2]. Hypertension, hyperlipidemia, and diabetes, commonly grouped under cardiometabolic syndrome, are considered leading chronic diseases and are closely linked to a variety of serious health complications. Managing these conditions through lifestyle changes and medication use is essential in reducing their long-term impact on health [2]. The prevalence of chronic diseases has escalated over the past decade, with noncommunicable diseases accounting for 67% of global deaths in 2010, rising to 74% by 2019 [1]. This trend is expected to impose an economic loss of approximately US $47 trillion worldwide by 2030 [3]. Effective management of chronic conditions relies heavily on medication adherence [4], defined as the extent to which individuals follow prescribed medication regimens as suggested by health care professionals [5,6]. High adherence rates are associated with improved clinical outcomes, reduced complications, and lower health care costs [7,8]. However, medication adherence remains suboptimal worldwide, with an estimated global adherence rate of only 50% [9]. This suboptimal adherence highlights the urgent need for effective and scalable strategies to improve adherence, particularly among patients with chronic diseases [10].

A variety of interventions have been implemented to improve medication adherence, including patient education, medication regimen management, pharmacist-led programs, cognitive behavioral therapies, medication reminders, and incentive-based measures [11]. While these approaches have demonstrated efficacy, they are often complex, resource-intensive, and difficult to scale [12-14]. For instance, pharmacist-driven initiatives require significant personnel and financial investments, thus limiting their feasibility in resource-constrained settings [11]. In addition, many existing adherence interventions demand frequent in-person interactions, which can be burdensome for patients and health care providers alike, especially in remote areas or those with mobility challenges [15]. A critical limitation of many adherence interventions is their lack of a solid theoretical foundation, leading to inconsistencies in their effectiveness and applicability across diverse populations [16]. Consequently, there is a growing need for cost-effective, scalable, and theory-driven interventions that can be delivered to reach a broader population [12,14,17].

The rapid proliferation of mobile phone use presents an opportunity to address these challenges [18]. With the widespread adoption of mobile technologies, mobile phone–based interventions present a promising and accessible solution for improving medication adherence [14,19]. As of 2024, approximately 4.3 billion people globally own smartphones [20], yet even those with limited technological literacy or lower socioeconomic status frequently use basic mobile phones [21]. SMS text messaging offers a widely accessible and cost-effective approach that does not require internet connectivity or smartphone technology [22]. Owing to its universal reach, simplicity, and potential for high patient engagement, SMS text messaging–based interventions are particularly well-suited for improving medication adherence, particularly in resource-limited settings [23,24]. The minimal technical demands of SMS text messaging make it particularly appropriate for older adults or individuals with limited digital proficiency [25]. Moreover, SMS text messaging interventions have shown considerable potential in enhancing medication adherence and clinical outcomes while remaining cost-effective. These SMS text messaging–based interventions deliver substantial health benefits while being affordable for both patients and health care systems [26].

Research consistently supports the effectiveness of SMS text messaging–based interventions in enhancing medication adherence, with randomized controlled trials reporting significant improvements in adherence rates [14,22,23,26-28]. Thakkar et al [14] reviewed 16 trials and found that SMS text messaging nearly doubled adherence rates among adults with chronic diseases. Similarly, the study conducted by Burn et al [26] demonstrated that SMS text messaging could reduce cardiovascular events and generate substantial cost savings. Despite these encouraging findings, many studies often overlook the specific content and theoretical grounding of the text messages, which may limit their long-term effectiveness [14]. Furthermore, a significant number of SMS text messaging–based programs either lack a strong theoretical foundation or apply behavioral theories ineffectively [29]. This raises questions about whether theory-driven interventions yield superior outcomes compared with those without a structured theoretical approach [29,30]. In addition, research on SMS text messaging has provided limited insights into their effectiveness across diverse populations and varying patient conditions [27]. The comparative benefits of tailored versus standardized messages remain uncertain, and few reviews have systematically analyzed the personalization strategies used in these interventions [31]. These gaps underscore the need for a comprehensive evaluation of the theoretical models and message-tailoring strategies used in text-based adherence interventions [32].

Given the increasing burden of diabetes, hypertension, and hyperlipidemia and the growing reliance on mobile health interventions, this study aimed to systematically review the literature on SMS text messaging interventions for medication adherence. Specifically, this review synthesized the theoretical frameworks that informed these interventions and investigated the methods used to tailor messages to individual patient needs. By addressing these gaps, this study aimed to contribute to the development of more effective and evidence-based strategies for enhancing medication adherence through mobile health interventions.

Methods

Study Design

We used the SPIDER (sample, phenomenon of interest, design, evaluation, research type) framework to develop our search strategy [33], guided by the research question, What are the key components of text messages in medication adherence intervention using digital devices? This study involved formulating a search strategy, applying predefined inclusion criteria for article screening, comparing and selecting relevant studies, and extracting key data. To ensure comprehensive and transparent reporting, this scoping review was conducted in accordance with the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) guidelines, including the development and implementation of the search strategy [34].

Search Strategy

We conducted a literature search in 2 electronic databases, PubMed and Scopus, to identify studies that contained text messages delivered via digital tools to enhance medication adherence. The SPIDER framework was applied to structure the research question and guide the search strategy (Multimedia Appendix 1). The key search terms included “diabetes,” “hypertension,” “hyperlipidemia,” “message,” “text,” “text message,” “app,” “application,” “digital,” “device,” “mobile,” “medical adherence,” and “medication adherence.” These terms were combined using the logical operators “OR” and “AND” and were restricted to searches within titles and abstracts only (Multimedia Appendix 2). We reviewed articles published in English from inception to October 2024, excluding those with “review” in the title.

Selection Process

The first 2 authors (YMY and TW) independently screened titles and abstracts from PubMed and Scopus databases and used EndNote to identify and remove duplicates. Following the screening process, they selected articles based on the predefined inclusion and exclusion criteria (Textbox 1). Studies were excluded if they lacked a clear rationale for message design, failed to describe the message content, or did not align text messages with the corresponding theoretical frameworks. Any disagreements among the reviewers were resolved through discussion, with a third reviewer (YMH) consulted when necessary to reach a consensus. In addition, gray literature was hand-searched using Google Scholar to identify further relevant studies during the full-text review process.

Data Extraction and Synthesis

Two reviewers (YMY and TW) extracted data from the full-text articles, including study design, author, year of publication, disease focus, behavioral theory, and the constructs or domains of messages. In addition, detailed information on text messages was collected, such as sending frequency and the content of tailored messages. These text message characteristics across the studies were then evaluated and compared in our review. Frequency tabulation was used to numerically summarize key variables, including disease focus, study design, behavioral theories applied, message frequency, and features of tailored messages.

Results

Overview

Initially, we identified 216 articles from PubMed and 131 from Scopus; we removed 86 (24.8%) duplicates, leaving 261 (75.2%) studies. After screening titles and abstracts, 49 (14.1%) articles were excluded because of irrelevance. The remaining 212 (61.1%) articles underwent full-text review for eligibility assessment. Of these, 160 (46.1%) studies were excluded for various reasons, including 31 (8.9%) studies not being primary research, 99 (28.5%) lacking detailed information on text messages, 19 (5.5%) not using text messages to improve medication adherence, 5 (1.4%) not focusing on chronic disease, and 6 (1.7%) being inaccessible. Most of the studies excluded because of lack of detail on text messages either failed to explain how the messages were designed, did not connect the messages to any theoretical framework, or referenced theory without showing how it was used to shape the messages. Without this information, we were unable to explore how the messages were tailored or how they related to the underlying theoretical frameworks. Ultimately, 52 (15%) studies [22,35-85] met the inclusion criteria and were included in the review (Figure 1).

‎

Characteristics of the Included Studies

Among the 52 eligible studies, most focused on a single disease (45/52, 87%). Diabetes (26/52, 50%) was the most commonly studied condition [22,35-37,39,40,44,51,52,56,58,59,63, 65-67,70-74,76,79,80,84,85], including type 1 diabetes (1/52, 2%) [35], type 2 diabetes (22/52, 42%) [22,36,37,40,44,51,52,56,59,63,66,67,70-74,76,79,80,84,85], unspecified diabetes (2/52, 4%) [38,64], and both type 1 and type 2 diabetes (1/52, 2%) [58]. Other conditions investigated included hypertension (16/52, 31%) [38,41-43,45,47,48,55,60-62,68,69,75,81,82], coronary heart disease (1/52, 2%) [57], cardiovascular disease (1/52, 2%) [53], overweight (1/52, 2%) [64], and recurrent stroke prevention (1/52, 2%) [83]. In addition, some studies included participants with multiple or different conditions, such as coronary heart disease and type 2 diabetes (1/52, 2%) [54], diabetes and coronary heart disease (1/52, 2%) [46], hypertension and bipolar disorder (1/52, 2%) [78], type 2 diabetes and hypertension (1/52, 2%) [50], and any chronic disease (2/52, 4%) [49,77] (Table 1).

Regarding study design, 39 (75%) studies used a randomized controlled trial approach [35,37,38,40-42,44-49,52-54,56,57,60,62-66,68-80,83-85], 7 (14%) were cross-sectional studies [22,43,51,55,67,81,82], and 5 (10%) focused on message development and feasibility [36,39,58,59,61] (Table 1). Furthermore, 33 (64%) studies incorporated behavioral theories or techniques [22,39-42,46-48,51-61,64,66,67,70,72,74-76,78,80-82,84,85], while an equal number (33/52, 64%) incorporated tailored message design [22,35,36,38,39,41, 42,45,46,49,50,52,54-58,61,64-67,69,70,72,74,75,78,79,81,83-85] (Table 1).

Features of the Behavioral Theory or Technique

Over half of the included studies incorporated behavioral theories or techniques (33/52, 64%) [22,39-42,46-48,51-61,64,66,67,70,72,74-76,78,80-82,84,85]. Among these, 19 distinct behavioral theories or techniques were used during message development (Table 2). The most frequently applied framework was the behavior change technique taxonomy (16/52, 31%) [42,48,51-57,59,60,72,74-76,84], followed by the capability, motivation, opportunity, behavior (COM-B) model (6/52, 12%) [56,60,67,70,72,75], the transtheoretical model of behavioral change (4/52, 8%) [40,53,66,80], and the social cognitive theory (4/52, 8%) [47,61,80,82]. Among the 33 studies incorporating theoretical frameworks or techniques in message design, 19 (37%) used a single theory or technique [22,42,46,47,51,52,54,55,57,58,61,66,70,74,76,78,81,84,85], while 14 (27%) applied multiple theories or techniques [39-41,48,53,56,59,60,64,67,72,75,80,82]. Consequently, the total number of theories or techniques used exceeded the number of studies.

Domain or Constructs

The design of the SMS text messages incorporated between 1 and 34 domains or constructs, although 3 (6%) studies did not specify the exact number used [40,47,63] (Table 2). Across all 52 studies, 8 distinct domain categories were identified: behavioral factors, knowledge and education, social and psychological support, lifestyle and health behaviors, clinical and management goals, barriers to access, cultural and belief systems, and theoretical constructs. Among these, behavioral factors were the most frequently addressed in interventions, followed by lifestyle and health behaviors, as well as knowledge and education. Behavioral factors focused on reducing the burden of medication-taking and encouraging behavior change by incorporating adherence reminders, providing cues or prompts, and fostering positive habits to support consistent medication use. Furthermore, lifestyle-related factors, including diet and nutrition as well as physical activity and exercise, played a significant role in intervention design.

Tailored Messages

In addition to generic messages sent uniformly to all recipients, 33 (64%) studies incorporated tailored messages customized to individual characteristics or needs [22,35,36,38,39,41,42,45,46,49,50,52,54-58,61,64-67,69, 70,72,74,75,78,79,81,83-85]. As summarized in Table 3, these studies used various methods and strategies to ensure that message content aligned with users’ specific conditions.

Among the studies using tailored text messages, the most common approach involved customizing message content based on users’ personal information. This strategy was applied in 21 (40%) studies [22,38,39,41,42,45,46,52,54-57,64,65,70,72,75,78,79,84,85], incorporating details such as users’ names, dates of birth, medication use, dosage and frequency, scheduled follow-up or refill appointments, smoking status, and preferred language. In addition, 14 (27%) studies allowed users to customize message timing and frequency [35,36,39,41,45,55,65,66,70,72,74,79,81,83], providing greater flexibility to meet individual needs. Moreover, 6 (12%) studies tailored messages based on assessments before the intervention [22,35,36,39,50,79]. Of these, 3 (6%) evaluated users’ top 3 or 4 barriers to medication adherence and provided targeted messages to address these challenges [22,35,36]. Two (4%) studies used baseline questionnaires to identify patients’ personal health goals and designed messages accordingly [50,79]. One (2%) study applied the health belief model and self-determination theory to categorize patients into different levels and tailored messages based on these classifications [39].

Furthermore, 6 (12%) studies tailored messages based on ongoing assessments conducted during the intervention [22,55,58,61,67,69]. Among these, 4 (8%) adjusted messages based on users’ daily survey or questionnaire responses [22,55,67,69], while 2 (4%) modified messages according to users’ clinical data, such as self-monitored blood pressure or blood glucose levels [58,61]. In addition, 3 (6%) studies personalized messages in response to users’ feedback [49,67,74], considering factors such as medication refill status, response rates to previous messages, and user preferences for content.

To further illustrate tailored messaging approaches, several representative studies were highlighted. These studies were carefully chosen to showcase a variety of tailoring strategies used across the broader group of included studies, such as providing clear explanations for message design, methods for crafting tailored messages, and incorporating patient feedback into the design and delivery of these messages. Mulvaney et al [35] developed messages addressing adherence barriers identified through the Barriers to Diabetes Adherence instrument. Participants received 75% of their messages tailored to their top 3 adherence barriers, with the remaining 25% randomly selected from a broader message pool. In addition, users could schedule messages at specific times in 15-minute intervals and set delivery frequency to daily, weekly, or weekends only. Gatwood et al [39] designed messages using constructs from the self-determination theory and health belief model. A Likert-type survey categorized users into high, medium, or low levels for each theoretical construct, with participants initially receiving 2 messages tailored to their level, followed by a third message from a higher level (eg, low-low-medium). The REACH program by Nelson et al [22] delivered daily interactive messages assessing medication adherence. If users responded “no,” a follow-up message prompted them to reflect on their reasons for nonadherence. The study conducted by Prayaga et al [49] used mPulse Mobile to deliver interactive messages focused on medication refill adherence. Users received reminders near refill dates and were asked to respond. If they did not reply within 2 hours or provided an irrelevant response, they received an additional reminder. Adu et al [58] developed My Care Hub, which provided semi-individualized feedback based on users’ logged blood glucose levels. If readings were within the clinically recommended range, messages encouraged continued self-management. If levels were outside the normal range, messages offered problem-solving suggestions for managing high or low blood glucose. Finally, Bartlett et al [74] implemented a preference-based tailoring approach. After receiving a message, users could respond with “like” or “dislike.” For medication adherence messages, liking a message doubled the likelihood of receiving future messages based on the same behavior change technique, while disliking a message halved the probability of receiving similar messages in the future.

Discussion

Main Findings

This review identified 52 SMS text messaging interventions designed to enhance medication adherence in patients with diabetes, hypertension, and hyperlipidemia. The findings underscore the varied strategies used to tailor these interventions to improve patient engagement and adherence. Consistent with previous research, personalized text messages that incorporate theoretical frameworks and adapt to real-world contexts show promise in supporting medication adherence [86].

The review categorized text messages into 2 main approaches: content-driven and theory-based designs. Content-driven messages focus on delivering specific information to participants, typically addressing goals such as medication adherence, healthy eating, and regular exercise. In contrast, theory-based messages are structured around behavioral theories, shaping their content based on key theoretical constructs. While some studies adopted only one of these approaches, others combined both strategies. For example, Zamanillo-Campos et al [85] demonstrated that their intervention integrated both targeted health objectives and behavior change techniques, effectively bridging theoretical foundations with practical applications in behavioral interventions. By merging the structured guidance of behavioral theories with the clarity of direct messaging, their study illustrates a comprehensive approach to intervention design.

Behavioral Theory or Technique Use in Interventions

The behavior change technique taxonomy is a comprehensive and standardized framework designed to identify, classify, and describe the active components of behavior change interventions [87]. The taxonomy, consisting of 93 distinct techniques, provides structured approaches for designing and implementing behavioral interventions. In this review, behavior change techniques emerged as the most frequently applied behavioral framework, either used independently or in combination with other models. Behavioral theories and techniques often serve as the foundation for interventions, guiding the development of text message content. For instance, Bartlett et al [59] structured and refined messages based on behavior change techniques through workshops, focus groups, acceptability surveys, and fidelity assessments. Their library of brief messages demonstrated both the applicability and acceptance of interventions based on behavior change techniques.

The COM-B model posits that behavior change requires 3 essential components: capability, opportunity, and motivation [88]. When integrated with the behavior change wheel, COM-B helps identify intervention functions and policy strategies to address behavioral barriers and enablers [88]. Schoenthaler et al [67] applied the COM-B model to text message design, showing that multiple constructs of capability, opportunity, and motivation could be incorporated into message components simultaneously. In addition, they linked their messaging system to patient-reported outcomes, allowing real-time feedback on how patients felt upon receiving messages and informing the development of future behavioral messages.

The transtheoretical model of behavior change conceptualizes behavior change as a dynamic, nonlinear process consisting of 5 stages: precontemplation, contemplation, preparation, action, and maintenance [89]. Sadanshiv et al [66] demonstrated how text messages aligned with these stages, delivering reminders on a predetermined schedule that corresponded to individuals’ progression through the behavior change process.

The social cognitive theory provides a framework for understanding and influencing behaviors through the interplay of personal, environmental, and behavioral factors [90]. A key concept within the social cognitive theory is self-efficacy, which refers to individuals’ belief in their ability to perform behaviors and achieve desired outcomes. Asgary et al [82] structured their messaging around 3 key priorities: knowledge, motivation, and maintenance. When aligned with the social cognitive theory principles, messages addressing disease management, diet, and medication adherence were found to enhance patients’ self-efficacy and confidence, ultimately promoting positive behavioral changes.

Previous medication adherence interventions that used a clearly defined theory tended to achieve better results than those without a theoretical basis [91,92]. Using a single theory to guide message design can simplify the development, analysis, and interpretation of interventions, particularly when the target behavior is clearly defined and the population and setting are specific. This approach allows for a focused exploration of key factors such as beliefs, motivation, or intention. However, it may overlook broader influences such as environmental or social determinants [59]. In contrast, a multitheory approach offers a more comprehensive understanding of the diverse factors influencing medication adherence and provides greater flexibility for designing multicomponent interventions. While this can enhance the richness of the intervention, it also introduces added complexity and requires careful integration to avoid redundancy or conflicting constructs [14,93]. Each behavioral theory contributes unique and validated constructs that can support the development of targeted messages, but overreliance on theory alone may oversimplify the nuanced realities of patient behavior [94]. Human motivation, social context, and environmental influences often do not conform neatly to theoretical models. Therefore, integrating patient feedback alongside theoretical frameworks can help ensure that message design reflects real-world experiences and communication needs [95]. Researchers must strike a balance between theoretical rigor and practical relevance, tailoring their approach to the research goals and intervention context. In exploratory phases, multiple theories may be useful for identifying a broad range of adherence-related factors, while in the development phase, a well-matched single theory or integrated model can guide strategy. During validation, empirical testing, such as structural equation modeling, can help determine which constructs are most predictive in practice. Ultimately, effective messaging depends not only on theoretical soundness but also on contextual relevance and feasibility.

Tailored or Personalized Messages in Interventions

Growing evidence suggests that, under certain conditions, tailored interventions are more effective in promoting behavior change [96]. Tailored text messages, in particular, have proven to be a valuable tool for improving medication adherence, as they seamlessly integrate into daily routines and can be adapted in real time to meet individual needs and health conditions [97]. Among various tailoring strategies, customization based on individual information is one of the simplest and most widely used approaches. Beyond incorporating basic personal details, such as a patient’s name, birthday, or preferred language, messages can also be personalized based on clinical characteristics, including smoking status or comorbidities, ensuring that the content remains relevant and meaningful to the recipient [75,84]. This level of personalization not only enhances patient engagement but also helps reduce user fatigue from receiving messages that do not align with their personal circumstances [98].

Most users find tailored text message timing and frequency highly beneficial for improving medication adherence. There is a strong demand for the ability to modify these features to better suit individual needs [41]. Allowing patients to adjust reminder messages in accordance with their medication regimen helps enhance adherence while mitigating the potential negative effects of excessive messaging [45,99]. Unlike other tailoring approaches where adjustments are made by the message provider, user-controlled customization of timing and frequency empowers individuals to tailor their messaging experience independently [41]. This emphasis on autonomy in tailored text messaging may significantly enhance engagement with health interventions [100].

Patient-reported assessments offer an effective means of tailoring text messages to improve both patient understanding of medication adherence and their acceptance of the messages themselves [36]. Tailoring can occur at different stages of an intervention. For example, assessments conducted before the intervention can help identify the most relevant and beneficial messages using theory-based tools to uncover key barriers to adherence or by evaluating lifestyle habits and disease status to establish personalized disease management goals [35,36,79]. Meanwhile, assessments conducted during the intervention can track patients’ actual medication use and disease management progress, enabling real-time adjustments—whether to reinforce adherence or to provide reminders for improvement [22,58]. Participants have responded positively to these tailored messages [36]. However, a relatively small proportion of the text messages reviewed in this study incorporated patient-reported assessments for tailoring. This is likely because of the significant resources required, including personnel for assessments and the development of tagging systems to link assessment results with personalized message content [35,36,67].

User Feedback in Text Messaging Design

Most studies relied on patients’ baseline characteristics at the time of enrollment, while relatively few incorporated changes in patients’ conditions or feedback during or after the intervention period [22,61,71,74]. Integrating weekly feedback via text messages allows researchers to track participants’ medication adherence over the past week, assess the effectiveness of the messages in supporting medication management, and identify the most impactful elements for promoting adherence [22]. In addition, messages can be tailored for subgroups with different disease management goals. For example, Champoux et al [61] demonstrated that weekly or monthly feedback messages provided patients with an overview of their blood pressure management, including comparisons with previous weeks or months, thereby enhancing personal relevance.

Lauffenburger et al [71] combined electronic pill bottles with tailored text messages in their intervention. This approach enabled real-time data collection on the frequency and timing of pill bottle openings, which were then linked to different types of messages. A reinforcement learning algorithm was used to determine which factors should be emphasized in each message, ensuring that the most relevant message was sent to each patient daily. Similarly, Bartlett et al [74] allowed participants to respond to messages with “like” or “dislike” feedback. Messages marked as “liked” were twice as likely to incorporate the same behavioral components in future messages, whereas those marked as “disliked” were half as likely to be repeated. This personalized approach aligned interventions with participants’ habits and preferences, leading to improved effectiveness. Participant feedback provided valuable user-centered insights that enhanced message design, ultimately improving engagement and adherence to both the messages and prescribed medications. However, implementing feedback systems poses challenges, including the increased complexity of managing a larger message bank, reliance on participants’ willingness to provide feedback, and potential privacy concerns owing to the collection of additional personal data.

In summary, the selected text messages demonstrate a range of content development and tailoring strategies designed to align with patients’ specific characteristics and needs. Each approach highlights its feasibility as a viable method for message development. However, the absence of comparative studies makes it unclear which theory-driven designs or tailored approaches yield the best outcomes for improving medication adherence. To develop more effective interventions, it is essential for designers to consider user preferences, the specific characteristics and needs of target patients, and the available resources for message development. Message design should be adapted to fit different situations. When patient background information or personal characteristics are not available in advance, simpler message designs may be more appropriate. However, whenever feasible, incorporating behavioral theory into message development is recommended, especially when it aligns with the goals of the intervention. Ultimately, a balance is needed between the level of message personalization and the overall feasibility of delivering the intervention.

Strengths and Limitations

This review provides an in-depth analysis of text message design strategies aimed at improving medication adherence. It outlines commonly used theoretical models for message development and tailored approaches, which offer valuable insights for future SMS text messaging interventions. However, several limitations should be acknowledged.

Because of time constraints and limited research capacity, we did not contact study authors or search for related formative publications to retrieve missing information. Consequently, some studies may have been excluded based on insufficient reporting rather than an actual absence of theoretical justification or rationale for message design. Future reviews may benefit from incorporating follow-up strategies to ensure a more comprehensive inclusion of relevant studies.

The study specifically focused on cardiometabolic syndrome, such as hypertension, diabetes, and hyperlipidemia, given their high prevalence and relatively stable treatment regimens. Acute and self-limiting conditions, such as influenza or fractures, typically resolve within a defined period and require minimal long-term intervention, making them less relevant to this review. In addition, this study primarily explored the design of text messages rather than the overall effectiveness of the interventions. Some of the selected studies described tailored message development without reporting outcomes, limiting the ability to evaluate the impact of specific message strategies on adherence. Future studies should evaluate how different text messaging approaches influence psychosocial and clinical outcomes.

Finally, interventions can incorporate multiple communication formats, including text and voice messages, phone calls, or video interactions. While this study focused exclusively on text messages, addressing different health behaviors may require varied approaches. A multifaceted strategy integrating multiple communication channels could offer a more comprehensive solution to overcoming barriers to medication nonadherence.

Conclusions

The comparison of text message designs offers unique insights beyond integrated quantitative effectiveness. By incorporating behavioral theory and tailored content, text messages can more effectively promote health behaviors. For patients with diabetes, hypertension, and hyperlipidemia on stable treatment, text messages are a practical tool for supporting medication adherence. Personalizing the information can enhance its relevance and strengthen patient engagement while incorporating behavioral theory provides a strong foundation for content design. Together, these strategies promise to promote sustained adherence and encourage positive health behaviors.