Designing a text messaging program to increase adherence to medication for the secondary prevention of cardiovascular disease

Background

People with a history of cardiovascular events have a five times higher risk of new events than people without no such history [1].Secondary prevention medications (anti-platelet therapy, ACE inhibitors, beta-blockers and lipid lowering therapy) are effective in reducing the risk of death and myocardial infarction in patients with coronary heart disease. Meta-analysis of randomized controlled trials show that in patients with existing coronary heart disease long term anti-platelet therapy reduces major vascular events (MI, stroke or vascular mortality) by about a quarter (OR 0.75 (95% CI 0.71-0.79); ACE inhibitors reduce cardiovascular mortality by just under a fifth (RR 0.83, 95% CI 0.72 to 0.96); beta blockers reduce mortality by almost a quarter (OR 0.77, 95% CI 0.69 to 0.85); and lipid lowering therapy reduces coronary mortality by about a fifth (RR 0.79, 95% CI 0.75 to 0.83) [2].

However, cardiovascular patients have poor medication adherence, the consequences of which include premature death, increased risk of further cardiovascular events, hospitalization, and high financial cost for the healthcare system [3] [4] [5] [6] [7]. Therefore, strategies to improve medication adherence to cardiovascular drugs must be a health priority [8] [6]. Since 2002, text messaging, or short message service (SMS), has been part of m-health strategies to improve health and change behavior [9]. Interventions using SMS have been shown to be effective for smoking cessation, adherence to antiretroviral therapy [10], diabetes self-management, weight loss, and physical activity [9]. As for cardiovascular treatment, several randomized clinical trials (RCT) using text messages to improve medical adherence in primary as well for secondary prevention have produced promising, although not yet conclusive, results [11] [12]. Systematic reviews conclude that existing RCTs have high risk of bias, so the evidence is uncertain, and most were performed in high-income countries, so the evidence may not generalize to low income countries [11].

The ongoing randomized clinical trial (RCT) “TXT2HEART COLOMBIA: Evaluation of the Efficacy and Safety of Text Messages to Improve Adherence to Cardiovascular Medications in Secondary Prevention” (https://clinicaltrials.gov), involves an SMS-based intervention aimed at improving adherence to cardiovascular medications amongst adults with history of arterial occlusive events living in a metropolitan area of Colombia. This paper presents the design and validation process of text messages developed for the RCT based on Transtheoretical Model of Health Behavior Change (TTM) [13]. This model, also known as the Stages of Change, integrates cognitive, behavioral, and motivational factors at different times in the behavior modification process. It has been recently applied to develop interventions that use SMS to change behavior related to nutrition [14], physical exercise [15] [16] and hypertension [17]. The publication of this phase of the study will contribute to filling gaps in the literature identified by Adler et al [11] by describing the development, message content, and the program design of the intervention. It also fulfils the requirement to publish detailed descriptions of interventions to avoid research waste [18] and allow a growing evidence base about which interventions work for what reason [19].

Method

Our methodological approach is described in Fig. 1. We used the model reported by Abroms and colleagues’ [20] for developing and evaluating text messages-based interventions. This model describes a process in which the intervention created is based on theory and evidence, the target audience is involved to ensure the intervention is engaging and useful, and there is a focus on implementation from the outset. Each step will be described in this section.

• Download figure
• Open in new tab

Figure 1. Process for text message intervention design

Literature research: the first step consisted in conducting literature research related to the target audience and behaviour change theories. A successful health program needs evidence-based strategies [21] [22] and must focus on understanding the health behavior that characterizes the population [20]. Literature research aimed to: 1) understand the factors associated with low adherence to cardiovascular medication and 2) identify the leading models and appropriate techniques for modifying health behaviors related to medication adherence. We performed literature research from February 2015 to June 2016 in the following database: Scopus, Ovid, Medline, Cochrane Library, EBSCO, Pub-Med, Science Direct and Google Scholar. The keyword terms were: Medication adherence, Medication adherence and cardiovascular Disease, Social support and cardiovascular risk, Health models, and M-Health. We reviewed the following behavior change models: The Theory of Reasoned Action [23], Social Learning Theory [24], the Transtheoretical Model of Change [13], the Precede Model [25] and the Behavior Change Wheel [26].

Objectives and techniques: As a result of this literature review, three main objectives for the intervention were established based on the behavior change goal and the theoretical aspects of the TTM: (1) raising awareness about medication effectiveness, (2) promoting self-care, and (3) offering guidelines for behavioral strategies to increase adherence (see Table 2). When the TTM was first developed, Prochaska and DiClemente [13] proposed a relationship between the perception of illness, or more appropriately the rise of awareness of illness, and the intention to change behavior and the subsequent change.

Text message program design process: Our behavior change goal was to improve medication adherence in secondary prevention for patients with history of occlusive arterial disease. In the literature search we found several studies indicating the importance of evaluating barriers that affect adherence to medications in cardiovascular patients [5] [27] [28]. We carried out qualitative research to identify barriers to adherence in our setting and compared our findings to the existing literature.

We used focus group methodology to interview patients with a previous history of cardiovascular events. Participant inclusion criteria were: male and female adults (18 years and older), ownership of a mobile cell phone with an active line, ability to read text messages and to give informed consent, and at least a six-month history of any of the following cardiovascular diagnoses: coronary artery disease (including non-primary coronary revascularizations), ischemic stroke, and peripheral artery disease or atherosclerotic aortic disease. These selection criteria mirrored the eligibility criteria for the TXT2Heart trial. Recruitment focused on patients who had an occlusive arterial disease event and been hospitalized at the Fundación Cardiovascular FCV, which provided clinical evidence of the event as well easier access to the patients. The inclusion criteria were narrowed to events occurring between six months and five years prior to the study. Participants were randomly selected from the FCV databases using SQL (Structured Query Language), with specification of the date of the event (between July 2011 and July 2015), ICD 10 diagnosis, and medical procedure such as angioplasty, revascularization, and other interventions for the cardiovascular diseases mentioned. After the patients gave their consent, a trained physician reviewed the patients’ electronic clinical histories to confirm the diagnoses and dates of the occlusive event.

Text Message Library: The text message designers worked as an interdisciplinary team, which included an epidemiologist, a pharmacologist, a specialist nurse in health and quality audits, and three psychologists. Each team member developed messages according to the following recommendations: length up to 160 characters (including spaces), use of appropriate language, avoidance of abbreviations and misspellings, and informal punctuation [20]. Most importantly, for the messages content the team considerate the findings from the focus groups (categories and subcategories) and TTM (Stages of Change). Once the messages were written, the interdisciplinary team held several discussion meetings to evaluate and classify each text message, considering aspects such as message pertinence, intention, and clarity.

The messages were classified as viable, editable or as messages to be discarded. Those identified as editable were rewritten, and some were also eliminated. Then, the messages were organized using a matrix and classified according to the categories and subcategories related to the barriers to adherence that were identified in the literature and the focus groups. This helped to visualize how many text messages corresponded to each category and subcategory, and to develop new ones in order to have at least 12 messages in each category.

Expert validation: The text messages were submitted for evaluation by 13 experts. The panel consisted of several different types of health professionals, including psychologists (clinical, psychometric, and general), nutritionists, nurses (practitioner, administrative, and researcher) and physicians (general practitioner, cardiologist, interventional cardiologist, and with expertise in prevention and health promotion). The experts were selected through convenience sampling and the inclusion criteria were knowledge of the topic (cardiovascular disease for physicians and behavior change for psychologists), research experience, and test/trial validation expertise. These professionals were invited to participate through a letter that explained the purpose of the validation. After accepting, they received two documents: (1) instructions for validating the text messages, which included descriptions of the categories, stages of the TTM, and aspects to be evaluated (clarity, coherence, and relevance), and (2) an Excel template containing the text messages, the categories and subcategories, aspects to be evaluated, and an additional space to record observations. The psychologists participating as experts also evaluated the coherence between the SMS and the TTM of change.

All the judges evaluated the messages for clarity, coherence, and relevance on a scale of 1 to 4, with 4 as the best score. An average score of 3.7 to 4 for the three aspects and a Kendall’s W of .85 or higher were indicators of strong agreement among the judges.

Pretesting and revising the text messaging program: the third step included feedback from prospective users about messages presented to them face to face, international researchers’ opinions, and finally a test for the deliverability of messages to cell phones. To obtain feedback form users, the participants from previous focus groups were invited to attend a new focus group and to evaluate the text messages. A total of 9 cardiovascular patients attended the meeting. The focus group was facilitated by a psychologist and had four observers (one physician, one nurse, and two psychologists). Participants were asked to evaluate clarity, coherence (according to the categories), and the importance and relevance of the messages. The final messages were presented to international experts in the development of messaging interventions for comments. Lastly, a test was conducted to assess the delivery of the text messages program using the CommCare and Telerivet platforms. To this end, some of the text messages were sent to the research team’s cell phones.

Results

This section describes the results from the different steps presented in the Methods section.

SMS Construction

A total of 415 messages were generated by the interdisciplinary group. They classified 212 messages as viable, 116 messages as editable messages and 87 messages to be discarded. The result of this stage was a text message library of 343 SMS to be evaluated by the judges (see Figure 2). The validation by experts resulted in a total of 133 messages that met the criteria of agreement (at least an average of 3.7 for the three aspects and a Kendall’s W of .85 or higher among the judges).

• Download figure
• Open in new tab

Figure 2. Process to Construct Text Messages

Discussion

The objective of this paper is to present the development of an intervention program aimed to increase medication adherence in cardiovascular patients, using technology and a behavior change model. The process was guided by a mixed methods approach that included focus groups with patients, an interdisciplinary team to create the text messages and expert validation. Beginning with the literature search, each step was informed by the findings of the previous one.

Strengths of the approach described include the focus on sustainable behavior change. Hence, the text messages were not reminders for medication intake or appointments, but an intervention to increase awareness and commitment to medication taking. The use of theory (TTM) to inform the content and delivery schedule of the text messages is a strength of this program design, as interventions based on theory have been shown to be more effective than those not using theory (ref), The use of the TTM resulted in a conceptualization of change as a process rather than a one-time event and resulted in a 12-month intervention program with SMS reflecting movement through the stages of change. While the participants are in the first’s stages, motivational and informative interventions are relevant. Therefore, in the program design, the messages sent to the cell phones during the beginning of the intervention, focused on these areas. This action is common for text messaging programs during important behavior change periods [20].As the participants move to subsequent stages, the messages are related to decision making and assuming responsibility. In the action stage, even though the frequency of SMS decreased, the stage has a greater number of messages in comparison with the others, since it aims to develop the adherence behavior. This variation in the frequency, especially decreasing the number of SMS, has been shown to be more effective than a fixed frequency [35].

Another strength of the development strategy was that the SMS were written by an interdisciplinary team of medical specialists and psychologists. This ensured the intervention reflected both clinical knowledge about cardiovascular diseases and the effect of medication and psychological knowledge about cognitive and motivational aspects of behavior change. For example, the physicians developed the messages related to knowledge of the disease, treatment, and the importance of taking medication, while the psychologists offered guidelines about the appropriate way to communicate the information through an SMS, in accordance with the TTM.

The focus groups with cardiovascular patients at the beginning and at the end of the process were important to identify the target population’s perceptions of important barriers to medication adherence and user test the SMS for clarity, acceptability and probable impact in the target population [36]. Expert validation provided additional quality assurance.

Conclusion

We demonstrated the feasibility of developing unidirectional messaging intervention targeting adherence to medication for secondary prevention of heart disease. Considering the long-term condition of the cardiovascular disease, a SMS intervention can be also contemplated as an accompanying experience using technology. It can be useful as a tool to offer support for the patient, with the conviction it comes from expert, as an incentive from the health system.