The design and conduct of the study were approved by the Malaysian Medical Research and Ethics Committee (NMRR ID-15-851-24737) in July 2015.

The study recruited patients followed up at the Outpatient Neurology facility at Hospital Kuala Lumpur (HKL), who were informed about the study and provided consent. HKL is the principle tertiary facility in Malaysia and receives a high number of patients with stroke from different territories in the Klang Valley and throughout Malaysia, with approximately 1000-1200 acute and recurrent stroke cases every year. The Neurology Department is a pioneer in setting up an acute stroke center in Malaysia, which is overseen by a group of neurologists, doctors, and other health services supporting staff.

For the interview, patients were randomly selected via a simple random sampling method, from a list of 89 patients with potential drug-related problems (DRP) [18], who were identified by a clinical pharmacist. DRP is a set of categories of medication issues used by medical personnel to conduct a strategic medication review in order to ensure optimization of the prescribed medication. The issues could be related to medication interactions, dosage appropriateness, adverse events, or adherence. Thus, a patient with DRP would be the best candidate to analyze medication-taking behavior [9]. The inclusion criteria were diagnosis with first stroke in the last 6 months from the initial date of screening and interview (January 1, 2016, until March 30, 2016); taking stroke-prevention medications such as statins, antiplatelets, or anticoagulants; and no memory problems and ability to converse, read, and write in Malay or English. We selected patients who had potential treatment issues such as a subtherapeutic effect with causes related to medication use process (eg, medication not taken). In-depth interviews were deemed most appropriate to build a trustful rapport between the researcher and patients and obtain more comprehensive views. This method also ensures that patients are comfortable and that their thoughts are not suppressed due to the presence of other patients. It was not possible to conduct a focus group discussion due to schedule and venue problems, as these patients had physical immobility or transportation issues. One researcher approached all patients and explained the purpose of the interview and study. The research team agreed to recruit more patients until data saturation [19]. However, of the 32 patients who showed interest, only 10 patients (31.2%) agreed and signed the consent form to volunteer for a personal interview for a maximum of 30 minutes. We completed full semistructured interviews with a total of 10 patients (5 women, 5 men; age range: 44-78 years; Table 1).

In-depth interviews were conducted on the basis of a qualitative open-ended interview guide developed by a nurse educationist, two pharmacist educationists, and a neurologist. This group represented views from different health care professionals involved in stroke treatment and care. The interview guide was pretested on nine poststroke patients for their relevance and suitability in the Outpatient Neurology clinic setting. Although short, the guide was precise in order to prevent burdening patients with many questions while giving them enough time to recall meaningful events (Multimedia Appendix 1). Each patient was given an appointment for the interview. One researcher (JA) led each scheduled participant to a quiet room at the Outpatient Neurology clinic, started the interview after a friendly chat, and continued to brief them about the study. The interviews focused on patient’s experiences of stroke and medication management. The interview was conducted at a relaxed pace, whereby the patient was allowed to pause or have a short break. The researcher (JA) posed prompts whenever necessary or when the conversation was mixed. After 15 minutes of interview, two online video vignettes were shown to the patients. The video vignettes of 2 minutes each were in an animated form, conveying messages on the importance of understanding the disease and its preventative medication and how people perceived their medications (Multimedia Appendix 2). The researchers (JA and SR) developed the video vignettes in English and Malay language to prompt patients to elicit a deeper thought of self-reflection, and in doing so, the video allowed the researchers to obtain a better understanding of the patients’ needs or barriers to improve their health conditions. Vignettes were suitable, as they empowered accumulation of delicate subjective information and are a successful device for inspiring judgments and discernments [20]. The videos that lasted less than 2 minutes had satisfactory face validity and content validity, as confirmed by three poststroke patients, a clinical nurse, two neurologists, and two pharmacists. Patients were asked to describe their experiences of adhering to prescribed medications after viewing the video. All responses were audio recorded, and the researchers ensured the confidentiality of the recordings.

All interviews were transcribed verbatim, checked by another researcher, and then documented for data analysis. All transcripts were manually coded and classified using the inductive thematic analysis methodology [21]. The methodology were as follows: (1) understanding the transcripts, (2) diagramming key points into codes, and (3) summarizing the mapped codes into subthemes and emergence of major themes. Six scripts in the Malay language were translated by an independent translator. Two researchers (JA and SR) reviewed the transcripts and met intermittently to discuss the themes, outlines, and issues established in the data. Once the themes were generalized, they were verified by two other reviewers (PW and CS) to ensure uniformity, precision, and quality [22].

Two major themes related to medication-taking self-efficacy were identified: self-efficacy in taking the effort to understand stroke and its preventative treatment for recurrent stroke and self-efficacy in taking prescribed medications to prevent stroke. A majority of the subthemes discussed by the patients referred to the individuals’ necessity, needs to attain the specific self-efficacy, and the challenges considered to be barriers to attaining those needs. Thus, the results section is organized into two sections that discuss the needs and barriers of both main themes. The first portion of each section discusses the needs and the second portion discusses the barriers. The themes emphasized on individual well-being, communication, and independence, which were elicited in response to specific probing of the elements related to medication-taking self-efficacy.

Normally, confidence is thought to be adequate to carry out a task, and it is simple to adhere to medications. Confidence has a positive effect on self-efficacy; however, this may not be the case vice-versa. The ability to understand, think, plan, and use prescribed medications to sustain medication adherence and ensure treatment effectiveness depends on the individual’s self-efficacy levels [15,16,23]. Hence, self-efficacy in understanding and taking medication appropriately found its specific niche in nonadherence and has been studied for more than a decade [24]. This qualitative study managed to obtain an overview of poststroke patients’ needs and barriers toward sustainable medication-taking self-efficacy.

Patients expressed a lack of the understanding that every stroke event portrays different symptoms, and inability to control stroke risk factors increases the risk of a recurring stroke event. This finding was consistent with a previous qualitative study of stroke patients’ perception, which highlighted the difficulty of identifying various atypical stroke symptoms [25]. It was clear that the patients were not proactive. About 5 of the 10 patients (50%) knew that relevant knowledge was important, but they were laid-back even though they were experiencing a stroke. Moreover, the lack of knowledge and awareness was limited to not only illness but also its preventative treatment. A trend of negative responses from those with a lower health literacy level was observed: 2 of the 10 patients (20%) had poor literacy levels. To achieve self-efficacy in order to gain an understanding of stroke and its preventative medication purpose, patients were faced with barriers such as inadequate or unreliable sources of information and poor communication with prescribers, which dampen their knowledge search. Importantly, unassessed low health literacy complements the knowledge attainment barriers [25].

Other than being proactive, perception and belief are essential in the patients’ decision making process in medication-taking behavior. Perception is subjective of what an individual thinks about an issue and how one is influenced by one’s beliefs. Thus, the intensity of medication-taking self-efficacy depends on the level of patients’ perception of their disease and its treatment [17,26]. Therefore, we could summarize that patients with similar underlying stroke risk factors or severity may not have the same perception of illness and belief about their medications even if they were assumed to have the same knowledge about stroke and its treatment. It is possible that varying physical and emotional experiences with stroke cause differences in perception and beliefs. Therefore, actions of poststroke patients’ toward adherence are steered by their perceived susceptibility and severity of stroke and led by their perceived benefits or barriers of medication taking, which are influenced by their beliefs [27,28]. Nevertheless, diverse populations and cultures have a strong influence on these beliefs too [29,30]. Hence, the success of a behavioral intervention depends on these factors as well.

Self-efficacious patients tend to develop or learn skills on their own in an effort to overcome worsening of the disease condition. They choose self-empowerment and a high responsibility of medication management to ensure optimal therapy effect. Some self-care examples are monitoring blood parameter, scheduling medication intake using the pill reminder, preparing appropriate medicine storage, and being aware of any allergy symptoms [31-33]. In the same way, self-efficacy is highly influenced by parallel changes in self-motivation and quality of life, and these changes vary between individuals according to the effect of personal experiences (mastery experience) and the perspectives of stroke and belief in the preventive treatment [12,34,35]. In other words, self-efficacious patients living with long-term illnesses tend to put effort into attaining an in-depth knowledge about their disease and treatment, creatively solving problems, and increasing their expertise of improving their own health [36]. Thus, in order to sustain medication-taking self-efficacy for optimizing treatment effectiveness, health professionals should ensure that their intervention fulfils patients’ needs of self-efficacy to understand and use medication appropriately and that it addresses their barriers toward the intended self-efficacy.

This study was based on the viewpoints of 10 poststroke patients. We experienced a high drop-out rate, as we were unable to secure appropriate funding when patients requested for their financial reimbursement. Therefore, generalizability of the study findings was limited by the small sample size. There were also challenges in distinguishing the needs and barriers of language or health literacy level. Hence, more in-depth interviews with validated tools are required to categorize varying health literacy levels among poststroke patients. This sample may not have captured the full range of the needs and barriers of poststroke patients. However, this did not detract from the clear importance of the two main needs of medication-taking self-efficacy: self-efficacy in gaining an understanding of stroke and managing the intake of preventative medications for stroke.

The strength of this study is that it revealed emotional experiences of poststroke patients under an unconstrained technique with the help of focused prompts and video vignettes to elicit deeper thoughts compared to close-ended questionnaires. The themes were discussed in specific probes of self-efficacy related to medication adherence dimensions. Hence, this study indirectly proposed the usefulness of video in modifying focused or planned behavior [37,38].

Our findings were exploratory, and hence, the outcomes on verifying the association between the needs and barriers and medication-taking self-efficacy should be confirmed using quantitative measures with adequate sample size. Nevertheless, the results provided a subjective perspective of poststroke patients based on their experiences, and thus, it is crucial to consider these viewpoints as a groundwork for future interventions related to understanding medication taking and its self-efficacy.