Insights into how Malaysian adults with limited health literacy self‐manage and live with asthma: A Photovoice qualitative study

Abstract Background Adjusting to life with a chronic condition is challenging, especially for people with limited health literacy, which is associated with low compliance with self‐management activities and poor clinical outcomes. Objective We explored how people with limited health literacy understand asthma and undertake self‐management practices. Design We adapted the arts‐based qualitative methodology Photovoice. Setting and Participants We sampled ethnically diverse adults with asthma and limited health literacy from four primary healthcare clinics in Malaysia. After a semistructured in‐depth interview, a subset of participants took part in the Photovoice component in which participants undertook a 2‐week photo‐taking activity and subsequent photo‐interview. Interviews, conducted in participants' preferred language, were audio‐recorded, transcribed verbatim, translated and analysed thematically. We used the Sorensen's framework (Domains: access, understand, appraise, apply) to describe participants' experience of living with asthma, what they understood about asthma and how they decided on self‐management practices. Results Twenty‐six participants provided interviews; eight completed the Photovoice activities. Participants with limited health literacy used various sources to access information about asthma and self‐management. Doctor–patient communication had a pivotal role in helping patients understand asthma. The lack of appraisal skills was significant and experiential knowledge influenced how they applied information. Self‐management decisions were influenced by sociocultural norms/practices, stigmatizing experiences, and available social support. Conclusion Locally tailored multilevel interventions (interpersonal, health system, community and policy) will be needed to support people with limited health literacy to live optimally with their asthma in an ethnically diverse population. Patient/Public Contribution Patients were involved in the study design, recruitment, analysis and dissemination.


| BACKGROUND
Limited health literacy is a global public health problem that limits people's ability to self-manage chronic conditions such as asthma. 1,2 Health literacy is defined as a person's cognitive and functional abilities to respond to the healthcare system's demands to care for their health. 1 About twothirds of people with asthma in a Malaysian population were found to have limited health literacy. 3 In people with asthma, studies have shown that limited health literacy is associated with increased unscheduled emergency visits and hospitalizations. 4 Support for self-management in the form of regular healthcare professional reviews and a written action plan improves asthma outcomes. 5 As recommended in asthma guidelines, 6,7 self-management requires a good understanding of asthma and informed action on the part of the patient. However, asthma selfmanagement may be challenging to implement and/or support in many health settings. 8,9 Tailoring self-management support for people with limited health literacy is crucial but is rarely provided. 10 Asthma affects almost 360 million globally, 11 yet it is a neglected noncommunicable disease in many health settings. 12 In Malaysia, the prevalence of adult asthma was 4.2%, with 1.2% of deaths related to asthma in 2006. 13 To date, qualitative explorations on asthma and health literacy issues are mainly from high-income countries. [14][15][16] A mismatch between patients' and doctors' expectations are among the challenges faced by people with limited health literacy. 14 Other identified barriers include the doctor's communication style, language and mistrust in the doctor-patient partnership due to perceived racial bias and lack of cultural sensitivity. [14][15][16] Culture, health beliefs and experiential knowledge influenced self-management behaviours leading to actions that may not align with evidence-based practice. 8,14,15 Individual choices and health decision-making are governed by deeply embedded sociocultural norms and practices within a community. 2,17 In this study, we recognized the sociocultural diversity of the population in Malaysia and the need to allow creative ways to explore health literacy, asthma and self-management practices.
Photovoice, an arts-based qualitative methodology that employs collective conversations, 18 to understand the issues and strengths of the community and the health system. 19 We adapted Photovoice by combining qualitative interviews and photo-taking activities to explore experiences and employ innovative approaches to disseminating the voice of the community. We aimed to explore how people with limited health literacy in Malaysia understand asthma and decide on self-management practices.

| STUDY DESIGN
We adapted Photovoice, 18 an arts-based qualitative methodology ( Figure 1) to generate understandings of health literacy challenges among people with asthma in Malaysia. The study included four stages. Stage 1 involved one-to-one in-depth interviews. Stage 2 involved training and photo-taking activity. Stage 3 involved one-toone Photovoice discussions with the subset of participants who took photos. A small-scale exhibition and sharing of findings through local social media networks took place at the end of the study to amplify people's experiences living with asthma among stakeholders (Stage 4). Wang's classic Photovoice description was of a one-way process where recruited participants underwent training, photo-taking activities and discussion as a group. 18 Although participants were given the freedom to take the photographs as individuals, only key photographs were chosen and discussed in focus group discussions. 18 The first group engagement was to train in Photovoice technique and understand the research assignment followed by group dialogue with policy-makers using key photographs. 18 Wang and Burris 18 completed the Photovoice stage with a final participants' evaluation of the process. We adapted this process to conform to our population's sociocultural needs. We introduced an in-depth interview in the first stage as a strategy to secure initial engagement, to foster relationships and trust among the research participants. We invited interviewees to take photographs and discussed all their photographs in a confidential interview.

| Setting, recruitment and sample
The study was conducted in 2019 at four primary health clinics in Klang District, a central region of Malaysia (see Table 1 for a description of Malaysia's healthcare system). These clinics were chosen to reflect a range of urban and suburban populations enabling the recruitment of a broad range of participants. Pictorial advertisements and easy-read information about the study were placed at the study sites. H. S. explained the study procedures to potentially interested participants (face-to-face or through phone calls) and assessed eligibility. Video information sheets helped potential participants, regardless of reading ability, visualize the research process, understand what was involved and make an informed decision about participation. The video was made available online where participants could watch it in their own time 9 (https://www.ed. ac.uk/usher/respire/health-literacy-asthma-malaysia/information-leaflets) before they gave written consent.
Adult patients (aged ≥18 years) with physician-diagnosed asthma on treatment from any of the four clinics, and with limited health literacy (defined as <33 on the validated Malay version of health literacy scale [HLS] 23 ). We purposively recruited a maximum variation sample of participants based on demography (age, gender, ethnicity) and asthma control assessed using the asthma control questionnaire (ACQ). 24 For those who agreed to participate, an initial interview was arranged at the most convenient clinic for the patient.

| Data collection
Written informed consent was obtained before each activity. In Stage 1, the participant was invited for an initial in-depth, semistructured interview. Using a topic guide (Supporting Information Appendix S1), the interviews started with a general discussion about personal experiences living with asthma. We then explored key areas of interest concerning participants' experiences of living with and managing asthma within and outside the home and interactions with healthcare professionals. H. S., a Malaysian family physician who designed the study in discussion with the wider research group, conducted all the interviews, assisted by an interpreter, T. P. Interviews were conducted in Malay, English, Tamil or Mandarin according to participants' preference, and they could move between languages if they so wished.
The photo-taking activity (Stages 2 and 3) was offered to all participants at the end of the initial interview-there was no selection • Malaysia is a multiracial country comprising three main ethnic groups: Malays, Chinese and Indians, each with their own culture and language.
• The national language, Malay, is used as the main medium of instruction in both primary and secondary national schools.
• English is learnt as a second language.
• At the primary school level, schools may also offer Mandarin and Tamil mediums of instruction. H. S. started the Photovoice interview using the photographs that the participant had agreed to share with the researcher, one at a time.
Participants were encouraged to describe where, why the photographs were taken and what it meant to them particularly around the aspects of living with asthma. During the interview, participants were asked if there had been any situations they decided not to capture during the photoactivity. H. S. encouraged participants to explain the reason behind these decisions (or why they decided not to share specific photographs with the researcher), though they were free not to explain their decision. The absence of these photographs challenged participants to reflect on pertinent-but hidden-issues such as social norms and practices. Participants were given space to add an explanation or accompanying text for each photograph. All interviews (in-depth interviews and Photovoice interviews) were digitally audio-recorded, using an encrypted audiorecorder.

| Data analysis
Recorded interviews were transcribed verbatim in the original language, checked for quality and accuracy and anonymized before translation into English using back-to-back translation methods. 22 Where it was not possible to translate metaphors, idioms, or culturally specific expressions into English, the original language was maintained, and explanatory footnotes provided. 22,25 Table 3 describes steps for data analysis.

| Patient and public involvement (PPI)
The research was discussed with PPI panels in Malaysia and the Asthma UK Centre for Applied Research (AUKCAR). 26 They reviewed research documents (participant information sheet and consent forms, animated videos to overcome literacy problems, flyers and photo-taking guidance) to improve readability and clarity. H. S. spoke with two PPI colleagues to learn more about how people with asthma view research and to discuss the topic guide. For example, one member suggested using common local terms to describe difficulty in breathing, such as semput for Malay participants. One of the PPI colleagues piloted the photo-taking activity and made some practical suggestions (e.g., allowing 2-week for the phototaking activity instead of 1).

T A B L E 2 Remit of the photo-taking activity
Please take pictures that show: (a) What is it like to live with asthma?
(b) What helps you, or what makes it more challenging to live with asthma?
(c) How do you learn about asthma, and where do you get information about it?
T A B L E 3 Steps taken for data analysis 1 1.
The analysis of the transcripts and the photographs were iteratively informed, but not restricted, by the health literacy framework by Sorensen.

2.
All the transcripts from the initial in-depth interviews and photo-interviews were uploaded in the computer-aided qualitative analysis software (Nvivo 11).

3.
After reading and rereading to immerse in the data, H.S. conducted a preliminary analysis, which was then explored with I. Y./P. Y. L./S. S. G./ H. P.

4.
H. S. then deductively coded the transcripts using the concerning asthma related-health information; access, understand, appraise and apply.

5.
Each interview statement was coded into one of the broad themes of Sorensen's health literacy framework.

6.
We added additional categories to each theme to ensure we captured unique themes and conforming/nonconforming concepts against the initial assumptions about health literacy.

7.
Refinement, agreement of categories and subsequent themes were done in iterative discussion with the multidisciplinary research team (I. Y., P. Y. L., S. S. G., H. P.), providing diverse clinical, health system and social research backgrounds.

8.
The team identified and structured the themes presented in this article. We discussed preliminary analysis and captions accompanying the photographs with the participants enabling additional or discordant themes to emerge.

| Trustworthiness, reflexivity and power dynamics of research
As researchers, we significantly impact on how data are collected, shaped and analysed. Trustworthiness refers to the degree of confidence in data, interpretation and methods used to ensure the quality of a study. 27 We adopted the Lincoln and Guba (1985Guba ( , 1989 criteria to evaluate the credibility, transferability, dependability and conformability of our qualitative work. 28,29 We provide a detailed description of the strategies used in Supporting Information Appendix S3 with a summary in Table 4.

| Description of data set
In total, 26 participants completed the initial interviews, 12 agreed to the photo-taking training session, but only eight completed the photo discussion.

| Presentation of findings
We describe participants' understanding of asthma and how they decided on self-management practices based on the four domains of health literacy: (i) access; (ii) understand; (iii) appraise; and (iv) apply. 1 For each domain, we report the themes and subthemes that influenced participants' engagement with health service support for asthma selfmanagement. See Table 6 for a summary. Figure 2 are the photographs taken by the participants, which are related to many of the quotes.

| Access to information on asthma and selfmanagement
Participants described two sources of information regarding asthma and Other than health information given by the healthcare professionals, information was typically assessed against their experience of living with asthma. For example, a participant highlighted how she weighed up the information about asthma triggers that she had found on social media and related this to her own experience before acting on the information ( Figure 2C).
If I see posts on social media and everyone else agreed on it, then probably it's true. People said that the airconditioning system in the car makes people cough.
It's the same with me, but, I still use it at times. Like Some who subscribed to the concept of Ying and Yang used warm water to counter asthma attacks. The hot and cold concept is widely understood in the Chinese culture, but it is also a concept accepted in Ayurvedic medicine in the Indian community. An Indian participant described and practised Ayurvedic medicine recommended by her mother-in-law to avoid being labelled as 'sick'.
My mother-in-law practised Ayurvedic medicines. Stigmatizing experiences were challenging and inevitably influenced self-management decisions. Sports and physical activities such as playing football were identified as essential activities in embodying health identities, particularly among two young men in this study.
Thus, for these participants, using an inhaler before a game or during a match demonstrated 'weakness' and invited unwanted social reactions. One participant shared a photograph of a football field where he used to play and said ( Figure 2F In contrast, formal social support helped some participants adopt a strong sense of identity and a positive attitude towards guidelinerecommended care. One participant who joined a patient support group to improve physical health described how the weekly gathering had helped him embrace his identity and enabled him to empower others ( Figure 2G).
I took this photo during one of our weekly gatherings. take the turn to organise this. Sometimes they get together and cook at their house, so we will enjoy some home-cooked meal. It is during the mealtime that we chatted, we shared stories-good and bad. We feel belonged to something, and we are not ashamed of who we are; a person with asthma or diabetes or whatever it is. So, as a person with asthma, I actively studies. 15,31,32 Although healthcare professionals are aware of this responsibility, they may selectively provide health information to patients, that is those with poor disease control or those whom they perceive may understand the health information. 8,31,32 Some patients may even wish to discuss CAM with their healthcare professionals but fear negative reactions. 33 The healthcare professional's barriers to discussing CAM with patients include lack of a trusting relationship, personal disapproval of CAM and lack of evidence making a conversation about CAM uncomfortable. 34 Our participants wanted to understand asthma and manage it but described challenges in their communications with the healthcare professionals. People with limited health literacy may not necessarily tackle communication barriers and may not use strategies such as establishing rapport and clarifying queries. 14,32,35 How health information is communicated (e.g., using universal health literacy precautions in providing health information) 36 is not only important to help patients understand their disease but is also crucial in dismantling paternalism in doctor-patient relationships. 37,38 Universal health literacy precautions include breaking down information and instructions into small steps, assessing comprehension through teach-back cycle and use of visual aids. 36,39 Information tailored to the patient's health literacy needs is more likely to be translated into positive health behaviours. 37,40 Knowledge and illness experiences from family, close friends and stories shared via social media were trusted informal sources of information among participants in this study. It has been shown that people look to the Internet (especially social media) to find others with similar illness experiences and use these stories to learn new strategies or confirm that their health behaviour is appropriate. 32,41 The danger is that these experiences may not reflect accurate health information or recommended behaviours. A recent review that highlighted social media as a powerful channel for health communications similarly cautioned about unreliable content, lack of privacy of personal information. 42 Others have suggested that superusers (users who write a large number of posts in online health communities) have the potential to reach a wide population and cost-effectively support such communities by providing accurate information about asthma and awareness about guidelinerecommended practice. 43 The lack of analytical skills to evaluate health information described by our participants could exacerbate the spread of false information and fearmongering against, for example, the use of preventer inhalers. Illness perceptions and personal health beliefs are grounded within social-cultural norms and practices. Nonadherence, for example, has been linked to mismatches between patients' common-sense interpretations of their long-term illness and treatment and medical reasoning. 44 31,50 Adjusting to life with asthma can be challenging, and the participants with limited health literacy in our study used various strategies to negotiate how to live with asthma despite difficulties with understanding essential health information. Healthcare professionals may not be aware of these struggles and may lack the skills to address the patient's life issues and sociocultural challenges in disease management. 8 Stigma directly impacted the lived experiences and decisions made by our participants. Goffman (1963) defines stigma as an 'attribute that was deeply discrediting and reduced the bearer from a whole person to a tainted, discounted one'. 51 In many societies, to be healthy is the norm to which ill-health ought to be restored. 52 Those with chronic illness (exemplified amongst our participants by the footballer and the daughter-in-law with asthma) struggled to meet this societal expectation and were assigned a sick identity, potentially affecting mental health. 53

| The role of adaptations in cross-cultural research
In classic Photovoice methodology, the target community's voicemaking was conducted collectively. 18 As participatory action research, it is bound to flexibility in terms of its conduct, 56

| Strengths and limitations
The Photovoice approach is conceived as participant-led research, giving voice to people to communicate their experiences in their own way. We adapted this to a more researcher-led approach, which has implications for the power balance between researcher and participants. For example, the researchers are clinicians, introducing a hierarchical power imbalance, 58 which might deter coconstructions of knowledge with the participants and hinder understanding illness experiences in research interviews. We, therefore, put in place strategies to minimize the impacts on our research findings. Firstly, we used strategies described by Lincoln and Guba (see Table 4) to ensure the patients' voices are represented in the findings, 58 which might deter coconstructions of knowledge with the participants and hinder understanding illness experiences in research interviews. We, therefore, put in place strategies to minimize the impacts on our research findings. Firstly, we used strategies described by Lincoln and Guba (see Table 4) to ensure the patients' voices are represented in the findings. 28 We provided a participant video to supplement written information and help improve understanding of the research. Infographics explained the conduct of Photovoice and encouraged creativity in completing photo-taking tasks.
The initial in-depth interviews were representative of the intended range of demographic characteristics and asthma control, though only a minority of participants took part in the Photovoice phase, in part because of the time and commitment needed to complete this. As a result, we lacked older participants and those of Chinese ethnicity among those who completed the Photovoice interviews. In classic Photovoice methodology, we would only have recruited participants willing to commit to the photo-taking. Our adapted approach with initial qualitative interviews meant we heard from a broader range of people enabling us to embed Photovoice insights in the broader context.
The use of a predetermined framework ensured a structured approach to developing our topic guides and informed our analysis.
We remained aware, however, that this could restrict our data collection and interpretation of novel themes. We, therefore, explicitly looked for discordant views and unexpected themes. The Photovoice interviews enabled the participants to determine the topics for discussion, and because these progressed in parallel with the interviews, we could accommodate novel insights into the ongoing interviews. The presence (or lack) of photographs triggered discussion of issues important to the participants enabling them to set the agenda. Language barriers were overcome as participants used photographs to express their experiences, and the interviewer prompted discussion about persons, objects, or settings depicted in the photographs. Conversations about photographs they were unable to capture, triggered reflection on important-but hidden-concerns like mental health. 17 The interviews and photographs provided a data source triangulation that enhanced understanding of health literacy experiences in real life and challenged us to think beyond the health system to how health literacy operates at the societal level. Interviews were conducted in the participant's spoken language to capture the cultural nuances, and we preserved this during the translation process.

| Implications for policy, practice and research
Tailored asthma care for patients' health literacy needs has a pivotal role in improving understanding of asthma and guidelinerecommended self-management practices. Multilevel interventions (interpersonal, health system, community and policy) may be needed to achieve this. Based on our findings, we propose a framework for considering changes that might tackle health literacy issues in asthma care ( Figure 3) and which potentially, may be adapted to other chronic conditions. Future research is needed to determine the feasibility and effectiveness of such multilevel interventions in low-middle-income countries (LMICs) settings.

| CONCLUSIONS
Health literacy and asthma self-management remain a challenge in many healthcare settings, with particular challenges in multiethnic LMICs. We The photographs included in this article were taken by participants of this study who had permitted the photographs to be used in any format within academic reports, publications and presentations of noncommercial nature.