Study design and setting

Embedded within the Medical Research Council framework for design and evaluation of complex interventions25, this pre–post feasibility study was conducted in an urban public primary-care clinic in the district of Klang, Selangor, Malaysia between September 2019 to July 2020. The study protocol was registered with BMC ISRCTN Registry [ISRCTN87128530; prospectively registered: September 5, 2019, http://www.isrctn.com/ISRCTN87128530]. The state of Selangor was chosen as it has a high prevalence of adults with asthma (22%), especially in urban communities as well as the highest prevalence of limited health literacy in Malaysia at 75%26.

Participants

The study participants were recruited from one of the primary-care clinics under the Klang Asthma Cohort (KAC) registry (a clinical asthma patients registry) using an Excel-generated simple random table by a research member, based on the inclusion and exclusion criteria in Table 1. Klang Asthma Cohort is a cohort of 1280 people with asthma recruited from six primary-care clinics in Klang who are willing to be approached for future research. A detailed description of KAC can be accessed at https://www.ed.ac.uk/usher/respire/chronic-respiratory-disorders/asthma-care.

Table 1 Participant eligibility criteria.

Participants were contacted via a telephone call (to avoid written communication in people with limited literacy) by a trained research assistant who provided a detailed description of the study. Those who agreed to participate in the study met face-to-face with the research assistant at the clinic to provide written informed consent and to answer the baseline questionnaire.

As this was a feasibility study, a formal sample size calculation was not required. Seventy participants were recruited, which was deemed to be adequate to inform the feasibility of delivering the intervention27.

Usual clinic care and self-management support

The selected primary-care clinic has a dedicated asthma clinic that operates one afternoon a week involving medical officers, pharmacists and nurses. Medical officers are doctors without postgraduate training who work in primary-care clinics under the leadership of specialist family physicians. They are trained to assess asthma control, check the use of peak expiratory flow rate, recommend appropriate treatment, and deliver supported self-management including a text-based asthma action plan, and as the participants’ usual doctor, continued to provide care throughout this study. The pharmacists taught inhaler technique and discussed adherence to medications and asthma action plans. The nurses provided asthma education. For participants in this feasibility study, the clinic management continued as usual, but a pictorial asthma action plan was provided instead of the standard text-based action plan.

Intervention

The intervention consisted of a pictorial asthma action plan (see Supplementary Fig. 1) instead of the text-based action plan incorporated within the existing self-management education and support, and is described in Table 2 using the Template for Intervention Description and Replication (TIDieR) checklist28.

Table 2 TIDieR Checklist.Outcome measures

All study outcomes were measured at baseline and at 1-, 3- and 6 months post intervention as in the questionnaire (see Supplementary Information: Questionnaire). We initially intended to follow up the participants over 12 months but had to stop data collection at 6 months to comply with restrictions during the COVID-19 pandemic.

Asthma control was the primary outcome and measured using the validated Global Initiative for Asthma (GINA) Asthma Symptoms Control9. This questionnaire comprises four questions that measure the adequacy of asthma treatment in the past four weeks. The questions focus on the day and night-time symptoms, use of reliever, and limitation of activity due to asthma. The option for each response is either 'Yes' or 'No'. Well-controlled was considered if the responses to all questions were 'No'. Any responses of 'Yes' were considered as not controlled.

The secondary outcomes measured in this study all related to the previous 1 month:

Number of times reliever medication (inhaled or oral bronchodilators) was used

Adherence to controller medication

Frequency of acute exacerbations (defined as episodes characterised by acute or subacute onset of progressively worsening symptoms, such as shortness of breath, cough, wheezing or chest tightness, which are worse than the patient's usual status and require a change in treatment)

Frequency of asthma-related emergency visits (to a health clinic and/or hospital emergency department)

Frequency of asthma-related admissions

Numbers of days lost from work for asthma treatment (defined as the number of days of medical leave taken by an employee, or unable to work if self-employed)

Number of times the participants reported using their pictorial asthma action plan in the previous month.

Data collection

Data were collected face-to-face using a pretested structured questionnaire in English or Malay language. At baseline, there were four sections to the questionnaire:

Section 1: Socio-demographic and socio-economic information, including age, gender, ethnicity, highest education level, occupation, marital status, personal and household incomes.

Section 2: Medical and healthcare information, including duration of asthma, triggers and allergies, frequency of attacks, use of healthcare resources, medications, vaccinations, current and history of alternative treatment use, smoking status, co-morbid conditions, previous asthma education and ownership/use of an asthma action plan, use of an asthma diary.

Section 3: Asthma control assessment using the GINA Asthma Symptom Control.

Section 4: Health literacy was measured using the validated 47-item Asian version of the Health Literacy Survey-Asia-Q47 (HLS-ASIA-Q47) which assesses the ability to access, understand, appraise, and use health information in the context of healthcare, disease prevention, and health promotion29. The HLS-Asia-Q47 has been shown to be valid and reliable for use in Malaysia30. It was rated on 4-Likert scale, ranged from 1 = very difficult to 4 = very easy. According to the instructions with the HLS-ASIA-Q47, an index of health literacy score was constructed using the mean-based scores of the 47 items. These were transformed into a unified metric ranging from 0 to 50 using the formula = (mean – 1)* (50/3)31. The index scores were grouped into two categories: limited and adequate health literacy. An index score of ≤33 indicates limited health literacy31.

Information on healthcare visits (emergency visits at the clinic for attacks) were verified by clinic doctors from the participants’ medical records. In case of any discrepancies, the information was checked with the patients, as patients in Malaysia might seek care from other health providers, and the medical record may not be complete.

Follow-up data on all the primary and secondary outcomes were collected at 1-, 3- and 6-month post intervention by trained enumerators who were medical doctors not involved in patients’ recruitment and baseline assessments. At every follow-up visit, primary and secondary outcomes were collected, and participants asked about reasons for using a pictorial action plan and any barriers and facilitators.

Data analysis

We used IBM SPSS version 26.032 and R software version 4.0.433,34,35, for the statistical analysis. Descriptive analysis of the baseline variables was reported using means and standard deviations for continuous variables and frequencies and percentages for categorical data. Chi-squared or Fisher’s Exact tests (for small numbers) for categorical variables and independent samples t test for continuous data were used to compare the difference in baseline characteristics between the participants who had completed, withdrawn or lost to follow-up.

The primary and secondary outcomes were categorised as binary data. We calculated the difference in paired percentages with well-controlled asthma, no reliever use, at least one missed day using controller medication, at least one acute exacerbation, at least one emergency visit, and at least one admission, for each of the follow-up time points compared to baseline. The analysis was completed using all data available with no imputation made for missing data (i.e., missing data were left as missing). The 'modified Wilson score method' or 'Newcombe score method' was used to calculate the Exact 95% confidence intervals for all paired differences.

Feasibility of assessing the cost of asthma-related care

An expert panel comprising a Ministry of Health (MoH) family medicine specialist and a pharmacist, and the research team who were family medicine specialists and a respiratory physician, reached consensus on the annual cost of care for a person with well-controlled and uncontrolled asthma. The unit costs of specialist and general outpatient visits were obtained from the legislated fee schedules for the Ministry of Health, Malaysia services which reflect the actual cost of services36. The fee schedule details fees for MoH facilities for non-citizens who were not eligible for subsidised healthcare in Malaysia. Thus, the fees are the estimated cost of care in public health facilities in the country. We estimated the cost savings over six months for the participants who completed the study. This estimation was based on the differences between the estimated cost incurred in the absence of the intervention and the actual costs as observed. However, resource use had only been captured for 3 months out of the 6-months follow-up (for months 1, 3 and 6 during the follow-up at 1 month, 3 months and 6 months). Therefore, in order to estimate medication costs, it was assumed that (a) asthma status at baseline remained throughout month 1; (b) asthma status at 1-month follow-up remained for months 2 and 3; and (c) asthma status at the 3-month follow-up remained for months 4, 5 and 6. The details are discussed in the Supplementary Information: Cost of asthma-related care and Supplementary Table 1.

Ethics approval

Regulatory approvals have been obtained in line with the operating procedures of the RESPIRE Global Unit, including approvals from the National Medical Research Ethics Committee, Ministry of Health, Malaysia [NMRR-18-2683-43494] and relevant authorities involved in the Klang District. Both verbal and written informed consent were obtained from eligible participants before the involvement of this study. Confidentiality of the participants was ensured; data were anonymised before publication or report writing. The study was conducted in accordance with the principles of the International Conference on Harmonisation Tripartite Guideline for Good Clinical Practice. This study also received sponsorship approval from the Academic and Clinical Central Office for Research & Development (ACCORD) at the University of Edinburgh.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.