Study design
This cross-sectional research utilized a convenience sample and was conducted from January to June 2022.
Study population
Stroke patients undergoing physiotherapy at four tertiary-level hospitals participated in this study. The research followed the 2013 Helsinki Declaration protocols and obtained ethical clearance from the Bangladesh Open University Ethics Committee (BOU/SST/MDMR/146/17/376).
Sample and sample size estimation
The sampling procedure for the study was done using the following equation: Z2pq/d2. Where Z2 =1.96, p = the expected health status by EQ-VAS (visual analog scale) is 60 % [10], d = 5% margin of error, and the attrition rate is 15 % to avoid bias. Applying the formula yielded a minimum sample size of 424.
Eligibility criteria
Inclusion criteria:
Seeking stroke patients, aged 18-80, with confirmed ischemic or hemorrhagic stroke diagnosed via CT scan or MRI, occurring at least one month ago. Participants should have no preexisting stroke-related disability (per mRS), and willingness to engage in the study is essential.
Exclusion criteria:
Exclusion criteria encompassed global aphasia and/or communication inability, medical instability, concurrent neurological disorders (brain tumor, traumatic brain injury, mental disorder, Parkinson’s disease, dementia), pre-existing musculoskeletal conditions (spinal cord injury, hip fracture, dislocations, contractures), ongoing psychoactive substance abuse, pre- and post-stroke psychiatric morbidity, and terminal illnesses like renal failure or end-stage cancer [11].
Data collection procedures
An interviewer-administered questionnaire was employed to collect data, translated into Bangla and back into English with expert input. A pretest involved sixteen respondents outside the study area in a similar setting. The interview was conducted by four trained data collectors (physiotherapy graduates with >3 years of experience). Additionally, clinical information, including type of stroke, side of weakness, duration of a stroke, and co-morbidity, was extracted from the patient's medical records.
Data collection tools
The questionnaire encompasses socio-demographic inquiries pertaining to age, sex, residence, educational status, marital status, occupation, income, smoking, tobacco habits, cohabitant situation, and care provider. Clinical profile included the type of stroke, side of weakness, duration of a stroke, co-morbidity, the person responsible for medical expenses, receiving rehabilitation care, and assistive device use.
EuroQL-5: This study utilized a modified and translated Bengali iteration of the EQ-5D-5L health-related quality of life questionnaire. The instrument comprises two integral components: the EQ-5D descriptive version and the EQ visual analog scale (EQ-VAS). The EQ-5D-5L description system encompasses five dimensions: mobility, self-care, usual activities, pain or discomfort, and anxiety or depression. Participants were instructed to evaluate their present health status using a 5-point Likert scale, wherein a score of 1 denoted the absence of problems, and a score of 5 signified significant problems. Respondents were also asked to assess "their health today" on a 0-100 vertical-thermometer-like scale (worst to best score) using EQ-VAS [10,13,14]. In our study, we computed the EQ summary index by assigning numerical values (ranging from 1 to 5) to denote the five levels of health across each of the five dimensions measured by EQ-5D-5L. These health states were then transformed into a consolidated measure known as utility, represented on a scale from -1 to 1. Here, 0 signifies death, values less than 0 indicate states worse than death, and 1 represents full health. To derive the EQ index values, we employed the calculation method established by Jyani et al. (14) in India, given the absence of a standard value set for Bangladesh and the geographical similarity between the two neighboring countries.
The Modified Rankin Scale (mRS) is frequently employed to assess disability post-stroke, aligning with the International Classification of Functioning, Disability, and Health (ICF) model encompassing body function, activity, and participation. The scale is a 7-point system ranging from 0 (no symptoms, completely normal) to 5 (severe disability), with 6 denoting death. Notably, the scale exhibits robust test-retest reliability, with a range of k=0.81 to 0.95 [15].
Measurement of variables
Independent variables: Initially, age was considered a continuous variable; however, it was subsequently categorized into three groups: 18–40 years, 41–60 years, and 61–80 years. Other variables were also categorized as follows: gender (male/female), place of residence (urban/rural), marital status (single or widowed/married), level of education (primary/secondary/tertiary), occupation (employed, including service, business, farming, daily labor; unemployed, including homemakers, retired, others), monthly family income (15000 taka/15000-30000 taka/31000-45000 taka/>45000 taka), smoking (No/Yes), tobacco habit (No/Yes), cohabitation situation (spouse & children/others, including parents, siblings, relatives), care providers (spouse & children/others), type of stroke (ischemic/hemorrhagic), duration of stroke (1-3 months/3-6 months/>6 months), side of weakness (right/left/both sides), number of stroke episodes (1st time/recurrent), co-morbidity (No/Yes), receiving rehabilitation care (No/Yes), the medical expenses provided by (self or spouse/children/others), and using assistive devices in daily life (No/Yes).
Dependent variables
The modified Rankin Scale (mRS): This scale is a 7-point scale categorised from 0 = normal, 1 = no significant disability, 2 = slight disability, 3 = moderate disability, 4 = moderately severe disability, and 5 = severe disability.
EuroQL-5: The EQ-5D descriptive version of the EQ-5D-5L (mobility, self-care, usual activities, pain/discomfort, and anxiety/depression) scores 1 = no problem, 2=slight problems, 3 = moderate problems, 4 = severe problems, and 5=extreme problems and the EQ visual analogue scale (EQ-VAS). Participants were asked to evaluate their current health using a vertical-thermometer-like scale ranging from 0 to 100 (from worst to best score). Furthermore, we calculated the EQ-5D summary index, with values between - 0.923 and 1 (0 indicating death and 1 indicating full health). A score below 0 is considered a health status worse than death.
Data Analysis
The data was analysed using SPSS-25 software to perform descriptive and inferential statistical analyses. Continuous variables were characterized using means and standard deviations (SD), while categorical variables were represented using frequencies and percentages. Univariate analysis employed the chi-square test and Fisher exact test to explore potential associations among socio-demographic status, clinical profiles, and EQ-5D dimensions. Consistent with previous research, it was noted that the EQ-5D instrument employs a pair of categorical variables for each domain. [10]. These variables were characterized by two discrete states: (a) absence of issues and (b) presence of problems categorized into degrees of severity (slight, moderate, severe, and extreme). The EQ summary index was subjected to statistical analyses, employing the Independent Samples T test for two groups and the one-way Analysis of Variance (ANOVA) test for more than two groups. Multiple binary Logistic regression was used to ascertain the estimated risk factor for Health-Related Quality of Life (HRQOL) among stroke survivors. This involved incorporating independent variables with a p-value <0.05 from the univariate analysis and using Adjusted Odds Ratios (AOR) and a 95% confidence interval (CI). The fitness of model for each EQ dimension was check by Hosmer-Lemshow’s test and classification table. The variance inflation factors (VIF) was used to test for multicollinearity amongst the independents variables (VIF= <5.0) [16]. The researchers established a significance level of less than 0.05.