This is a randomized clinical trial that was conducted from Nov 2015 until March 2016 in two groups of intervention and control and registered with code IRCT2015070214333n38 in the Iranian clinical trial site.
In this study, all patients with hemiplegic stroke who were admitted to the neurology unit of Farabi Hospital in Kermanshah agreed in participating in the study were eligible to include in the work. Referring to the neurology unit of the center, the researchers selected subjects from eligible samples and as available and divided them randomly through the flip coin method in two intervention and control groups.
The minimum sample size required in this study was based on the formula for comparing the mean of a quantitative trait in the two groups provided in the following, and it was calculated considering confidence interval of 95% (α-1) and probability of 90% (β-1) using the results of various studies [15], and the minimum sample size required for each intervention and control group was 9. However, in terms of percentage as a drop, as well as increasing the test power, the researchers considered the minimum sample size of 20 for each group and a total of 40 people.
Questions for the caregiver include: Living with a hemiplegia patient with stroke, lack of mental disorders, ability to do care services, non-use of drugs and psychotropic, having at least a third-grade high school certificate and receiving previous training in this field. Exclusion criteria include the decision to discontinue the collaboration at any moment, early discharge, or the death of the patient.
Data Collection
Sampling was performed among patients with hemiplegic stroke admitted to the neurology unit of Farabi hospital in Kermanshah city. After obtaining written consent from the patients and their carers, the patients were randomly assigned to the two intervention and control groups according to the flip coin method. Through questionnaires, the demographic information of both groups was gathered including age, sex, marital status, educational level, occupation, family relationships, economic status of the patient's family, place of residence in the hospital, furthermore, the health status of both groups was assessed through completing the SQ2/0 questionnaires. In the intervention group, based on the educational needs of hemiplegic patients and based on the valid sources (1, 8, 16-18), the training program of these patients was designed with family-based attitude by the researcher and the research team's collaborators, and to obtain the validity the valuable comments of 3 neurologists, 4 faculty members, and 3 neurologist nurses were used. The training included information on the disease, symptoms, complications, influencing factors, nutrition, position change, and preventing the formation of ulcer, control of bowel and bladder function and mobility in the form of a care program that lasted four consecutive days, each day for 50 to 60 min. The plan was provided for the patient’s carer during the hospitalization period. In some cases, training sessions incremented from 6 to 8 sessions. After training the care program, the carers were asked to perform the care activities in the hospital according to the program, and their actions were controlled by the researcher using the checklist (this checklist also validated using the mentioned experts’ comments). The carers were trained comprehensively and evaluated by a person. After assuring the carers' empowerment (more than 95%), they were asked to provide a care program for the patient for a month at home, and the training pamphlets and contact numbers were provided for the caretakers for being responded within the 24 hours a day. During this one-month, the caregivers and patients were contacted once 4 or 5 days during and the continuity of care was ensured. One month after the intervention, the control and intervention groups were coordinated with the research unit to refer to the patient training unit in the hospital to re-complete the health status questionnaire. For the people unable to attend, a researcher assistant was referred to the patients' home for completing the questionnaire. All of questionnaires were collected by a researcher assistant that he was not the research team.
Data collection tools
The data were collected using a demographic data collection form for the patient and the caregiver and the health status questionnaire.
Health Questionnaire, SQ2/ 0, of the patients consisted of 37 health status questions (19) in three aspects: physical, mental, and social, including subscales of General Health (2 questions), Physical Functioning (10 questions), Role Limitations Attributed to: Physical Health (4 questions) , Role Limitations Attributed to: Emotional Problems (3 questions), Bodily Pain (2 questions), Energy / Fatigue (4 questions), Mental Health (5 questions), Social Function (2 questions) and Health Perception (5 questions). Some questions were responded with “yes” and “no” and the rest with 3 or 5 options Likert scale. A higher score in each subscale means a better status in that sub-scale [16]. The content validity and reliability of the health status questionnaire were reviewed in the study of Kafami et al.[17] Using the retest and pre and post-correlation coefficient the reliability was obtained and verified for each health status subscale including general health (r = 0.76, α = 0.87), Physical Functioning (r = 0.68, α= 0.79), Role Limitations Attributed to: Physical Health (r = 0.66, α= 0.86), Role Limitations Attributed to: Emotional Problems (r= 0.66, α = 0.86), pain (r= 0.80, α= 0.89), Energy/Fatigue (R= 0.89, α= 0.94), Mental Health (r = 0.92, α= 0.94) and Health Perception (r= 0.76, α= 0.87) (20). To further ensure the validity confirmation, the questionnaire with the objectives of the study was provided to 5 neurologists and 3 faculty members, who all confirmed it. Moreover, for reconfirmation of the reliability of the tool, using the 20 similar samples before initiating the study, the alpha Cronbach coefficient was re-calculated as 0.885.
Data analysis
SPSS V.21 was used to analyze the data. The Kolmogorov Smirnov test was used to determine the normal distribution of the data, and appropriate parametric tests or their nonparametric equivalents were used based on the distribution of data. The significance level for all tests was 0.05.