Study design and period
A descriptive cross-sectional study was carried out between August 9 and September 14, 2018. A survey which included all medical students who were enrolled at Orotta School of Medicine (current name - Orotta College of Medicine and Health Sciences) was undertaken.
Study population and setting
A total of 201 medical students, in their 2nd, 3rd, 4th, 5th and 6th academic years, were invited to participate voluntarily and anonymously at the selected OSM. However, students who met the following inclusion criteria: medical students between 2nd and 6th year of study, who stayed in the college for more than six months, and had not been diagnosed with critical illness in the previous 4 weeks were 192.
Data collection tool and techniques
The three main components of the data collection tool used in the survey were socio-demographic characteristics, sub-optimal health measurement scale version 1.0 (SHMS V1.0), and health promoting life style II (HPLP-II) scale. Time to fill the questionnaires through self-administration ranged from 20 to 30 minutes.
Socio-demographic characteristics
Self-administered questionnaire was disseminated to obtain data on participants socio-demographic information including; age, gender, level of education, smoking habits, alcohol consumption, financial status, length of stay in college, and body mass index (BMI) (kg/m2).
Self-rated health status (SRH)
The second part was evaluation of SRH, which was performed according to the clinical guidelines for SRH published by the China association of Chinese medicine. To measure SRH, we used the Sub-Optimal Health Measurement Scale Version 1.0 (SHMS V1.0), which is a multidimensional and self-report symptom inventory [3]. SHMS V1.0 comprises 39 items with 3 dimensions: physiological health (14 items) includes physical condition (3 items), organ function (6 items), body movement function (3 items) and vigor (2 items); psychological health (12 items) comprises positive emotion (4 items), psychological symptoms (6 items) and cognitive function (2 items); and social health (9 items) contains social adjustment (4 items), social resources (3 items) and social support (2 items), and 4 other items for health status evaluation, in which participants were asked: "What is your general feeling in terms of physical/psychological/ social/general health?" The 35 items of five-point Likert-type (1= never, 2 =occasionally, 3 = sometimes, 4 = often, 5= routinely) were used to measure the participants self-reported health problems [5, 6].
The original score of every factor was equivalent to the total score of items included in this factor, and the original score of every dimension was equivalent to the total score of factors included. Subsequently, the original raw score was converted to obtain the final score accordingly. Ultimately, the converted scores were used to analyze the outcomes-that is, the total scores for each SHMS V1.0 domains were transformed into a range of 0 (worst possible health status measured by the questionnaire) to 100 (best possible health status), with a highest score representing better SRH [2].
Health-promoting lifestyle (HPL)
The health-promoting lifestyle profile (HPLP) was initially developed by Walker, Sechrist, and Pender in 1987 and later revised as Health Promoting Lifestyle Profile-II (HPLP-II) in 1997 [9]. HPLP-II containing 52-items was used to evaluate the students’ HPL. It consists of 6 dimensions namely: health responsibility (9 items), spiritual growth (9 items), physical activity (8 items), nutrition (9 items), interpersonal relationships (9 items), and stress management (8 items). This measuring instrument can be used to assess the frequency of health-promoting behavior using a self-reporting. To determine the frequency of each behavior, a 4-point Likert scale (1 = ‘‘never’’, 2 = ‘‘sometimes’’, 3 = ‘‘often’’, and 4 =‘‘routinely’’) was used, in which all items of HPLP-II are affirmative, with no reverse questions [5, 6].
The score for overall HPL is obtained by calculating the mean of the individual's responses to all 52 items; six subscale scores are obtained similarly by calculating a mean of the responses to subscale items. The use of means rather than sums of scale items is recommended to retain the 1 to 4 metric of item responses and to allow meaningful comparisons of scores across subscales. Hence, the final total score was obtained by adding the scores for all the items and dividing by the total number of items [5, 6].To determine the levels of HPLP-II subscales, the scores were divided by quartiles into low (1.00–2.38), moderate (2.39–2.61), good (2.62-2.91) and excellent (2.92–4.00). High scores indicated a greater frequency of health-promoting behaviors. A mean of 2.50 was considered to be a positive response, in line with previous specifications [6].
Validity and reliability
We used a previously validated English version of SHMS V1.0 and HPLP-II questionnaires. The composite Cronbach’s alphas for SRH (after converting the negatively worded items) and HPL were 0.819 and 0.874 respectively. Moreover, Cronbach’s alphas for physiological health, psychological health, and social health were 0.657, 0.669, and 0.785 respectively. On the other hand, Cronbach’s alphas for health responsibility, spiritual growth, physical activity, nutrition, interpersonal relationship, and stress management were 0.769, 0.772, 0.572, 0.678, 0.788, and 0.629 respectively.
Pre-test
A pre-test was conducted on 20 students at Asmara College of Health Sciences in August 2018, to evaluate the clarity, applicability and understandability of the tools, to estimate the time needed to complete the questionnaire, and to get comments with regard to the number of questions. The questionnaire was distributed by data collectors who were fluent in English. The pre-designed questions which were unclear or ambiguous were subsequently simplified by the investigators. Finally, necessary modifications were undertaken after the pre-test.
Data analysis
SPSS version 22 was the statistical program used to analyze the data. Descriptive analyses for the demographic data was done using frequency (percent), and median (IQR). Tables were used to provide an overall and comprehensible presentation and description of data. The internal consistency of both SRH and HPL at subscale level and composite were assessed using Cronbach’s alphas. An independent samples t-test, ANOVA, Chi-square/Fishers exact test for independence were used to compare mean differences and proportions between the groups, as appropriate. Bonferroni post-hoc test was also performed for variables that were found to be significant in ANOVA. The correlation of SRH/HPL with socio-cultural factors were tested using Pearson’s coefficient correlation. Binary multivariate logistic regression was used to explore the association between categories of SRH and HPL among medical college students. A p value of less than or equal to 0.05 was considered as statistically significant.