Physical activity and health-related quality of life among adults living in Jeddah city Saudi Arabia

Study setting, study population and sampling technique

This cross-sectional interview-based study was conducted in three shopping malls, three walking tracks and fitness centres/gyms and three governmental hospitals in Jeddah, Kingdom of Saudi Arabia. Multistage stratified random sampling technique was used to identify these shopping malls, walking tracks and fitness centres/gyms and hospitals. The study duration was 6 months, from March to September 2022.

Jeddah is divided into five administrative units, each having a tertiary care hospital, malls and many gyms. Three administrative units were initially selected using a lottery method. These three areas exhibit significant socioeconomic differences in their population. From each selected administrative unit, one hospital, one gym or a walking track and one shopping mall were selected using a simple random selection approach. Individuals coming to each selected shopping mall and visitors of admitted patients coming to the hospital who were willing to participate were selected through systematic sampling with a random start approach (every 3^rd person). Similarly, regular gymgoers (attending the gym at least twice a week for the last 2 months) were randomly selected. A minimum duration of 2 months was considered since many gyms offer a month of the free trial, and people usually do not follow up.

The sample size was calculated using the online Epitools sample size calculator. Based on a study done in Al Khobar (Al-Shehri et al., 2008), Saudi Arabia, the mean physical component score was 41.3, with a standard deviation of 6. Using these figures, keeping the significance level at 95%, the desired precision at 0.5; the calculated sample size was 546. We increased the sample by 25% to 685 for any dropouts/incomplete forms.

Study tool, data collection & scoring

After collecting the demographic profile data, a widely used, reliable and validated instrument, 36-Item Short Form Survey (SF-36), was used to gather the data in this survey. A questionnaire translated into Arabic was adapted from a study conducted in the Jazan region (Sheikh et al., 2015) for this study. The general practice physical activity questionnaire (GPPAQ) was used to assess the physical activity level.

Before the start of data collection, 1-day training on the use of SF-36 and GPPAQ was carried out. A four-member team was trained for the data collection, including a doctor, a female nurse and two volunteers (a male and a female) from the Research and Studies department, Ministry of Health (MoH). Keeping in view the cultural norms, only female team members collected data from gyms at the ladies-only time and from visitors in the female wards.

Visitors were approached and explained the purpose of this campaign. Details on the demographics and the SF-36 questionnaire items were collected from the participants. The participant’s physical activity level was assessed using the electronic version of the GPPAQ with a tablet provided by MoH (available at https://www.gov.uk/government/publications/general-practice-physical-activity-questionnaire-gppaq), and the physical activity index was calculated electronically. The physical activity index categorises the participants into active, moderately active, moderately inactive, and inactive (Table S1). Each interview took around 25–30 min, which was communicated and agreed upon in advance. At the end of the interview, a 5–10 min health education session was given regarding the importance of physical activity and a healthy diet using Information, Education and Communication (IEC) material approved and provided by MoH.

The SF-36 is a brief questionnaire that generates scores across eight dimensions of health. The items include; physical functioning (10 questions); physical role limitations (four questions); bodily pain (two questions); general health perceptions (five questions); energy/vitality (four questions); social functioning (two questions); emotional role limitations (three questions) and mental health (five questions). Patients were asked to rate their responses using 2-, 3-, 5-, or 6-point polytomous response options. For each dimension, a scoring algorithm was used to convert the raw scores item to a scale ranging from zero (worst health) to 100 (best health). Additionally, the SF-36 domain scores were summed together to calculate the two-component summary score; physical component summary (PCS) and mental component summary (MCS).

Data on the current smoking status (cigarette and/or sheesha) was taken. Current smokers were defined as having smoked twice or more in the last week (Abdulrashid et al., 2018). Participants were identified to have a comorbid condition if they had a history of diabetes, hypertension, cardiovascular diseases, renal diseases, joint/back pains or any diagnosed autoimmune disease or if they were currently on any medication. Considering it a sensitive issue and because of cultural values, the history of any other form of addiction or drug dependence was not asked.

Outcome variables

The primary outcome of interest was the health-related quality of life (HRQoL) and the effect of physical activity on the HRQoL. HRQoL was inferred by calculating means for the eight domains and two-component summary score of SF-36. The secondary outcome was to determine factors associated with the PCS and MCS global scores of HRQoL.

Selection criteria

• Inclusion criteria: Shopping mall and hospital visitors and gym/walking track users between 18–70 years old and willing to participate.

• Exclusion criteria: Visitors on wheelchairs or clutches, pregnant females, who had undergone any surgery in the recent past (within the last 2 months), or people with special needs were excluded.

• Individuals identified as having any mental health issues or drug dependency.

Data analysis

Data analysis was done using Stata 16.1 (Stata Corp, College Station, TX, USA). First, we calculated descriptive statistics, including means and standard deviations for continuous variables and frequencies with percentages for categorical variables. We described demographic characteristics among all people by the level of their physical activity. Mean scores for all the SF-36 eight domains and PCS and MCS scores by physical activity levels were calculated according to the scoring guidelines (https://www.rand.org/health-care/surveys_tools/mos/36-item-short-form/scoring.html). Additionally, PCS and MCS mean scores by physical activity levels were calculated. The higher score indicates greater HRQoL (Patel, Donegan & Albert, 2007). To reduce the potential for gyms/ walking tracks individuals to skew the physical activity results, a stratified analysis calculating the mean PCS scores (of all three subgroups) was carried out using ANOVA with a Bonferroni correction test.

Unadjusted and adjusted linear regression analyses with 95% CI were performed to evaluate factors associated with PCS and MCS global scores. In the unadjusted analysis, variables that did not violate the assumption of proportionality were considered for inclusion in the adjusted analysis. Covariates included age groups (18–29, 30–44, 45–60 and 60+), sex, nationality, educational and marital status, history of smoking and comorbid conditions, interview location, and physical activity levels.

Ethics statement

Ethical approval was granted by the Ministry of Health and Directorate of Health Affairs Jeddah (H‑02‑J‑002-A01448) and the administration of shopping malls, hospitals and gym/ fitness centres. Before the interview, verbal consent was taken from the visitors and confidentiality of data was ensured.

Study participants

A total of 693 adults participated in this study, with a mean age of 36 (±11) years. Overall, 59% were females, 88% were Saudi nationals, 62% had education up to a bachelor’s or a diploma, and around 58% were married. Approximately 9% were current smokers, and 6% had a comorbid condition. The majority of the study participants were interviewed in shopping malls (48%), followed by gyms/walking tracks (31%) and hospital visitors (21%) (Table 1).

The number of active or moderately active people was around 28% and 34%, respectively. Compared to the inactive or moderately inactive people, individuals who were active or moderately active were mostly men (70% vs 30%) and were single (68% vs 32%). A higher proportion of active or moderately active people had a bachelor’s or a higher degree (60% vs 40%) and were frequent gym/walking track users (77% vs 23%). The majority of the individuals with any comorbid condition and who had a history of smoking were also inactive (Table 1).

SF-36 domain scores and the component summary scores

SF-36 scores for the eight domains were calculated, showing a higher mean for active and moderately active individuals. Additionally, the PCS and MCS scores were calculated. Compared to inactive, the mean PCS and MCS scores for the active category were 79 vs 65 and 71 vs 57, respectively (Table 2). Additional stratified analysis shows no statistical difference (p = 0.108) in the mean PCS scores across the three subgroups (gym/walking tracks, hospital visitors and mall visitors; 75, 71 and 72, respectively).

Component summary scores (physical and mental components) according to the physical activity levels are shown in Fig. 1. The physical component domain scores were higher in all activity levels (active 79 vs 70; moderately active 74 vs 66; moderately inactive 68 vs 56; and inactive 66 vs 58) than the mental component domain scores (Fig. 1), highlighting the importance of physical activity.

Factors associated with physical component summary (PCS) scores and health-related quality of life

In adjusted analysis, PCS scores were significantly associated with male gender (adjusted β 4.43, 95% CI [1.44–7.41]) and physical activity levels (moderately active- adjusted β 5.73, 95% CI [1.28–10.17] and active- adjusted β 10.11, 95% CI [5.44–14.77]). Compared to the individuals between 18–29 years of age, individuals in age groups 30–44, 45–60 and 60+ had lower PCS scores (adjusted β −4.06, 95% CI [−7.87 to −0.25]; −6.27, 95% CI [−11.50 to −1.03]; and −8.44, 95% CI [−17.59 to 0.71]), respectively. History of smoking and having any comorbid conditions significantly affected PCS scores (adjusted β −5.52, 95% CI [−10.47 to −0.56] and −11.21, 95% CI [−17.39 to −5.03]), respectively. The adjusted R-squared value for the final model was 0.11. Marital status and nationality had no significant association with PCS (Fig. 2 and Table S2).

Factors associated with mental component summary (MCS) scores and health-related quality of life

In adjusted analysis, MCS scores were significantly associated with male gender (adjusted β 6.51, 95% CI [3.40–9.63]) and physical activity levels (moderately active-adjusted β 7.57, 95% CI [2.92–12.21] and active-adjusted β 9.77, 95% CI [4.90–14.64]). Compared to the individuals between 18–29 years of age, individuals in age groups 30–44 had less MCS scores (adjusted β −4.76 95% CI [−8.74 to −0.78]). History of having any comorbid conditions significantly affected MCS scores (adjusted β −13.08, 95% CI [−19.52 to −6.63]). Smoking status was not significantly associated with MCS scores (adjusted β −4.03, 95% CI [−9.20 to 1.13]). The adjusted R-squared value for the final model was 0.12. Marital status and nationality had no significant association with MCS (Fig. 3 and Table S3).