Quantifying the benefits of healthy lifestyle behaviors and emotional expressivity in lowering the risk of COVID-19 infection: a national survey of Chinese population | BMC Public Health

Quantifying the benefits of healthy lifestyle behaviors and emotional expressivity in lowering the risk of COVID-19 infection: a national survey of Chinese population | BMC Public Health

Examine setting, members and treatment

From June 29, 2022 to July 2, 2022, we recruited on the web volunteers and made use of a self-made questionnaire to carry out a preliminary study. Then we conducted a dynamic research of grownups (aged ≥ 18 years outdated) in four parts of Jiangsu province (jap region), Henan province (central area), Heilongjiang province (northeast location), and Qinghai province (western location) of China were being selected to variety a consultant sample utilizing stratified random sampling. Subsequently, inside every single area, additional than two rural areas and two urban places from each and every metropolis had been selected randomly. From August. 3, 2022 to August. 14, 2022, baseline study was carried out in the aforementioned 4 regions, followed by a subsequent follow-up with the baseline inhabitants in from February. 1, 2023 to February. 18, 2023. This timeline authorized for the collection of detailed data to examine the evolution of things below investigation more than time.

For the duration of the baseline survey carried out in July 2022, information was gathered from 8002 individuals. Soon after making use of exclusion conditions, info from 6781 contributors was finally involved in the recent review. In January 2023, China commenced to put into action the “Class B epidemic and B Management” coverage. Then, in February 2023, a adhere to-up study was conducted, in which we followed up 5780 members in the review populace primarily based on the baseline survey, with a abide by-up fee of 85.24%. We then excluded four adults who experienced been infected with the COVID-19 virus in the baseline survey, and the remaining 5776 participants have been provided in our analysis and review (Fig. 1). We make assurance that there have not transpired any variations in the violence and variety of virus throughout abide by up period. The research was accepted by the Lifestyle Science Ethics Evaluation Committee of the Life Sciences Ethics Review Committee of Zhengzhou College, and contributors signed informed consent varieties on the questionnaire.

Fig. 1
figure 1

Inclusion and exclusion of participants

Evaluation

In the course of the baseline study, we investigated all legitimate data about contributors ahead of an infection, and through the abide by-up survey, we continued to investigate the stick to-up information and facts of these contributors, the questionnaire for some unbiased variables measurements is found in Added file 1: Table S1.

Sociodemographic covariates

The research provided 7 covariates: gender (classified as guys or ladies), age(divided into 5 teams: 18–29, 30–39, 40–49, 50–59, and ≥ 60), location of home (categorized as Jiangsu, Henan, Heilongjiang, or Qinghai province primarily based on populace density data from the seventh population census of China, Extra file 1: Fig. S1), marital standing (categorized as married or other), training degree (classified into junior and down below, senior, undergraduate, graduate and over), existence of long-term condition (yes/no), and background of allergic (classified as certainly, no, or not very clear).

Lifestyle behaviors evaluation

The research assessed 4 possible life-style behavior aspects dependent on definitions and requirements derived from the American Higher education of Lifestyle Drugs (ACLM). These factors incorporated using tobacco (classified as smoke, stop, or never. We outlined the smoking cigarettes as “refers to private use of cigarettes or other tobacco merchandise, such as cigars, pipes, waterpipes, and so on.”), ingesting (categorized as consume, give up, or under no circumstances. We described the consuming as “personal consumption of any beverage containing extra than .5% alcohol, which includes beer, wine, spirits, etc.”), physical workout (classified as in no way, < 1 time, 1–2 times, 3–5 times, or ≥ 6 times per week. We defined the physical exercise as “a sport, exercise, or athletic activity that requires physical exertion and energy expenditure.”), and adherence to preventive measures such as wearing masks, washing hands, and keeping distance (categorized as never, seldom, sometimes, often, or always. We defined the wearing masks, washing hands, and keeping distance as adherence to preventive measures) (Additional file 1: Table S2).

To effectively reflect the lifestyle of different groups, we calculated the lifestyle behaviors score using the simple way to sum scores, we assigned values to each lifestyle option based on the criteria established for the study [25]. These values ranged from negatively to positively, or from less frequency to higher frequency, reflecting the extent to which each lifestyle choice aligned with the desired criteria. We allocated numeric values to denote the various smoking statuses (smoke = 1 point, quit = 2 points, never = 3 points), scoring ranged 1–3 points. Similarly, drinking was assigned as (drink = 1 point, quit = 2 points, never = 3 points), scoring ranged 1–3 points. Moreover, physical exercise was assigned as (never = 1 point, < 1 time = 2 points, 1–2 times = 3 points, 3–5 times = 4 points, ≥ 6 times = 5 points), scoring ranged 1–5 points. Additionally, wearing masks, washing hands, and keeping distance were scored accordingly (never = 1 point, seldom = 2 points, sometimes = 3 points, often = 4 points, always = 5 points), scoring ranged 1–5 points. With higher scores indicating healthier lifestyle behaviors.

To conduct the attributable risk (AR) analysis, we classified lifestyles as either health or unhealthy based on the mean value of the lifestyle behaviors scores. Based on the mean scores obtained, we categorized the smoking status options. Since the mean score for smoking status was 2.51, we defined a mean score of 2.51 as whether or not a criterion for whether or not a healthy smoking status is, we classified option 3 (never) as healthy lifestyle choice, while option 1 (smoke) and 2 (quit) were classified as unhealthy. Similarly, for drinking status, with a mean score of 2.44, we defined a mean score of 2.44 as whether or not a criterion for whether or not a healthy drinking status is, we categorized option 3 (never) as a health choice, while option 1 (drink) and 2 (quit) were considered unhealthy. In terms of physical exercise, where the mean score was 3.40, we defined a mean score of 3.40 as whether or not a criterion for whether or not healthy physical exercise is, options 4 (3–5 times) and 5 (≥ 6 times) were classified as healthy choices, whereas options 1 (never), 2 (< 1time), and 3 (1–2 times) were classified as unhealthy. Lastly, for wearing masks, washing hands, and keeping distance, the mean score was 4.15. we defined a mean score of 4.15 as whether or not a criterion for whether or not healthy preventive measures are. Here, option 5 (always) was deemed a healthy choice, while options 1 (never), 2 (seldom), 3 (sometimes), and 4 (often) were classified as unhealthy. The AR formula is as following:

$$rmAR,rm(Attribute,rmRisk),rm = fracIe – IoIe*100%$$

Notes: Ie indicates the incidence of the exposed group and Io indicates the incidence of the non-exposed group.

Emotional expressivity assessment

We evaluated emotional expressivity through three emotional factors, namely anxiety, depression and stress. The assessment of these scores utilized the internationally validated emotional self-assessed “DASS-21” scale (Depression, Anxiety and Stress Scale, a maturity scale validated in China [26, 27], Additional file 1: Table S3). Anxiety was measured through seven questions, including items: “2. I was aware of dryness of my mouth”, “4. I experienced breathing difficulty (e.g., excessively rapid breathing, breathlessness in the absence of physical exertion)”, “7. I experience trembling (e.g., in the hands)”, “9. I was worried about situations in which I might panic and make a fool of myself)”, “15. I felt I was close to panic”, “19. I was aware of the action of my heart in the absence of physical exertion (eg, sense of heart rate increase, heart missing a beat)”, “20. I felt scared without any good reason”. Depression was assessed using seven questions with items: “3. I couldn’t seem to experience any positive feeling at all”, “5. I found it difficult to work up the initiative to do things”, “10. I felt I had nothing to look forward to”, “13. I felt down-heart and blue”, “16. I was unable to become enthusiastic about anything”, “17. I felt I wasn’t worth much as a person”, “21. I felt that life was meaningless”. Stress level was evaluated through seven questions: “1. I found hard to wind down”, “6. I tend to over-react to situations”, “8. I felt that I was using a lot of nervous energy”, “11. I found myself getting agitated”, “12. I found it difficult to relax”, “14. I was intolerant of anything that kept me from getting on with what I was doing”, “18. I felt that I was rather touchy” [28, 29]. We tested the DASS-21 scale for reliability and validity. The results showed that the Cronbach’s alpha coefficient was 0.955, indicating that the scale reliability was reliable.

For the calculation of the emotional score, we assigned values (no = 0 points, sometimes = 1 point, often = 2 points, always = 3 points) for the 4 options (no, sometimes, often, always) of 21 questions according to the emotional evaluation criteria in DASS-21. According to international standards, the total score for anxiety, depression and stress was multiplied by 2 to get the final score. The total score for anxiety, depression, and stress ranged from 0 to 42, with higher scores indicating higher levels of anxiety, depression, and stress [30].

For the purpose of conducting AR analysis, we established cut-off values for anxiety, depression, and stress based on internationally recognized standards. According to these standards, a score of ≤ 7 in anxiety was categorized as normal, while a score of ≥ 8 was considered abnormal. For depression, a score of ≤ 9 was classified as normal, while a score of ≥ 10 was deemed abnormal. Similarly, in the case of stress, a score of ≤ 14 was considered normal, while a score of ≥ 15 was classified as abnormal [31].

Interaction of lifestyle behaviors and emotional factors

To capture the combined effects of lifestyle behaviors and emotional factors, we employed multiplication interaction by multiplying the respective factors. In the current study, we investigated four lifestyle behavior factors and three emotional expressivity factors. Based on the correlation observed between these factors, we identified a total of eight interactions. These interactions included the interaction between drinking status and stress, the interaction between physical exercise and anxiety, the interaction between physical exercise and depression, the interaction between physical exercise and stress, the interaction between wearing masks, washing hands, and keeping distance and anxiety, the interaction between wearing masks, washing hands, and keeping distance and depression, and the interaction between wearing masks, washing hands, and keeping distance and stress.

Dependent variable

We used the question in the questionnaire “Did you get infected with the COVID-19 virus?“ The option 1 (yes, confirmed cases by doctors in medical institutions), option 2 (yes, abnormal nucleic acid testing, antigen testing, virus culture isolation or serological test results), option 3 (presence of clinical symptoms of COVID-19 infection), option 4 (abnormal clinical test results: such as chest X-ray, chest CT test, lung function, blood oxygen saturation, blood routine, etc.) were combined into 1 = had infected COVID-19 virus while the option 5 (no) and option 6 (not clear)were combined into 0 = no. If “had infected COVID-19 virus” was selected, participants were asked which symptoms they had experienced.

Statistical analysis

We used the chi-square (χ2) test to analyze the difference in COVID-19 infection rates of residents with different characteristics, used the mean ± standard deviation to describe the lifestyle behaviors of Chinese residents prior to infection, used the median (upper and lower quartiles) to describe the emotional status of Chinese residents prior to infection. Considering that the infection rate of the population may have regional clustering, we gave priority to using the multi-level statistical model (also known as the random-effects model), and first fit the two-level empty model, taking the region (province) as the high level. The intra-class correlation (ICC) was used to determine whether the variation of the data was clustered in the high level (province). The test result showed that ICC was 0.0322046, indicating that only 3.22046% of the variation of dependent variables was caused by the high level (Province), much less than the general level of 0.1 (10%), indicating that the degree of variation in the dependent variable (infection) is low and the aggregation is low, which may not fit the multilevel model analysis, then we used multi-factor regression analysis. Binary logistic regression was used to explore the influencing factors of COVID-19 infection condition while correlation analysis was used to assess the correlation between lifestyle behavior factors and emotional expressivity factors. We used correlation analysis to analyze the correlation between lifestyle and emotional expressivity, then we used multiplicative interactions to calculate interaction terms, and finally we put the interaction terms as dependent variables into the regression model to explore the magnitude of the interaction effect [32,33,34]. A collinearity test using the variance inflation factor (VIF) (< 6, less than the cut-off value of VIF “10”) was used to determine the correlation between independent variables. No collinearity was detected between these covariates. To assess the impact of unhealthy lifestyle behavior factors and abnormal emotional factors on COVID-19 infections, we estimated the attributable risk (AR) analysis that enables us to estimate the proportion of infections that could be attributed to these factors. Additionally, sensitivity analysis was performed to ensure the stability and robustness of the results whereas residents with history of allergic were excluded to ensure consistency of the findings. All analyses were performed using SPSS 27 and STATA 17 software, with P < 0.05 considered as statistically significant.