Evaluation of the association of physical activity levels with self-perceived health, depression, and anxiety in Spanish individuals with high cholesterol levels: a retrospective cross-sectional study

Theorical framework: cholesterol and mental disorders

Although less discussed in the literature, total cholesterol levels (Includes both types: low-density lipoprotein (LDL) cholesterol and high-density lipoprotein (HDL) cholesterol) are also associated with mental disorders or mental health problems (Pereira, 2017; Troisi, 2009). Anxiety and depression are the most prevalent mental disorders in the general population (Kessler et al., 2009). The study by Han (2022) reports an odd ratio of presenting depressive symptoms of 2.28 (1.07–4.86) in the high cholesterol group with respect to the group with normal cholesterol values. In addition, in people without psychiatric pathology, an association has been found between elevated cholesterol levels and symptoms of anxiety or mild depressive states (Rafter, 2001; Cheon, 2023). In the same way, high cholesterol levels have also been reported in female patients with panic disorders (Bajwa et al., 1992; You et al., 2023). In contrast, some studies suggest that naturally low concentrations of lipids and lipoproteins are associated with depression as well as poor psychological health (Rafter, 2001; Suarez, 1999). One hypothesis would be that low cholesterol levels influence depression as they may reduce the availability of serotonin, making the patient more susceptible to depression (Rafter, 2001; Papakostas et al., 2004). In this respect, a recent meta-analysis revealed that elevated triglyceride levels and reduced HDL levels were associated with the first episode of a major depressive disorder (Wei et al., 2020). Contrary to these findings, other studies report that LDL levels is also associated with depression (Tedders et al., 2011; Andruškiene et al., 2013). Regarding the inconsistencies of the results in the literature between cholesterol levels and depression, it could be that a modifying factor could be sex, making the relationship between cholesterol and depression different in men and women, but it is not yet clear (Tedders et al., 2011). For example, a study performed with an American population found that low HDL levels were associated with higher levels of major depression in women but not in men, while in men, low and high LDL levels and total cholesterol were associated with major depression (Tedders et al., 2011). Although there is evidence of an association between cholesterol and psychological symptoms, a causal relationship cannot be established as cholesterol itself does not directly cause behaviours, it may induce chemical changes affecting the likelihood of certain behavioural outcomes by modulating certain neural pathways (Pereira, 2017). As commented previously, literature also shows that self-rated health status is associated with dyslipidaemia (Nogueira de Sá et al., 2022). In this sense, a study in a Brazilian population found an association between high cholesterol levels and poorer self-rated health status. It is important to consider that self-rated health status as well as symptoms related to anxiety or depression when caring for a person because they are important predictors of mortality and morbidity  (Nogueira de Sá et al., 2022; Larsen et al., 2014).

Physical activity, mental health and cholesterol

On the other hand, another important factor associated with dyslipidaemia is lifestyle. Literature establishes the importance of regular physical activity to prevent and control dyslipidaemia (Tian & Meng, 2019). In addition, evidence suggests that sedentary or physically inactive behaviour is associated with mental disorders such as depression and anxiety in general population (Harris, 2018). In this sense, several studies reported an association of PA with cholesterol levels, for example, higher levels of PA are associated with higher HDL levels (Kokkinos & Fernhall, 1999). Moreover to the apparent physical health benefits, PA also positively affects mental health, being associated with reduced depression and anxiety symptoms and improved cognitive functioning and psychological well-being (Carek, Laibstain & Carek, 2011). A previous study carried out our investigator group found associations between mental health and PA levels in the Spanish population in the pre-pandemic period analysing the data of the Spanish National Health Survey (SNHS 2017) (Denche-Zamorano et al., 2022a). Specifically, Spanish general population groups with higher level of PA had fewer mental disorders evaluate with the Goldberg General Health Questionnaire (GHQ-12) (Denche-Zamorano et al., 2022a). The National Health Survey (NHS) is carried out in Spain by the Ministry of Health, Consumer Affairs and Social Welfare (MSCBS) every 5 years, so the SNHS 2017 is the last health survey before pandemic period. The GHQ-12 has been used in the SNHS to analyse the evolution of the population’s mental health or check the impact of specific situations, such as economic crises or pandemics (Cabrera-León et al., 2017). In other way, a previous study has also reported that mental illness was associated with fewer walking days in a Spanish population with a common disease such as asthma (De-Miguel-Diez et al., 2024). In addition, other study show that patients with asthma who practice PA have less psychological distress (Denche-Zamorano et al., 2022e). However, to the authors’ knowledge there are no data on the association between mental disorders such as depression and anxiety and physical activity levels in a sample of Spanish population with high cholesterol also in the pre-pandemic period in order to be able to compare the impact of the pandemic in the future. The author considerate that improving physical fitness could be a strategy to address the impact of an unhealthy lifestyle on depression and anxiety in Spanish people with high cholesterol (Alam & Rufo, 2019).

According to the literature cited above, the association between cholesterol and depressive or anxiety disorders is complex, probably runs in one direction or the other, the underlying mechanism remains unknown, and causality has not been established. This suggests that more studies are needed to investigate and deepen the links between these variables and how are modulated for other factors like PA.

The results of this study may inform and help health promotion programs to develop more specific protocols aimed at targeting people with hypercholesterolemia as well as their possible related symptoms such as depression or anxiety through physical activity. In addition, the results will provide baseline data on mental health in this population before the COVID-19 pandemic, which will allow the establishment of lines of research to evaluate the effects of the pandemic on mental health in people with high cholesterol.

The current study aimed to analyse associations and differences between different PA levels and mental health, self-perceived health, anxiety, depression and sex in Spanish adults with high cholesterol were assessed. In addition, predictors and risk factors for mental health, anxiety and depression.

Study design and sample

The data used in this study is a second analysis of data from the SNHS 2017 and can be downloaded freely and without restriction of: website of the Spanish Ministry of Health (Ministry of Health, 2017a). A retrospective cross-sectional study was conducted with an adult household resident population (over 15 years of age) living in Spain. in collaboration with the National Institute of Statistics (NIS), collecting health information on the population residing in Spain to ascertain their state of health. For this purpose, the adult questionnaire was used in the SNHS 2017, covering four broad areas: socio-demographic variables, health status, use of health services and health determinants (Ministry of Health, 2017a). A stratified three-phase random sampling system was used to obtain the sample for the SNHS 2017. The surveys were conducted between October 2016 and October 2017. The SNHS 2017 methodology includes all the information on the regulations governing the survey, the objectives, the design of the questionnaire, the sample design (type of sampling, sample size, sample selection, distribution over time, estimators and sampling errors), information collection, treatment of erroneous or missing data, among other important information (Ministerio de Sanidad—Sanidad En Datos—Encuesta Nacional de Salud de España, 2017b).

The SNHS 2017 sample size was 23,089 participants, being the initial sample of 23,089 adults, aged 15 years and older, living in family dwellings in Spain. The following selection criteria were applied to achieve the final sample for current study: being under 70 years of age, reporting high cholesterol, submitting data on depression and anxiety, submitting data on mental health variables, and submitting data on PA variables. Public files with anonymized SNHS 2017 microdata were downloaded.

In order to select participants with high cholesterol, information was extracted from the following survey questions:

1. Do you have, or have you ever had, high cholesterol? (Yes, No, Don’t know or No answer (DK/NA)).

2. Have you suffered from it in the last 12 months? (Yes, No, DK/NA).

3. Has a doctor told you that you have it? (Yes, No, DK/NA).

To be considered as having high cholesterol, a participant had to have answered “Yes” to (a) and “Yes” to (b) or (c). A detailed sample selection procedure is explained in Fig. 1.

Procedures and variables

Data were extracted from the following questions:

Depression status

1. Do you suffer or have you ever suffered from depression? (Yes, No, DK/NA).

2. Have you suffered from it in the last 12 months? (Yes, No, DK/NA).

3. Has a doctor told you that you have it? (Yes, No, DK/NA).

To be considered as having depression, a participant had to have answered “Yes” to (a) and “Yes” to (b) or (c).

Statistical analysis

The distribution followed by the data of the variables of interest was analysed using the Kolgomorov-Smirnov test. A descriptive analysis was performed to assess the gender differences in age and mental health (continuous variables) and the prevalence of SPH, depression status, anxiety status and physical activity level (PAL) (categorical variables). To evaluate sex differences in the variables of interest was using the Mann–Whitney U test for the continuous variables or the Chi-square test with a post-hoc test pairwise z-test for categorical variables with Bonferroni correction. In addition, the Chi-square test was used to evaluate the relationship between PAL and depression status, anxiety status and SPH; as well as the difference in proportions according to the PAL group. Then to interpret the strength of association, Phi coefficient (2x2 contingency tables) or Cramer’s V Coefficient (contingency tables greater than 2x2) were calculated according to the following interpretation: insignificant (0.0–0.10), weak (0.10–0.20), moderate (0.20–0.40), relatively strong (0.40–0.60), strong (0.60–0.80), very strong (0.80–1.00) (Lee, 2016). To test mental health scores according to the PAL with the Kruskal-Wallis test to analyse possible differences in scores at baseline, and the Mann–Whitney U test to analyse differences between the different levels.

Multiple binary logistic regressions were performed, taking the depression status, anxiety status and mental health status respectively, as dependent variables, and sex, age, PAL and social class as independent variables. For all logistic binary regressions, the assumptions of the normality of residuals, the absence of influential factors were tested using Cook’s distance and the absence of outliers. For all analyses, a significance level of less than 0.05 was assumed as statistically significant, and IBM SPSS Statistical v.25 (Armonk, NY, USA) was used.

Relationships between PAL and depression, anxiety and SPH

Self-reported depression was found to be associated with PAL (X² = 78.7, df = 3, p < 0.001, V = 0.157), and the highest and the lowest prevalence of depression was found in “Inactive” and “Active” groups, respectively (28.1% vs 9.6%, p < 0.05) (Table 2). PAL and self-reported anxiety were also found to be associated (X² = 30.6, df = 3, p < 0.001, V = 0.098). The highest and lowest proportions of anxiety were found in the “Inactive” and “Active” groups, respectively (23.9% vs 12.5%, p < 0.05). Similarly, dependency relationships were found between PAL and SPH (X² = 241.3, df = 6, p < 0.001, V = 0.195). The 30.1% of people in the “Inactive” group presented negative SPH, compared to 4.3% in the “Very active” group, p < 0.05. Conversely, 73.3% of people in the “Very active” group presented positive SPH, compared to 34.9% of people in the “Inactive” group, p < 0.05 (Table 2). Figures S1–S3 present the prevalences of positive SPH, depression and anxiety according to PAL.

Moreover, associations were found between mental health according to the GHQ-12 and PAL scores in the total sample and both sexes, p < 0.001. The highest medians GHQ-12 scores were presented by the “Inactive” group in the total sample (13.8 points), in women (14.9 points) and men (12.7 points), presenting significant differences with the rest of the PAL groups, p < 0.001. In contrast, the lowest medians scores were found in the “Very active” groups, in the total sample (9.6 points) and in women (10.0) and men (9.5 points). In none of these groups were significant differences found between “Active” and “Very active” groups (Table 3).

Multiple binary logistic regression model for the risk factors for self-reported depression and anxiety

Being female, older, having lower social class and being physically inactive were risk factors for depression (Nagelkerke’s R² = 11.0%) according to the binary multiple logistic regression model for depression risk factors (Table 4). In addition, being female, older, of lower social class and physically inactive were also risk factors for self-reported anxiety (Nagelkerke’s R² = 7.5%) according to the binary multiple logistic regression model for anxiety risk factors. The same results were obtained in the models for mental health status (Nagelkerke’s R² = 8.8%) (Table 4).

Main findings and theoretical applications

Practical implications and future directions

Although it should be kept in mind that these are preliminary results and it is a cross-sectional study, these results establishing physical inactivity as a risk factor for depression, anxiety and mental health scores is a relevant finding, since the level of physical activity is a modifiable risk factor. Therefore, these results support the need to implement community health promotion programs through physical activity. from the authors’ point of view, the introduction of physical exercise programs in this population could be a useful form of intervention to improve their mental health and help them meet physical activity recommendations (Bull et al., 2020; ACSM’s Guidelines for Exercise Testing and Prescription, 2021).

On the other hand, one of the potentials of our results, in addition to the robustness of the results and the representativeness of the sample analysed, is that they are the latest data recorded in the pre-pandemic period in this Spanish population. This will allow us to conduct to carry out future studies to analyse the im-pact of the pandemic on the relationships between the level of physical activity and mental disorders in this population. In addition, lines should compare the mental health of inactive people with high cholesterol with that of the general population, and possible interaction with sex, in order to see whether might benefit more, or what the optimal dose of physical activity for this type of population. Likewise, the development of longitudinal studies could establish the real impact of physical activity on people with high cholesterol.

Limitations

This study contains some limitations. (1) inherent limitation to the cross-sectional study design so that causal relationships between variables cannot be established. (2) other variables that could influence depression and anxiety were not included, such as some sociodemographic, cultural, alcohol consumption or economic characteristics of the participants. It is advisable to develop future lines of research that address other behaviours such as nutritional or health habits such as taking medication, or the intake of substances such as alcohol as well as the prevalence of access to and use of health services and their characteristics.