Recreational exercise is associated with lower prevalence of depression and anxiety and better quality of life in German people living with HIV

ABSTRACT

Sedentarism is a risk factor for depression and anxiety. People living with the human immunodeficiency virus (PLWH) have a higher prevalence of anxiety and depression compared to HIV-negative individuals. This cross-sectional study (n = 450, median age 44 (19–75), 7.3% females) evaluates the prevalence rates and prevalence ratio (PR) of anxiety and/or depression in PLWH associated with recreational exercise. A decreased likelihood of having anxiety (PR=0.57; 0.36-0.91; p = 0.01), depression (PR=0.41; 0.36-0.94; p=0.01), and comorbid anxiety and depression (PR = 0,43; 0.24-0.75; p=0.002) was found in exercising compared to non-exercising PLWH. Recreational exercise is associated with a lower risk for anxiety and/or depression. Further prospective studies are needed to provide insights on the direction of this association.

KEYWORDS:

• HIV
• depression
• anxiety
• exercise

Introduction

The most common mental disorders in people living with HIV (PLWH) are depression and anxiety (Steel et al., 2014). This is of importance because mental health disorders negatively affect PLWH’s quality of life (QoL) and antiretroviral therapy (ART) success (Uthman et al., 2014).

PLWH are 50% less likely to meet the physical activity recommendations than HIV-negative individuals (Vancampfort et al., 2018), which might be explained by the higher prevalence of physical comorbidities related to the chronic inflammatory state in HIV (Ceccarelli et al., 2013; Ettorre et al., 2011), disabilities (Myezwa et al., 2018) and fatigue (Barroso et al., 2014).

Sedentary PLWH report worse mental well-being and lower QoL scores compared to the physically active PLWH (O’Brien et al., 2016). Physical activity (PA) is related to reductions in anxiety and depression symptoms and serves as a protective factor for depression and anxiety in non-HIV population (Schuch et al., 2018, 2019), given that PA influences mood, cognitive functions, and depression symptoms (Basso & Suzuki, n.d.). It also effects the central nervous system (Firth et al., 2018; Kandola et al., 2019) and HIV chronic inflammation/oxidative stress (Ettorre et al., 2011; Ivanov et al., 2016). Meta-analyses have further shown significant improvements in fitness (including aerobic capacity and strength) in exercising vs. sedentary PLWH (Pérez et al., 2017, 2018). This suggests that planned, structured and repetitive PA done to improve or maintain fitness, may be beneficial in preventing anxiety and/or depression in PLWH.

To date, there is no published data on recreational exercise as a protective factor for depression and/or anxiety in PLWH. Therefore, the aim of the present study was to (a) assess the likelihood of PLWH experiencing anxiety or depression and (b) QoL, in relation to recreational exercise.

Methods

Individuals were recruited from German institutions involved in HIV/AIDS care. Questionnaires were completed in hardcopy or online format. Inclusion criteria were: ≥18 years of age, diagnosed with HIV. Self-reported questionnaires were used to assess exercise, QoL, HIV and ART characteristics, and the presence of anxiety and/or depression. The EuroQoL 5-dimensions (EQ-5D) (Cooper et al., 2017) and the medical outcome study-HIV survey (MOS-HIV) (Wu et al., 1997) were used to assess QoL. Participants scoring 2–5 on dimension #5 on the EQ-5D were classified as having depression/anxiety. The survey investigated regular recreational exercise (RE) defined as “planned, structured and repetitive bodily movement done to improve or maintain one or more components of physical fitness” (Caspersen et al., 1985) performed at least once a week for at least three months (Pollock et al., 1998). To determine exercise time, participants were asked whether they were currently performing exercise or had performed exercise in the past. The time length of exercise practice was measured in years. Exercise frequency was measured in number of sessions per week. Exercise intensity was measured in metabolic equivalent of task (MET), and each reported type of sport was converted into METs according to (Ainsworth et al., 2011). Each exercise session duration was measured in hours per week. Participants were categorized into recreational exercising (i.e., Performed RE in the past and were also still performing RE and/or never performed RE and started regular RE at least once a week in the last 3 months, prior to study) and sedentary (individuals who never performed exercise or used to perform recreational exercise more than 3 months, before completing the survey) PLWH.

The immunological parameters of cluster differentiation four (CD4) white blood cell count was a self-report question. For the total ART duration, participants indicated the year they commenced ART. Age commencing ART was calculated based on the year of HIV diagnosis and the total time under ART. Body mass index (BMI) was calculated after the HIV diagnosis (there was no access to clinical records). Participants who did not answer the questions about exercise frequency, duration or type were excluded from this study. For more study information, refer to methods supplementary material and (Zech et al., 2020).

Statistical analysis

A prior descriptive analysis indicated a non-normal distribution of the variables. The Mann–Whitney U test evaluated differences for continuous variables between the two groups (physically active X sedentary). The Chi-square test with a post-hoc analysis applying the Bonferroni correction (García & Núñez, 2003) was used to evaluate group differences for ordinal and nominal variables. Fisher’s exact test was chosen in cases where frequencies were less than five. A Poisson regression with robust variance was performed to calculate the prevalence ratio (PR) and 95% confidence intervals (95% CI) (Barros & Hirakata, 2003) of anxiety and/or depression. The relationship between depression or anxiety and the different variables in the whole sample were analyzed as follows: the relationship to anthropometric characteristics (age, height, weight, BMI) in model 1; HIV and ART characteristics in model 2; physical/mental health summary scores and the EQ-5D health state index in model 3; exercise characteristics in model 4; and to exercise status: exercising and sedentary in model 5. Model 3–4 were adjusted for years living with HIV. The significance alpha level was set to < 0.05. Data are presented as median and interquartile range (IQR, 1st quartile – 3rd quartile) unless otherwise indicated. All statistical analyses were performed using SPSS (IBM Corp., Armonk, NY, USA).

Ethical aspects

The study was approved by the ethics committee of the Charité Berlin (Protocol No. EA1/084/11). Participant’s data were anonymously collected following participants informed consent.

Results

PLWH affected by anxiety and depression

In the RE group the anxiety prevalence was 33 (12.7%), the depression prevalence was 43 (16.5%), and the prevalence of comorbid anxiety and depression was 32 (12.3%). For sedentary PLWH, the anxiety prevalence was 41 (21.6%), depression was 49 (25.8%), and comorbid anxiety and depression was 28 (14.7%). A significant difference between groups for anxiety (χ²=6.3; df=1; p=0.01) and depression (χ²=5.7; df=1; p=0.01) emerged, with prevalence rates being higher for sedentary PLWH. No significant group difference was found for comorbid anxiety and depression. Please refer to Table 1 for cohort characteristics.

Table 1. Cohort characteristics.

Parameter	Overall	Sedentary	Recreational exercise
	n = 450	n = 190 (42%)	n = 260 (58%)
Gender
Female, n (%)	33(7.3)	23 (69.7)	10 (30.3)*
Male, n (%)	415 (92.2)	165 (39.8)	250 (60.2)*
Other, n (%)	2 (0.4)	2 (100)	0
Age (yrs.)	44 (19–75)	45 (22–75)	43 (19–69)
Height (cm)	180 (150–198)	178 (155–197)	180 (150–198)
Weight (kg)	77 (32–178)	75.5 (32–138)	77 (52–178)
BMI (kg·m2-1)	23.5 (9–63.1)	23.4 (9–41.7)	23.5 (17–63.1)
Years with HIV	8 (1–32)	7.5 (1–31)	8 (1–32)
Years with ART	5.5 (0-29)	6 (0–29)	5 (0–23)
CD4 cell count category
Unknown, n (%)	39 (8.7)	15 (38.5)	24 (61.5)
> 500 cell/µL, n (%)	272(60.7)	106 (39)	166 (61)
200–499 cell/µL, n (%)	110 (24.6)	51 (46.4)	59 (53.6)
< 200 cell/µL, n (%)	27 (6)	16 (59.3)	11 (40.7)
Missing	2	2	0
Exercise
Started after HIV dx, n (%)	–	–	149 (33%)
Started before HIV dx, n (%)	–	–	111 (25%)

Note. Data presented in median and interquartile range (IQR), number of participants (n), and percentage (%). Patient self-reported: antiretroviral treatment (ART), CD4 cell count expressed in cells per microliter (cell·µL^-1). Diagnosis (dx). Significant differences at the 0.05 level between exercising and sedentary PLWH (*).

Qol, anxiety, and depression

An analysis in depressed and/or anxiety PLWH engaging in RE or sedentary activities revealed no differences in QoL scores and the EQ-5D anxiety/depression dimension categories. Data are displayed in Table 2.

Table 2. MOS-HIV and EQ-5D questionnaires.

Parameter	Sedentary	Recreational exercise	X^2	Df	p-value
PHSS	49.5 (20.5–65.1)	56.7 (17.6–63.4)			< 0.05
MHSS	44.4 (13.7–66.9)	51.6 (13.5–67.1)			< 0.05
EQ-5D HSI	0.909 (0.156–1)	0.910 (0.127–1)			< 0.05
EQ-5D VAS	70 (5–100)	85 (11–100)			< 0.05
Anxiety / depression			14.6	4	0.005
Category 1	67 (34.2)	129 (65.8)*
Category 2	59 (34.4)	77 (56.6)
Category 3	40 (50)	40 (50)
Category 4	21 (65.6)	11 (34.4)*
Category 5	3 (50)	3 (50)

Note. Chi square (X²), degrees of freedom (Df), physical health summary score (PHSS), mental health summary score (MHSS), EQ-5D health state index (HSI), EQ-5D visual analog score (VAS). Dimension categories: (1) no problems, (2) slight problems, (3) moderate problems, (4) severe problems, and (5) extreme problems. Group participants: overall n = 450; sedentary n = 190; Recreational exercise n = 260. Significant differences at the 0.05 level between recreational exercising and sedentary PLWH (*). Please refer to results supplementary material for dimension #1-4 of the EQ-5D.

Discussion

The objective was to investigate the relationship between RE and the prevalence of anxiety and/or depression, and QoL in PLWH. The rates of anxiety and depression were lower, QoL scores were better in exercising compared to sedentary PLWH. These findings are in line with previous research in HIV negative (Schuch et al., 2018) and HIV-positive (Heissel et al., 2019).

The prevalence for comorbid anxiety and depression did not differ between groups. This accords with previous cross-sectional studies with a reported prevalence between 8.1% (Camara et al., 2020) and 25% (Tesfaw et al., 2016) in PLWH explained by the impact that a chronic disease (like HIV) has on depression and anxiety (DeJean et al., 2013). Notably, no differences were found in the severity of anxiety and/or depression between exercising and sedentary depressed and/or anxious PLWH.

The reduced depression/anxiety symptoms (Table 3) and better QoL in exercising PLWH, may be attributed to different pathways in the physiopathology of depression and anxiety. Exercise can increase brain-derived neurotrophic factor (BDNF) circulating concentrations (lowered in depressed individuals (Kandola et al., 2019)) and anxiety (Castrén & Kojima, 2017). In PLWH, a chronic inflammatory state is present (Ettorre et al., 2011) and pro-inflammatory markers identified as a cause of depression and anxiety (Renna et al., 2018), are downregulated by exercise (Kandola et al., 2019). Oxidative stress connected to HIV (Ivanov et al., 2016), and ART elevates reactive oxygen species (ROS) in PLWH (Popoola & Awodele, 2016). Depression and anxiety are associated with oxidative stress (Hovatta et al., 2010; Black et al., 2015), and regular/long-term exercise produces an adaptive response to ROS by upregulating the production of antioxidant enzymes and enzymes that repair ROS damage (Kandola et al., 2019). Finally, exercise stimulates the parasympathetic nervous system activity (reduced in people with anxiety), ultimately regulating anxiety symptoms (Stubbs et al., 2017).

Table 3. Risk of anxiety and depression in relation to anthropometric characteristics, HIV and ART, RE characteristics, and RE after the HIV diagnosis in PLWH.

Variables	Depression			Anxiety			Anxiety & depression
	PR	95% CI	p	PR	95% CI	p	PR	95% CI	p
Model 1
Age (yrs.)	1.01	1–1.03	0.03	1.01	1–1.03	0.4	1.02	0.99–1.04	0.05
Weight (kg)	1.12	0.95–1.32	0.11	1.24	1–1.45	0.04	1.22	0.98–1.52	0.06
BMI (kg·m^2−1)	0.71	0.41–1.17	0.64	0.14	0.34–1	0.05	0.52	0.26–1.05	0.08
Model 2
Years since the HIV diagnosis (yrs.)	1.04	0.97–1.11	0.16	1.05	1–1.11	0.1	1	0.90–1.10	0.88
Years under ART (yrs.)	1	0.92–1.09	1	0.96	0.89–1.03	0.05	1	0.91–1.16	0.67
CD4 cell count (cell·µL^−1)	1	0.99–1	0.67	1	0.99–1	0.36	1	0.99–1	0.86
Model 3
PHSS score	0.99	0.96–1.01	0.31	0.98	0.95–1.02	0.2	0.96	0.94–1	0.02
MHSS score	0.94	0.92–0.97	0.001	0.94	0.91–0.96	0.001	0.96	0.93 - 0.99	0.007
EQ-5D health index	0.30	0.08–1.13	0.01	0.26	0.06–1.15	0.02	0.22	0.04–1.31	0.03
Visual analog health score	0.98	0.97–0.99	0.001	0.98	0.97–0.99	0.001	0.98	0.97–1	0.03
Model 4
Frequency (session·week^−1)	1.14	0.93–1.37	0.06	1.06	0.83–1.31	0.5	1	0.73–1.31	0.97
Intensity (METs)	0.80	0.64–1	0.04*	0.85	0.66–1.1	0.15	0.88	0.65–1.2	0.38
Duration (hrs·week^−1)	0.95	0.87–1.03	0.17	0.98	0.89–1.06	0.59	0.98	0.87–1.08	0.69
Model 5
Exercise	0.62	0.41–0.94	0.01	0.57	0.36–0.91	0.01	0.43	0.24–0.75	0.002
Sedentary	Ref	Ref	Ref	Ref	Ref	Ref	Ref	Ref	Ref

Note. Prevalence ratio (PR), 95% confidence interval (95% CI). CD4 cell count expressed in cells per microliter (cell·µL^-1), Exercise intensity expressed in metabolic equivalent of task (METs), reference group (Ref). (*) this relation disappeared after controlling for years living the HIV diagnosis.

Lower QoL scores in sedentary compared to exercising PLWH (Table 2), may be related to health/physical limitations (i.e., Frail PLWH (Blanco et al., 2019)), that hinder aptitude to perform exercise or, alternatively, discontinued exercise due to worse HIV-status.

Limitations

Although a strength was the sample size (N=450), it was not nationally representative due to the convenience sampling, and although predominantly male with a distribution representative for PLWH (Reinsch et al., 2011), exercise anti-depressive effects vary among HIV-negative females and males (Bhui & Fletcher, 2000; Zhang & Yen, 2015). A self-report assessment was used; this may have increased the risk of memory and social desirability bias. Quantification of RE was not assessed through accelerometry or a validated PA questionnaire.

Conclusion and implications for practice

A lower likelihood of experiencing depression and/or anxiety was found in exercising PLWH. Exercise anxiolytic and anti-depressive effect is instrumental in informing exercise lifestyle interventions for PLWH to enhance their QoL. Exercise guidelines to control and adjust intensity, time, and frequency of exercise should be followed to prevent the occurrence of anxiety and/or depressive symptoms.

Authors contribution

Conceptualization: Camilo Pérez-Chaparro and Philipp Zech; Methodology: Camilo Pérez-Chaparro and Philipp Zech; Formal analysis and investigation: Camilo Pérez-Chaparro; Writing – original draft preparation: Camilo Pérez-Chaparro; Writing – review and editing: Maria Kangas, Philipp Zech, Felipe B. Schuch, Michael Rapp, Andreas Heissel, Supervision: Andreas Heissel.

Data availabity statement

The authors confirm that the data supporting the findings of this study are available within the supplementary materials.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

Camilo Pérez-Chaparro was funded by the Fundación para el futuro de Colombia and Deutsche Akademische Austauschdienst. Andreas Heissel was funded by the University of Potsdam. Philipp Zech was funded by the FAZIT Stiftung. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.