Exercise and mental health of people living with HIV: A systematic review

Abstract

Objective

Mental illness is highly prevalent among people living with HIV. Poor mental health is linked to HIV disease progression, making the treatment of mental illness alongside HIV essential. While the benefits of exercise on the physical health of people living with HIV are well established, the effect of exercise on mental health in this population is less examined. Therefore, this study aimed to conduct a systematic literature review of the effects of exercise on mental health in people living with HIV.

Methods

A search of electronic databases (PubMed, Web of Science, PsycINFO) through 30 November 2016 was completed. The methodological framework for scoping studies was used to conduct the review process. RISMA guidelines were used to report the results.

Results

The search resulted in 2273 articles and 52 were determined to be relevant. After review of the full text of potentially relevant studies, 24 studies were included for the analysis.

Discussion

Both aerobic and resistance exercise have independent and combined positive effects on various indicators of mental health in people living with HIV. Major limitations include high attrition rate, small sample size, and poor study designs. Higher quality studies with more diverse populations such as women, older adults, and transgender individuals are required.

Keywords

Depression quality of life HIV/AIDS resistance exercise aerobic exercise

Introduction

Globally, 36.7 million people were living with human immunodeficiency virus (HIV) at the end of 2015 and, in 2015, approximately 2.1 million people became newly infected.¹ HIV infection can now be considered a chronic condition in countries where antiretroviral therapy (ART) is available and accessible to people living with HIV (PLWH). In fact, the HIV mortality rate dropped by 94% from 1996 to 2006 in countries with high rates of access to ART;² and a study of HIV-infected adults in the United Kingdom has shown that this life expectancy improvement is mainly attributed to CD4+ cell count restoration and viral load suppression achieved through ART adherence.³ Optimal ART adherence not only increases the life expectancy of PLWH, but also decreases the HIV transmission risk dramatically.^4,5

As a chronic condition, HIV is often accompanied by a high prevalence of co-occurring mental illness. For example, in the United States, it is estimated that 16% of those with HIV have a comorbid anxiety, while up to 36% may have comorbid depression.^6–8 In fact, in the United States, the prevalence of depression is estimated to be three times higher among PLWH compared to their seronegative counterparts.⁹ Similarly, studies conducted with newly diagnosed HIV-infected adults from South Africa¹⁰ and Cameroon¹¹ show that 55% and 63% of the patients were suffering from depression, respectively. These high rates of depression are particularly concerning because negative affective states have been linked to poor medication adherence and the clinical progression of the disease.^12–14 More specifically, a meta-analysis of cross-sectional studies, which examined the association between depression treatment and ART adherence, reported that the odds of ART adherence is 83% better for those being treated for depression.¹⁴ Moreover, a review of longitudinal studies conducted from 1990 to 2007 indicated that poor mental health was associated with higher viral load levels, a decline in CD4+ cell count, and an increased risk of mortality.¹³

Although the specific biological mechanisms involved in these relationships are currently unknown, it is clear that disease progression is impacted by negative affective states and depressive symptoms. As such, there is a strong rationale to investigate interventions that are designed to improve mental health in PLWH. Exercise may be one such intervention as some recent evidence suggests. Specifically, a recent meta-analysis on the effect of aerobic exercise on the health of PLWH found significant improvements in quality of life and depressive symptoms.¹⁵ However, another meta-analysis was inconclusive on the effect of resistance exercise on mental health of PLWH.¹⁶ Importantly, both meta-analyses were limited to randomized controlled trials and exercise interventions that were carried out at least three times per week for at least 4 weeks.^15,16 Therefore, the purpose of this paper was to further investigate the effects of aerobic and resistance exercise on mental health outcomes in PLWH by identifying and critically reviewing any published study to date. In order to demonstrate the scope of the studies conducted so far, the studies in this review were not limited by study design (i.e. both randomized controlled trials and quasi-experimental), or level of exercise frequency, intensity, or duration.

Methods

Eligibility criteria

The inclusion criteria were determined prior to the literature search. Studies had to include the following: (a) participants were people living with HIV/AIDS, (b) exercise (i.e. resistance, aerobic exercise or combined) was used as an intervention, (c) at least one mental health outcome was measured pre-to-post intervention, and (d) published in English. Both experimental and quasi-experimental designs were included. We also included interventions that combined aerobic or resistance exercise with other movement modalities (e.g. Tai Chi) or nonexercise modalities (e.g. pharmacological interventions or nutritional interventions). The methodological framework for scoping studies was used to conduct the review process.¹⁷ PRISMA guidelines were used to report the results.¹⁸

Search strategy

A systematic review of published (i.e. in print and online) studies testing the effects of exercise on mental health of PLWH was conducted. An electronic database literature search through 30 November 2016 was conducted using “PubMed”, “Web of Science”, and “PsycINFO”. The keywords used for the search were (“hiv” OR “aids”), (“exercise” OR “physical activity” OR “aerobic” OR “resistance” OR “walk”), (“mental health” OR “depression” OR “depressive symptoms” OR “quality of life” OR “mood” OR “fatigue” OR “anxiety”) and the syntax was adapted for each database accordingly. A manual search through the references of the review articles was also conducted to include all the relevant literature.

Data extraction and synthesis

The information related to study eligibility was extracted by one reviewer. In case of any uncertainty about whether an article met the inclusion criteria, the article was discussed by the coauthors to determine exclusion or inclusion in the review. The following information was extracted for each included study: (a) characteristics of the participants; (b) characteristics of the exercise intervention (e.g. type, frequency, intensity, duration); (c) characteristics of the control condition; (d) type of outcome measured (e.g. depressive symptoms, mood, anxiety, quality of life); (e) the instruments administered to measure the outcomes; (f) study design; and (g) program attrition rate (i.e. the percentage of participants who dropped out). Formal assessment of risk of bias was not included due to variability in study design and the limited number of studies meeting the inclusion criteria.¹⁹

Results

The search resulted in 2273 articles. After excluding duplicates across the databases and articles that clearly did not match the inclusion criteria, 52 were determined to be relevant. Of these, 24 studies were eligible and are included in this review. The search procedure and study selection methodology are shown in Figure 1, and the participants’ characteristics, study designs, and outcomes are presented in Table 1. Characteristics of the intervention and control group are presented in Table 2. The results are categorized by mode of the exercise interventions.

Figure 1.

The search procedure and study selection methodology.

Table 1.

Participants’ characteristics, study designs, and outcomes.

Author (year)	Study design	Sample size (at baseline)	Men (%)	Mean age (y)	Attrition rate	Mental health outcome measured	Mental health results
LaPerriere et al. (1990)²⁰	RCT 2 groups: AE; CON	50 (never tested for HIV before)	100	30.9	NR	Anxiety and depressive symptoms	Upon HIV status notification: sig. ↓ in anxiety and depressive symptoms for AE compared to CON
MacArthur et al. (1993)²¹	Randomized uncontrolled trial 2 groups: vigorous intensity AE; moderate intensity AE	25	96	35.4	19/25 (76%)	Anxiety and depressive symptoms	No sig. change
Lox et al. (1995)²⁵	RCT 3 groups: AE; RE; stretching and flexibility (FLEX)	34	100	36.4	1/34 (3%)	Mood, Life satisfaction, Physical self-efficacy	Sig. ↑ in positive mood, life satisfaction, and sig. ↓ in negative mood for both AE and RE compared to CON (FLEX)
Stringer et al. (1998)²⁴	RCT 3 groups: vigorous intensity AE; moderate intensity AE; CON	34	92	36	8/34 (23%)	QOL	Sig. ↑ in QOL for both AE groups compared to CON
Terry et al. (1999)²⁶	Randomized uncontrolled trial 2 groups: vigorous intensity AE; moderate intensity AE	31	67	30.5	10/31 (32%)	Depressive Symptoms	No sig. change
Bhasin et al. (2000)⁶⁷	RCT 4 groups: placebo, no RE (PNR); testosterone, no RE (TNR); placebo, RE (PRE); testosterone, RE (TRE)	61 (with testosterone deficiency)	100	41.8	12/61 (20%)	QOL	No sig. change
Agin et al. (2001)⁶¹	RCT 3 groups: whey protein (PRO); RE; combined whey protein and RE (PRO + RE)	37	0	40.8	7/37 (19%)	QOL	Sig. ↑ in physical activity, health perceptions and vitality subscales for RE; sig. ↑ in social functioning subscale for PRO + RE; sig. ↓ in physical activity subscale for PRO
Baigis et al. (2002)²³	RCT 2 groups: AE; CON	123	80	37	24/123 (19%)	QOL	Sig. ↑ in overall health subscale of MOS-HIV for AE compared to CON
Neidig et al. (2003)²²	RCT 2 groups: AE; CON	60	92	36	12/60 (20%; 40% in AE, 0% in CON)	Depressive symptoms and mood	Sig. ↓ in CES-D and POMS score, and sig. ↓ in depression/dejection subscales of POMS for AE compared to CON
Rojas et al. (2003)⁵⁰	Nonrandomized controlled trial 2 groups: AE + RE; CON	39	70	46.2	6/39 (15%)	Depressive symptoms, anxiety, QOL	Sig. ↑ in emotional well-being, physical strength and energy subscale for AE + RE compared to CON. No sig. changes in depressive symptoms and anxiety
Shevitz et al. (2005)⁶⁸	Randomized uncontrolled trial 3 groups: nutrition intervention combined with placebo pills (NI + Placebo); nutrition intervention combined with 10 mg of oxandrolone twice daily (NI + OX); nutrition intervention combined with RE (NI + RE)	50	70	42.5	3/50 (6%)	QOL	Sig. ↑ in physical functioning subscale for NI + RE pre to post
Galantino et al. (2005)⁴⁷	RCT 3 groups: combined AE and RE; Tai chi; CON	51	100	NR	13/51 (25%)	Mood, QOL, spirituality	Sig. ↑ in overall health subscale of QOL for both EX groups compared to CON. Sig. ↓ in tension/anxiety and confusion/bewilderment subscales of POMS for both EX groups compared to CON.
Fillipas et al. (2006)⁴⁹	RCT 2 groups: combined AE and RE; CON	40	100	43.5	5/40 (12%)	Self-efficacy QOL	Sig. ↑ in self-efficacy for the AE + RE compared to CON. Sig. ↑ in overall health and cognitive function subscale of QOL for the AE + RE compared to CON
Engelson et al. (2006)⁶²	Pre-post-test 1 group: combined AE, RE and nutrition education class	28	0	41.5	10/28 (36%)	QOL, depressive symptoms and anxiety, overall happiness, sense of coherence	Sig. ↑ in all the items except for anxiety subscale on BSI and general mental health subscale of QOL
Perez-Moreno et al. (2007)⁵²	RCT 2 groups: combined AE and RE; CON	27	100	37	8/27 (30%)	QOL	Sig. ↑ in QOL in the AE + RE pre to post; no sig difference between groups
Mutimura et al. (2008)²⁷	RCT 2 groups: AE; CON	100	40	38	3/100 (3%)	QOL	Sig. ↑ in overall QOL for AE compared to CON. Specifically, sig. improvement in psychological, independence, and HIV ART specific domains of QOL
Gomes et al. (2010)⁵¹	Nonrandomized controlled trial 2 groups: combined AE and RE; CON	29	59	45	0/29 0% The control group consisted of waitlist and participants who.	Life satisfaction	Sig. ↑ in life satisfaction in the AE + RT group compared to the CON
Maharaj et al. (2011)²⁸	RCT 2 groups: AE; CON	52	78	33.2	16/52 (31%)	QOL	Sig. ↑ in all domains of QOL for AE compared to CON
Ogalha et al. (2011)⁶³	RCT 2 groups: combined AE and nutrition counseling; CON (nutrition counseling only)	70	54	43.1	7/70 (10%; 20% in CON, no participants dropped out of the AE + counseling)	QOL	Sig. ↑ in general health, vitality and mental health subscale of QOL for AE+ counseling compared to CON
Jaggers et al. (2014)⁴⁸	RCT 2 groups: combined AE and RE (AE + RE); CON	93	76	42.3	44/93 (47%)	Mood, HIV related distress symptoms, psychological stress, salivary cortisol (immediately after awakening, 1 and 2 hours after awakening)	Sig. ↓ in total POMS score and depression subscale for AE + RE compared to CON; sig. ↓ in cortisol levels for AE + RE compared to CON; sig. ↑ in perceived stress for CON compared to AE + RE
Hamid et al. (2015)⁴⁴	Randomized uncontrolled trial 2 groups: RE using resistance band; RE not using resistance band (i.e. using bodyweight)	29	100	41.7	10/29 (34%)	QOL	Sig. ↑ in QOL in the RE using bodyweight. No sig. difference between groups
Guerra et al. (2016)⁴³	Pre-post-test 1 group: RE	15	73	43.7	0/15 (0%)	QOL	Sig. ↑ in spirituality, perception of QOL, environment, and general health subscale of QOL pre to post.
Aweto et al. (2016)⁶⁴	RCT 2 groups: combined AE and counseling; CON (counseling only)	40	30	31.3	7/40 (18%)	Depressive symptoms	Sig. ↓in depression scores for combined AE and counseling group pre to post; no change for CON
Mkandla et al. (2016)⁴⁵	RCT 2 groups: RE; CON	160	29	42.2	96/160 (60%; 63% in CON; 56% in RE)	QOL	Sig. ↑ in QOL in the RE compared to CON

RCT: randomized controlled trial. NR: not reported. AE: aerobic exercise group; RE: resistance exercise group; CON: control group; EX: exercise group; QOL: quality of life. POMS: profile of mood states. CES-D: Center for Epidemiological Studies-Depression Scale.

Table 2.

Characteristics of the intervention and the control group.

Author (Year)	Exercise mode	Intensity	Exercise frequency	Exercise duration^a (min)	Control condition
La Perriere et al. (1990)²⁰	Cycle ergometer	Interval training: 70–80% age-predicted MHR	5 wks 3× per wk	45	Nonexercise
MacArthur et al. (1993)²¹	Various aerobic interval training (i.e. treadmill walking, jogging, rowing, stair stepper)	Vigorous: 75–85% VO_2max Low: 50–60% VO_2max	24 wks 3× per wk	24 40	None
Lox et al. (1995)²⁵	Cycle ergometer	50–80% HRR	12 wks 3× per wk	24	Stretching and flexibility (unsupervised)
	Progressive resistance exercise	3 sets of 10 repetitions at 60% of 1 RM		24
Stringer et al. (1998)²⁴	Cycle ergometer	Vigorous: 50% of difference between LT and VO_2max	6 wks 3× per wk	30–40	Nonexercise
		Moderate: 80% of LT		60
Terry et al. (1999)²⁶	Run Walk	Vigorous: 75–85% MHR Moderate: 55–60% MHR	12 wks 3× per wk	30	None
Bhasin et al. (2000)⁶⁷	Progressive resistance exercise	60% of 1RM for the initial 4 weeks followed by 12 weeks of periodized program: 90% of 1RM on heavy days, 80% of 1 RM on medium days, 70% of 1 RM on light days	16 wks 3× per wk	NR	Placebo
Agin et al. (2001)⁶¹	Progressive resistance exercise	3 sets of 8–10 repetitions at 75% of 1 RM	14 wks 3× per wk	NR	Participants were their own controls by attending 6-week pre-intervention assessment period
Baigis et al. (2002)²³	Ski machine	75–85% of measured MHR	15 wks 3× per wk	20	Attention only (once per week visit from research staff to collect completed diary sheets)
Neidig et al. (2003)²²	Treadmill and cycle ergometer	60–80% VO_2max	12 wks 3× per wk	60	Waitlist
Rojas et al. (2003)⁵⁰	Combination of various aerobic exercises (circuit training, cross-country run, obstacle course), and progressive resistance exercise	Moderate (heart rate does not exceed 140 bpm) 50–80% of 1 RM	16 wks 2× per wk	25 15	AIDS-related support group
Shevitz et al. (2005)⁶⁸	Progressive resistance exercise	3 sets of 5 exercises at 80% of 1 RM	15 wks 3× per wk	NR	None
Galantino et al. (2005)⁴⁷	Combination of various aerobic exercises followed by resistance exercise using resistance band; Tai chi	60–70% of HRR	8 wks 2× per wk	25	Nonexercise
Fillipas et al. (2006)⁴⁹	Combination of aerobic (bike, treadmill, cross trainer, stair stepper) and progressive resistance exercise	60–75% MHR 60% of 1 RM progressed to 80% of 1 RM	24 wks 2× per wk	20 30	Unsupervised walking program twice per week and attended a forum monthly
Engelson et al. (2006)⁶²	Combination of treadmill and progressive resistance exercise	70–80% of age-predicted MHR	12 wks 3× per wk	30	None
		3 sets of 10 exercises at 8–10 repetitions		60
Perez-Moreno et al. (2007)⁵²	Combination of cycle ergometer and progressive resistance exercise	70% of age-predicted MHR progressed to 80% of age-predicted MHR 2–3 sets of 12–15 RM for the initial 4 weeks progressed to 2–3 sets of 8–10 RM	12 wks 3× per wk	70 (aerobic and resistance combined)	Nonexercise
Mutimura et al. (2008)²⁷	Jogging, running, stair stepper	45% of age-predicted MHR for the initial 3 weeks, 60% of age-predicted MHR for next 6 weeks, 75% of age-predicted MHR for last 6 weeks	24 wks 3× per wk	45–60	Nonexercise
Gomes et al. (2010)⁵¹	Combination of treadmill, cycle ergometer and progressive resistance exercise	Moderate (heart rate does not exceed 150 bpm) 3 sets of 5 exercises at 12 repetitions (60–80% of 1 RM)	12 wks 3× wk	30 NR	Nonexercise
Maharaj et al. (2011)²⁸	Cycle ergometer, treadmill	50–70% age-predicted MHR at research site 30 min (10 min brisk walk, 10 min jog in place, 10 min bodyweight squat) at home	12 wks 4× per wk (1× per wk at research site, 3× per wk at home)	40 min at research site; 30 min at home	Attention only
Ogalha et al. (2011)⁶³	Gym class	75% of MHR	24 wks 3× wk	60	Nutritional counseling (once per month)
Jaggers et al. (2014)⁴⁸	Combination of treadmill and progressive resistance exercise	50–70% age-predicted MHR	6 wks 2× per wk	30	Attention only (twice per week; sessions included reading books, talking, watching television, or participating in other sedentary behaviors)
		12 repetitions		20
Hamid et al. (2015)⁴⁴	Progressive resistance exercise using resistance band	The group using resistance band: 1 sets of 25 repetitions to fatigue for the initial 4 weeks progressed to 3 sets of 12 repetitions to fatigue for the last 8 weeks; the group not using resistance band: performed the same exercises without any equipment in group setting	12 wks 6× per wk	NR	None
Guerra et al. (2016)⁴³	Progressive resistance exercise	1 set of 8 exercises at 15 RM for the initial 4 weeks progressed to 3 sets of 8 exercises at 12 RM	16 wks 3× per wk	40	None
Aweto et al. (2016)⁶⁴	Cycle ergometer	50–60% HRR	6 wks 3× per wk	30	Counseling (once every 2 weeks)
Mkandla et al. (2016)⁴⁵	Progressive lower body resistance exercise	NR	12 wks NR	NR	Nonexercise

The total time is the time of each session excluding warm-up and cool-down.

MHR: maximum heart rate; HRR: heart rate reserve; LT: lactate threshold; RM: repetition maximum; bpm: beats per minute; None: no control group; Nonexercise: the control group did not participate in any sessions and were asked to continue with their activities of daily living; NR: not reported.

Aerobic exercise interventions

Nine studies^20–28 examined the effect of aerobic exercise on mental health of PLWH. Researchers tested both continuous and interval exercise at different intensities. Three studies^20,21,26 examined interval aerobic exercise and the other six^{22–25,27,28} tested continuous aerobic exercise. The intensities tested were low,^21,27 moderate,^{20,23,24,26–28} and vigorous.^{21,22,24–26} Interventions ranged from 5 weeks²⁰ to 24 weeks,^21,27 with the duration of each exercise session lasting between 20 min²³ and 60 min.^24,27 Exercise frequency was three times per week across all experiments except for one study,²⁸ which had the participants exercise once per week at the research site and three times at home.

Four studies^23,24,27,28 measured quality of life (QOL), four^20–22,26 measured depressive symptoms, two^20,21 measured levels of anxiety, and one²⁵ measured affect, life satisfaction, and self-efficacy.

Two studies^20,22 used the profile of mood states (POMS),²⁹ one study²¹ used the general health questionnaire,³⁰ and one²⁶ used the Montgomery–Asberg³¹ to measure the anxiety and depressive symptoms. Additionally, the Center for Epidemiological Studies-Depression Scale (CES-D) and the Beck Depression Inventory (BDI)^32,33 were used to measure the depressive symptoms.²² Affect was measured²⁵ with the Positive And Negative Affect Schedule (PANAS),³⁴ life satisfaction with the Satisfaction With Life Scale (SWLS),³⁵ and self-efficacy with the Physical Self-Efficacy Scale (PSES).³⁶ QOL was measured²⁴ with a subset of a QOL questionnaire,³⁷ the Sickness Impact Profile (SIP),³⁸ the Medical Outcome Study in HIV-infection (MOS-HIV),³⁹ the Duke Activity Status Index (DASI),^23,40 the World Health Organization Quality of Life (WHOQOL),^27,41 or the Medical Outcome Survey 36-item Short-Form Health Survey (SF-36).^28,42

The findings of these studies are mixed while the majority of the studies show significant improvement in one mental health outcome. In one study,²⁰ seropositive controls scored significantly higher in anxiety and depressive symptoms at the 1-week post HIV status notification when compared to the seropositive exercisers. Other researchers^22,25 reported significant improvements in the depressive symptoms, significantly higher positive mood, higher life satisfaction, and lower negative mood in the exercise group when compared to the control group. However, in two other studies,^21,26 the researchers reported no significant change in anxiety and depressive symptoms. Out of the four studies that measured QOL, one²³ reported significant improvement only in the health subscale of QOL in the exercise group compared to the control. Two^27,28 reported significant improvement in the overall QOL domain in the exercise group compared to the control. The other one²⁴ reported significant improvement in QOL for both exercise groups compared to the control with no significant differences between the exercise groups (i.e. vigorous vs. moderate intensity).

On the whole, aerobic exercise seems to be beneficial for mental health of PLWH. However, when drawing conclusions from these results, it is important to consider the shortcomings of the current body of research. First, the attrition rate varies across the studies. Specifically, the attrition rate ranged from 3% in two studies^25,27 to 76% in another study.²¹ Second, participants were 100% men in two studies,^20,25 and between 40% and 96% men in other studies.^{21–24,26–28} Third, two studies^21,26 did not have a control group and the contact time in the control group was not matched in two other studies.^23,25 Lastly, only one study²² controlled for HIV disease stage and balanced the groups based on a number of potentially confounding variables (e.g. social support, perceived stress, anti-depressant use, and HIV-related symptoms).

Resistance exercise versus aerobic exercise interventions

There is only one study²⁵ that tested the effects of resistance versus aerobic exercise. In this study,²⁵ mental health status was compared between aerobic exercise, resistance exercise, and an unsupervised stretching control group. The aerobic exercise group exercised continuously at moderate to vigorous intensity, while the resistance exercise group completed progressive resistance exercise. Affect, life satisfaction, and self-efficacy was compared between groups pre- and post-intervention. Results indicated that both the aerobic and resistance exercise groups had significantly higher positive affect, higher life satisfaction, and lower negative affect when compared to the stretching control group, with significantly greater improvements found in the aerobic group. Of note, the participants were all men and there were more participants diagnosed with AIDS in the resistance exercise group compared to the aerobic group.

Resistance exercise interventions

There are four studies^25,43–45 that examined the effect of resistance exercise on mental health of PLWH. All of these studies reported administering progressive resistance exercise. One study⁴⁴ had two resistance exercise groups. In this study,⁴⁴ one group used resistance bands and initiated the exercise protocol at 1 set of 25 repetitions to fatigue and progressed to 2–3 sets of 8–12 repetitions to fatigue. The other group used bodyweight (i.e. no resistance band) and performed the same exercises in a group setting. Another study,⁴³ progressed the participants from 1 set of 15 repetition maximum (RM) to 3 sets of 12 RM. Lox et al.²⁵ had participants complete 3 sets of 10 RM throughout the intervention. One study,⁴⁵ however, has not reported the details of the resistance exercise protocol. The frequency of the exercise sessions were three times per week in two studies,^25,43 and six times per week in another.⁴⁴ The length of the interventions were 12^25,44,45 to 16 weeks.⁴³

Three studies measured QOL.^43–45 One study²⁵ measured mood, life satisfaction, and physical self-efficacy. QOL was measured through the WHOQOL,⁴¹ the Euro QOL-5 Dimensions,⁴⁶ and the SF-36.⁴² Two studies,^43,45 reported significant improvement in QOL for the exercise group compared to the control group, while Hamid et al.⁴⁴ reported no significant differences in QOL between the two exercise groups. Additionally, Lox et al.²⁵ reported significant improvement in mood and life satisfaction for the exercise group compared to the control.

Of note, two studies did not have a control group,^43,44 and the contact time in the control group was not matched in one study.²⁵ Except for one study⁴³ with perfect attendance, the attrition rate ranged from 3%²⁵ to 60%.⁴⁵ Participants were 100% men in two studies,^25,44 73% men in one,⁴³ and 29% in the other.⁴⁵

Combined aerobic and resistance exercise interventions

There are six studies^47–52 that combined aerobic and resistance exercise. The aerobic exercise protocols tested were low,⁴⁷ moderate,^48–51 and vigorous.⁵² The duration of aerobic exercise sessions were between 20⁴⁹ and 45 min.⁵² The resistance exercise protocols were mostly initiated at 50–60% of 1 RM and progressed to 80% of 1 RM.^49–51,52 One study,⁴⁷ however, used resistance bands and one study⁴⁸ maintained the resistance at 12 RM during the entire intervention. The frequency of the sessions were 2^47–50 or 3^51,52 times per week. All studies compared combined aerobic and resistance exercise with a control group only, except for one study,⁴⁷ which had two exercise groups (i.e. combined aerobic and resistance exercise and a group performing Tai Chi movements) and a control group.

Three studies^47,51,52 asked the control group not to participate in any sessions and to continue with their activities of daily living. Other studies had participants attend an AIDS-related support group,⁵⁰ attend sessions which included reading books, talking, watching television, or participating in other sedentary behaviors,⁴⁸ or attend an unsupervised walking program and a monthly forum.⁴⁹

Two studies^47,48 measured mood. Four studies^47,49,50,52 measured QOL. Additionally, Galantino et al.⁴⁷ measured spirituality, Fillipas et al.⁴⁹ measured self-efficacy, and Rojas et al.⁵⁰ measured psychological well-being. Jaggers et al.⁴⁸ measured psychological distress, self-reported HIV symptoms (e.g. insomnia, pain, fatigue) and salivary cortisol, and Gomes et al.⁵¹ measured life satisfaction. Mood was measured with the brief POMS (POMS-30).⁵³ QOL was assessed with the MOS-HIV³⁹ and a QOL questionnaire developed for Spanish speakers undergoing treatment for drug addiction.⁵⁴ Spirituality was measured with the Spirituality Well-being Scale (SWB),⁵⁵ and self-efficacy was measured through the General Self-Efficacy Scale.⁵⁶ Self-reported HIV symptoms were measured with the HIV Symptom Distress Scale (SDS),⁵⁷ psychological stress was measured with the Perceived Stress Scale (PSS),⁵⁸ psychological well-being was measured through the Symptom Checklist Revised (SCL-90),⁵⁹ and life satisfaction was measured through the Life Satisfaction Index.⁶⁰

Except for one study,⁵² which did not report significant differences between groups, all others show significant improvement in mental health status. Three studies^47,49,50 reported improvement in QOL for the exercise group compared to the control group. Specifically, improvements are reported in emotional well-being, physical strength, energy, overall health, and cognitive function subscales of QOL.^49,50 Additionally, Fillipas et al.⁴⁹ reported significant improvement in self-efficacy in the exercise group compared to the control group. Jaggers et al.⁴⁸ reported a significant reduction in the depressive symptoms and mood disturbances as well as significant decrease in salivary cortisol levels in the exercise group. Interestingly, the control group scored significantly higher in perceived stress post intervention, suggesting a protective role of exercise against stress.⁴⁸ Significant improvements in life satisfaction are also reported in the exercise group compared to the control.⁵¹

Three studies^47,49,50 had participants exercise in groups, and two studies^50,51 did not conduct random assignments. Moreover, in one study⁵¹ the control group comprised both the waitlist and the nonadherent participants in the exercise group. Participants were 100% men in three studies,^47,49,52 and between 59% to 76% men in others.^48,50,51 The attrition rate ranged from 12%⁴⁹ to 47%.⁴⁸

Combined exercise and nutritional supplementation/counseling interventions

There are four studies^61–64 that combined exercise and nutritional supplementation or counseling. In one,⁶¹ participants were randomized to 14 weeks of either whey protein supplementation group (1.0 g/kg/day of whey protein), progressive resistance exercise group, or combined whey protein supplementation and progressive resistance exercise group. Prior to the intervention, participants attended a 6-week pre-intervention assessment period to serve as their own control. In another one,⁶² researchers tested the effect of a weight loss program on psychological parameters in a 12-week longitudinal study.

Two other studies^63,64 tested the effect of combined aerobic exercise with counseling compared to a control group. In one,⁶¹ the exercise protocol was progressive resistance exercise. In the other,⁶² the exercise protocol was a combination of aerobic and progressive resistance exercise and participants attended group nutrition education classes once a week as well. In the studies with aerobic exercise intervention,^63,64 participants either had to complete moderate intensity aerobic exercise three times per week combined with counseling once a week⁶⁴ or had to attend an hour gym class three times per week combined with nutritional counseling once per month.⁶³

QOL was measured through the MOS-HIV³⁹ and the SF-36.^42,61–63 Both the Brief Symptom Inventory (BSI)⁶⁵ and the BDI³² were used to measure the depressive symptoms.^62,64 The 5-item SWLS³⁵ measured overall happiness, and the 13-item Sense of Coherence Scale⁶⁶ was used to measure individual’s beliefs regarding the fairness, manageability, and meaningfulness of their circumstances.⁶²

All of the studies^61–64 indicated significant improvement in the mental health status with exercise. Specifically, Agin et al.⁶¹ reported that the progressive resistance exercise group showed significant improvement in subscales of physical activity, vitality, and general health perception of QOL and the combined resistance exercise and whey supplementation group significantly improved on the social functioning subscale of the QOL. Interestingly, there was a significant decline in the physical activity subscale of QOL for the whey protein supplementation group. Engelson et al.⁶² reported that QOL scores significantly improved in all the dimensions measured, except for general mental health on SF-36 and anxiety subscale on BSI. Aweto et al.⁶⁴ reported significant improvement in the depressive symptoms for the exercise group while there was no change for the control group. Finally, Ogalha et al.⁶³ reported significant improvement in general health, vitality, and mental health subscale of SF-36 in the exercise group compared to the control group.

Importantly, attrition rate ranged from 10%⁶³ to 36%.⁶² One study⁶² did not have a control group and the contact time in the control group was not matched in two other studies.^63,64 Participants were mostly women in three of these studies (i.e. 100% in two,^61,62 and 70% in another.)⁶⁴

Combined exercise and pharmacological treatment

There are two studies^67,68 that looked at the effect of combined exercise and testosterone treatment. The first one⁶⁷ used a four-group double-blind randomized controlled trial design. The groups were (1) placebo, no exercise; (2) testosterone replacement, no exercise; (3) placebo combined with progressive resistance exercise; and (4) testosterone combined with progressive resistance exercise. Testosterone treatment was 100 mg per week of testosterone enanthate. In the second study,⁶⁸ participants were randomized to either a nutrition intervention (i.e. increased caloric intake to 50 kcal/kg/day and increased protein intake to 1.6 g/kg/day) with placebo pills, a nutrition intervention with 10 mg of oxandrolone twice daily, or a nutrition intervention with progressive resistance exercise.

Both studies^67,68 used progressive resistance exercise. Both the frequency of the sessions (i.e. three times per week) and the length of the interventions (i.e. 16 weeks in the first one⁶⁷ and 15 weeks in the second one⁶⁸) were similar.

In the first study,⁶⁷ QOL was measured through a nonspecified self-report questionnaire including multi-item measures of physical functioning, general health perception, emotional well-being, and cognitive and sexual functioning. In the second one,⁶⁸ QOL was measured through the subscale of physical functioning from SF-36.⁴²

The findings of these two studies are mixed. Bhasin et al.⁶⁷ did not report any significant differences in QOL measures, while Shevitz et al.⁶⁸ reported significant improvement in QOL in the combined nutrition intervention and resistance exercise group pre- to post-intervention with no difference between the groups. Attrition rate was also different across these studies. Bhasin et al.⁶⁷ had an attrition rate of 20% while Shevitz et al.⁶⁸ had an attrition rate of 6%. Participants were mostly men in these two studies (i.e. 100% men in one⁶⁷ and 70% men in the other).⁶⁸

Discussion

The effects of both aerobic and resistance exercise on mental health have been tested, with positive results shown independently for each type of exercise, at different intensities, and when they are combined. Importantly, no adverse effects were reported with exercise in this population. These results are also supported by the recent meta-analytic research on the beneficial effects of aerobic exercise with PLWH¹⁵ and other meta-analysis on the effect of exercise on mental health with the general population.⁶⁹ However, it would be difficult to draw a detailed exercise recommendation for improvement of mental health from the reviewed studies due to limitations of the current body of research.

Many of the studies had a small sample size and were inadequately powered. For example, sample size ranged from 15⁴³ to 160.⁴⁵ Many experiments also reported a high level of attrition with seven studies reporting more than 30%.^{21,26,28,44,45,48,62} Unfortunately, it is not clear why some studies experienced such high attrition, but it could be attributable to the exercise protocol (i.e. frequency, intensity, duration, and type), lack of remuneration, or logistical barriers, such as scheduling or access. The high attrition rate clearly biases these findings because it is unknown if the results found to date only represent a unique subset of PLWH who completed the study’s requirements.

Another limitation is the quality of the reviewed studies. Six^{21,26,43,44,62,68} of these studies did not have a control group and two studies^50,51 did not perform random assignment. These potential threats to internal validity of the experiments make it difficult to draw inferences from the results or link the outcome to the intervention.⁷⁰ For instance, during the course of the intervention, a change in an unobserved variable such as progression of the HIV disease could have confounded the results. Also, lack of randomization can have detrimental effects on the outcome. Without random assignments, it is not clear if the groups were different in regards to the confounding variables related to physical activity behavior (e.g. self-efficacy) and, therefore, responded differently to the intervention.^70,71 Moreover, the contact time between groups was not matched in five studies,^{23,25,49,63,64} and four studies^44,47,49,50 utilized group exercise; while another used group nutrition classes.⁶² Given the well-known effects of social support on mental health,^72,73 it is not clear if the positive effects shown were an independent effect of the exercise, a consequence of social support by other participants, or a combination of those effects.

The current studies are also limited by the sampled population characteristics. Participants were aged between 30.5 to 46.2 years and mostly men living with HIV. In fact, only four studies^45,61,62,64 were on average mostly women (i.e. 85% women). Six studies^{20,25,44,47,49,52,67} were only men (i.e. 100% men), and the majority of the participants (i.e. >50%) were men in 12 other studies.^{21–24,26,28,43,48,50,51,63,68} Considering that women, transgender individuals, and older adults living with HIV are under-represented in the current research, it is imperative that future trials consider including all genders and age groups. Additionally, many studies did not acknowledge whether or not participants were seeking mental health treatment, or whether there were baseline differences with respect to stages of HIV disease. This is clearly problematic given that poor mental health might be confounded with symptoms of physical illness related to HIV and neuropsychological impairment.⁷⁴

Finally, nearly half of the research included in this review has been conducted more than 10 years ago, during which important gains have been made in the advancement of HIV treatment and care. It is unlikely that the results of a study examining the psychological benefits of exercise more than 10 years ago would be easily replicated now. More importantly, these studies may represent a different sample of PLWH compromising their generalizability to the current population of PLWH.

Recommendations for future research

The current body of work suffers from several consistent methodological limitations. Despite this, there is a clear rationale to continue to examine exercise as a tool to help with mental health symptom management (e.g. depressive symptoms, anxiety, and components of QOL) for PLWH, as most of the studies show significant improvement in these symptoms. Importantly, one recent study with seronegative individuals living with depression reported that the majority had an interest in participating in an exercise program for symptom management and mood enhancement.⁷⁵ Although these preferences need to be reexamined with seropositive individuals, it seems that an individualized exercise program might be a viable approach for improving mental health.

Future studies should use intent-to-treat analysis in order to include all the randomized participants in the final analysis,⁷⁶ and help with deeper understanding of the effectiveness of exercise interventions.^76,77 The field can also benefit from future meta-analyses to help detect the bias in the current research and increase the precision of the results specifically with inconclusive or underpowered trials.^78,79 Also, researchers should actively seek out and include more diverse populations such as women, transgender individuals, and older adults and account for confounding variables such as stages of HIV disease, baseline psychological profiles of the participants, and/or social support. Future research could also consider using activity monitors to minimize the burden of access to the research site for the participants. Finally, more adequately powered randomized clinical trials are required to fully understand the acute and long-term effects of exercise dose (e.g. intensity, frequency, volume) on mental health outcomes.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

UNAIDS. Global AIDS update 2016, http://www.unaids.org/en/resources/documents/2016/Global-AIDS-update-2016.

Bhaskaran

Hamouda

Sannes

et al.

Changes in the risk of death after HIV seroconversion compared with mortality in the general population. JAMA 2008; 300: 51–59.

May

Gompels

Delpech

et al.

Impact on life expectancy of HIV-1 positive individuals of CD4+ cell count and viral load response to antiretroviral therapy. AIDS (London, England) 2014; 28: 1193–1202.

Attia

Egger

Muller

et al.

Sexual transmission of HIV according to viral load and antiretroviral therapy: Systematic review and meta-analysis. AIDS (London, England) 2009; 23: 1397–1404.

Donnell

Baeten

Kiarie

et al.

Heterosexual HIV-1 transmission after initiation of antiretroviral therapy: A prospective cohort analysis. Lancet 2010; 375: 2092–2098.

Chander

Himelhoch

Moore

. Substance abuse and psychiatric disorders in HIV-positive patients: Epidemiology and impact on antiretroviral therapy. Drugs 2006; 66: 769–789.

Bing

Burnam

Longshore

et al.

Psychiatric disorders and drug use among human immunodeficiency virus-infected adults in the United States. Arch Gen Psychiatry 2001; 58: 721–728.

Blair

Fagan

Frazier

et al.

Behavioral and clinical characteristics of persons receiving medical care for HIV infection - Medical Monitoring Project, United States, 2009. MMWR Suppl 2014; 63: 1–22.

Rosenberg

Sullivan

et al.

Excess burden of depression among HIV-infected persons receiving medical care in the united states: Data from the medical monitoring project and the behavioral risk factor surveillance system. PLoS One 2014; 9: e92842–e92842.

10.

Ramirez-Avila

Regan

Giddy

et al.

Depressive symptoms and their impact on health-seeking behaviors in newly-diagnosed HIV-infected patients in Durban, South Africa. AIDS Behav 2012; 16: 2226–2235.

11.

L'Akoa

Noubiap

Fang

et al.

Prevalence and correlates of depressive symptoms in HIV-positive patients: A cross-sectional study among newly diagnosed patients in Yaounde, Cameroon. BMC Psychiatry 2013; 13: 228–228.

12.

Ickovics

Milan

Boland

et al.

Psychological resources protect health: 5-year survival and immune function among HIV-infected women from four US cities. AIDS (London, England) 2006; 20: 1851–1860.

13.

Leserman

. Role of depression, stress, and trauma in HIV disease progression. Psychosom Med 2008; 70: 539–545.

14.

Sin

DiMatteo

. Depression treatment enhances adherence to antiretroviral therapy: A meta-analysis. Ann Behav Med 2014; 47: 259–269.

15.

O'Brien

Tynan

Nixon

et al.

Effectiveness of aerobic exercise for adults living with HIV: Systematic review and meta-analysis using the Cochrane Collaboration protocol. BMC Infect Dis 2016; 16: 182–182.

16.

O'Brien

Tynan

Nixon

et al.

Effects of progressive resistive exercise in adults living with HIV/AIDS: Systematic review and meta-analysis of randomized trials. AIDS Care 2008; 20: 631–653.

17.

Arksey

O'Malley

. Scoping studies: Towards a methodological framework. Int J Soc Res Methodol 2005; 8: 19–32.

18.

Hutton

Salanti

Caldwell

et al.

The PRISMA extension statement for reporting of systematic reviews incorporating network meta-analyses of health care interventions: Checklist and explanations. Ann Intern Med 2015; 162: 777–784.

19.

Liberati

Altman

Tetzlaff

et al.

The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: Explanation and elaboration. J Clin Epidemiol 2009; 62: e1–34.

20.

LaPerriere

Antoni

Schneiderman

et al.

Exercise intervention attenuates emotional distress and natural killer cell decrements following notification of positive serologic status for HIV-1. Biofeedback Self-regul 1990; 15: 229–242.

21.

MacArthur

Levine

Birk

. Supervised exercise training improves cardiopulmonary fitness in HIV-infected persons. Med Sci Sports Exer 1993; 25: 684–688.

22.

Neidig

Smith

Brashers

. Aerobic exercise training for depressive symptom management in adults living with HIV infection. J Assoc Nurses AIDS Care 2003; 14: 30–40.

23.

Baigis

Korniewicz

Chase

et al.

Effectiveness of a home-based exercise intervention for HIV-infected adults: A randomized trial. J Assoc Nurses AIDS Care 2002; 13: 33–45.

24.

Stringer

Berezovskaya

O'Brien

et al.

The effect of exercise training on aerobic fitness, immune indices, and quality of life in HIV+patients. Medicine & Science in Sports & Exercise 1998; 30: 11–6.

25.

Lox

McAuley

Tucker

. Exercise as an intervention for enhancing subjective well-being in an HIV-I population. J Sport Exer Psychol 1995; 17: 345–362.

26.

Terry

Sprinz

Ribeiro

. Moderate and high intensity exercise training in HIV-1 seropositive individuals: A randomized trial. Int J Sports Med 1999; 20: 142–146.

27.

Mutimura

Stewart

Crowther

et al.

The effects of exercise training on quality of life in HAART-treated HIV-positive Rwandan subjects with body fat redistribution. Qual Life Res 2008; 17: 377–385.

28.

Maharaj

Chetty

. Rehabilitation program for the quality of life for individuals on highly active antiretroviral therapy in KwaZulu-Natal, South Africa: A short report. Int J Rehab Res 2011; 34: 360–365.

29.

NcNair

Lorr

Droppleman

. EITS manual for the profile of mood states, San Diego, CA: Educational and Industrial Testing Service, 1971.

30.

Goldberg

. The detection of psychiatric illness by questionnaire: A technique for the identification and assessment of non-psychotic psychiatric illness, Oxford: Oxford University Press, 1972.

31.

Montgomery

Asberg

. A new depression scale designed to be sensitive to change. Br J Psychiatry 1979; 134: 382–389.

32.

Beck

Steer

. BDI, Beck depression inventory: Manual, San Antonio, TX: Psychological Corporation, 1987.

33.

Radloff

. The CES-D scale a self-report depression scale for research in the general population. Appl Psychol Meas 1977; 1: 385–401.

34.

Watson

Clark

Tellegen

. Development and validation of brief measures of positive and negative affect: The PANAS scales. J Pers Soc Psychol 1988; 54: 1063–1070.

35.

Diener

Emmons

Larsen

et al.

The satisfaction with life scale. J Pers Assess 1985; 49: 71–75.

36.

Ryckman

Robbins

Thornton

et al.

Development and validation of a physical self-efficacy scale. J Pers Soc Psychol 1982; 42: 891–900.

37.

Spiker

. Quality of life and pharmacoeconomics in clinical trials, New York: Lippencott Raven Publishers, 1996.

38.

Bergner

Bobbitt

Carter

et al.

The Sickness Impact Profile: Development and final revision of a health status measure. Med Care 1981; 19: 787–805.

39.

Revicki

Jacobson

et al.

Evidence for reliability, validity and usefulness of the Medical Outcomes Study HIV Health Survey (MOS-HIV). Qual Life Res 1997; 6: 481–493.

40.

Hlatky

Boineau

Higginbotham

et al.

A brief self-administered questionnaire to determine functional capacity (the Duke Activity Status Index). Am J Cardiol 1989; 64: 651–654.

41.

Hsiung

Fang

Chang

et al.

Comparison of WHOQOL-bREF and SF-36 in patients with HIV infection. Qual Life Res 2005; 14: 141–150.

42.

McHorney

Ware

Jr Lu

et al.

The MOS 36-item Short-Form Health Survey (SF-36): III. Tests of data quality, scaling assumptions, and reliability across diverse patient groups. Med Care 1994; 32: 40–66.

43.

De Medeiros Guerra

Galvao De Souza

Mesquita Soares

et al.

Resisted exercise, morphological and functional standards, and quality of life of people living with HIV/AIDS. J Sports Med Phys Fitness 2016; 56: 470–475.

44.

Hamid

MSA

Rajah

SJJ

. Effects of progressive resistance exercises on quality of life, body composition and muscle strength in patients with HIV. Sains Malaysiana 2015; 44: 1473–1479.

45.

Mkandla

Myezwa

Musenge

. The effects of progressive-resisted exercises on muscle strength and health-related quality of life in persons with HIV-related poly-neuropathy in Zimbabwe. AIDS Care 2016; 28: 639–643.

46.

Jelsma

Mhundwa

De Weerdt

et al.

The reliability of the Shona version of the EQ-5D. Central African J Med 2001; 47: 8–13.

47.

Galantino

Shepard

Krafft

et al.

The effect of group aerobic exercise and t'ai chi on functional outcomes and quality of life for persons living with acquired immunodeficiency syndrome. J Altern Complement Med 2005; 11: 1085–1092.

48.

Jaggers

Hand

Dudgeon

et al.

Aerobic and resistance training improves mood state among adults living with HIV. Int J Sports Med 2015; 36: 175–181.

49.

Fillipas

Oldmeadow

Bailey

et al.

A six-month, supervised, aerobic and resistance exercise program improves self-efficacy in people with human immunodeficiency virus: A randomised controlled trial. Aust J Physiother 2006; 52: 185–190.

50.

Rojas

Schlicht

Hautzinger

. Effects of exercise training on quality of life, psychological well-being, immune status, and cardiopulmonary fitness in an HIV-1 positive population. J Sport Exer Psychol 2003; 25: 440–455.

51.

Gomes

Borges

Lima

et al.

Effects of physical exercise in the perception of life satisfaction and immunological function in HIV-infected patients: Non-randomized clinical trial. Rev Bras Fisioter 2010; 14: 390–395.

52.

Perez-Moreno

Camara-Sanchez

Tremblay

Riera-Rubio

Gil-Paisan

Lucia

. Benefits of exercise training in Spanish prison inmates. International journal of sports medicine 2007; 28: 1046–52.

53.

McNair

Lorr

Droppleman

. Manual: profile of mood states, San Diego, CA: Educational and Industrial Testing Service, 1971.

54.

Rojas OML. Construcción de un test para medir la calidad de vida relacionada con la salud en drogodependientes: aplicación de un modelo politómico de la teoría de respuesta al ítem. 2005.

55.

Paloutzian

Ellison

Loneliness: a sourcebook of current theory, research, and therapy. In: Peplau

Perlman

(eds). Wiley series on personality processes, New York: Wiley, 1982, pp. xvii–xvii. 430p.

56.

Schwarzer R and Jerusalem M. The General Self Efficacy Scale (GSE) English language version. 2006.

57.

Murdaugh C and Walther S. Development of a scale to measure symptom frequency and distress in persons with HIV. AIDS (London, England). 1995.

58.

Cohen

Kamarck

Mermelstein

. A global measure of perceived stress. J Health Soc Behav 1983; 24: 385–96.

59.

Derogatis L. Administration, scoring and procedures manual–I for the R (evised) version. Baltimore MD: John Hopkins University School of Medicine, Clinical Psychometric Unit. 1977.

60.

Neugarten

Havighurst

Tobin

. The Measurement of Life Satisfaction. J Gerontol 1961; 16: 134–43.

61.

Agin

Gallagher

Wang

Heymsfield

Pierson

Jr. Kotler

. Effects of whey protein and resistance exercise on body cell mass, muscle strength, and quality of life in women with HIV. AIDS (London, England) 2001; 15: 2431–40.

62.

Engelson

Agin

Kenya

et al.

Body composition and metabolic effects of a diet and exercise weight loss regimen on obese, HIV-infected women. Metabolism: clinical and experimental 2006; 55: 1327–36.

63.

Ogalha

Luz

Sampaio

et al.

A randomized, clinical trial to evaluate the impact of regular physical activity on the quality of life, body morphology and metabolic parameters of patients with AIDS in Salvador, Brazil. Journal of acquired immune deficiency syndromes (1999) 2011; 57(Suppl 3): S179–85.

64.

Aweto

Aiyegbusi

Ugonabo

Adeyemo

. Effects of Aerobic Exercise on the Pulmonary Functions, Respiratory Symptoms and Psychological Status of People Living With HIV. Journal of research in health sciences 2016; 16: 17–21.

65.

Derogatis

Melisaratos

. The Brief Symptom Inventory: an introductory report. Psychol Med 1983; 13: 595–605.

66.

Antonovsky A. Unraveling the mystery of health: How people manage stress and stay well. Jossey-Bass, 1987.

67.

Bhasin

Storer

Javanbakht

et al.

Testosterone replacement and resistance exercise in HIV-infected men with weight loss and low testosterone levels. JAMA: the journal of the American Medical Association 2000; 283: 763–70.

68.

Shevitz

Wilson

McDermott

et al.

A comparison of the clinical and cost-effectiveness of 3 intervention strategies for AIDS wasting. Journal of acquired immune deficiency syndromes (1999) 2005; 38: 399–406.

69.

Ekkekakis

. Honey, I shrunk the pooled SMD! Guide to critical appraisal of systematic reviews and meta-analyses using the Cochrane review on exercise for depression as example. Mental health and physical activity 2015; 8: 21–36.

70.

Campbell DT and Stanley JC. Experimental and quasi-experimental designs for research. Ravenio Books, 2015.

71.

Bauman

Sallis

Dzewaltowski

et al.

Toward a better understanding of the influences on physical activity: The role of determinants, correlates, causal variables, mediators, moderators, and confounders. Am J Prev Med 2002; 23: 5–14.

72.

Hughes

Jaremka

Alfano

et al.

Social support predicts inflammation, pain, and depressive symptoms: longitudinal relationships among breast cancer survivors. Psychoneuroendocrinology 2014; 42: 38–44.

73.

Eisenberger

Taylor

Gable

et al.

Neural pathways link social support to attenuated neuroendocrine stress responses. NeuroImage 2007; 35: 1601–1612.

74.

Drebing

Van Gorp

Hinkin

et al.

Confounding factors in the measurement of depression in HIV. J Pers Assess 1994; 62: 68–83.

75.

Busch

Ciccolo

Puspitasari

et al.

Preferences for exercise as a treatment for depression. Mental Health Phys Activ 2016; 10: 68–72.

76.

Ten Have

Normand

S-LT

Marcus

et al.

Intent-to-treat vs. non-intent-to-treat analyses under treatment non-adherence in mental health randomized trials. Psychiatr Ann 2008; 38: 772–783.

77.

Brown

Wang

Kellam

et al.

Methods for testing theory and evaluating impact in randomized field trials: Intent-to-treat analyses for integrating the perspectives of person, place, and time. Drug Alcohol Depend 2008; 95: S74–S104.

78.

Chalmers

Hedges

Cooper

. A brief history of research synthesis. Eval Health Prof 2002; 25: 12–37.

79.

Ioannidis

Lau

. Pooling research results: Benefits and limitations of meta-analysis. Jt Comm J Qual Improv 1999; 25: 462–469.