Supervised Physical Activity and Improved Functional Capacity among Adults Living with HIV: A Systematic Review

Abstract

Physical activity (PA) combats the effects of multimorbidity and antiretroviral therapy in people living with HIV (PLWH), but PLWH often don’t meet recommended PA guidelines. The purpose of our review was to investigate whether supervised PA improved functional capacity in PLWH. Preferred Reporting Items for Systematic Reviews and Meta-Analyses were followed. Five databases were searched for randomized controlled trials in English, with participants ages 18 years and older, and a supervised PA intervention. A database search yielded 8,267 articles, with 15 eligible for review inclusion. We found a low risk of bias within and across studies. Combined aerobic/progressive resistance training (PRT) improved strength, cardiovascular, and flexibility outcomes; aerobic interventions alone showed no significant improvements; PRT improved strength outcomes; yoga or yoga/meditation showed no outcome differences; and t’ai chi showed cardiovascular and flexibility improvements. We found that supervised PA increased functional capacity in PLWH and that self-report was not a reliable assessment.

There are currently 1.1 million people living with HIV (PLWH) in the United States (Centers for Disease Control and Prevention [CDC], 2017). Due to advances in antiretroviral therapy (ART), PLWH are living longer (Antiretroviral Therapy Cohort Collaboration, 2008, Boyd, 2009) and HIV is now managed as a chronic disease in much of the developed world (Deeks, Lewin, & Havlir, 2013). Despite the steady increase in life expectancy (Samji et al., 2013), non-HIV-related mortality, specifically, cardiovascular disease, has eclipsed HIV-related mortality as the major cause of death among PLWH (Farahani, Mulinder, Farahani, & Marlink, 2017).

Evidence has suggested that physical activity (PA) is safe and beneficial in promoting the health and medical stability of PLWH (Hand et al., 2009, Jaggers, 2018, Nixon et al., 2005, O’Brien et al., 2016, Yahiaoui et al., 2012). Specifically, PA can effectively mitigate the effects of cardiovascular disease (Hand et al., 2009, Kamitani et al., 2017, O’Brien et al., 2016), the leading cause of death among PLWH (Farahani et al., 2017), and symptoms of long-term HIV exposure and long-term ART (O’Brien et al., 2016, Webel et al., 2015, Yahiaoui et al., 2012). For example, PLWH who participate in PA have a reduced risk for heart disease, increased energy, improved regulation of bowel function, improved sleep, and lower stress (Hand et al., 2009, U.S. Department of Veterans Affairs, 2015, Yahiaoui et al., 2012). PA is also an important strategy for improving aerobic capacity, muscle strength, and flexibility in this population (Haskell et al., 2007, Jones and Carter, 2000, O’Brien et al., 2016, Poton et al., 2016).

The CDC (2015) defined PA as any expenditure of energy by skeletal muscles to produce any bodily movement. PA is a broader concept than exercise, as exercise, a subset of PA, is defined as a planned, structured, and repetitive activity (Caspersen, Powell, & Christenson, 1985). Recommended PA for adult PLWH includes 20 to 40 minutes of combined aerobic and resistance exercise at least three times per week (O’Brien et al., 2016, Yahiaoui et al., 2012). These recommendations were shown to reduce the severity of comorbidities (Dirajlal-Fargo et al., 2016, Kamitani et al., 2017, Yahiaoui et al., 2012) and reduce inflammation associated with long-term HIV and long-term ART (d’Ettorre et al., 2014, Hand et al., 2009, Webel et al., 2015). Given the benefits of PA, regular participation for PLWH is essential; however, their PA participation has remained below recommended levels (Montoya et al., 2015, Simonik et al., 2016, Vancampfort et al., 2016, Webel et al., 2015). Weekly PA recommendations make no mention of supervision, despite being based on studies of both supervised and unsupervised PA interventions (O’Brien et al., 2016, Yahiaoui et al., 2012).

Supervised PA has the potential to encourage PLWH to participate in PA regularly. Studies of adults with other chronic conditions (e.g., arthritis, obesity, kidney failure patients on dialysis, heart failure, diabetes, and cancer) were shown to have increased PA participation after involvement in a PA intervention supervised by a clinician or allied health professional (Akbaba et al., 2016, Casla et al., 2014, Colak et al., 2017, Daul et al., 2004, Klempfner et al., 2015, Negri et al., 2010, Nicolai et al., 2009). In some cases, regular participation in PA continued well beyond the end of the study (Azad et al., 2012, Casla et al., 2014, Trinh et al., 2014).

Additionally, when outcomes were compared between supervised and unsupervised PA interventions, more pronounced improvements were seen as a result of supervised PA. To detail, participants of supervised PA interventions showed significant improvement in cardiovascular fitness, muscle strength, and significantly decreased fat mass, body mass index (BMI), and body weight, compared with controls (Akbaba et al., 2016, Boshuizen et al., 2005, Dalager et al., 2015, Nicolai et al., 2009, Rossomanno et al., 2012). However, the evidence to support the benefits of supervised PA as the main intervention for PLWH is missing from the current body of literature. Past systematic reviews have examined the effects of both supervised and unsupervised PA interventions for PLWH and reported aggregate results (Gomes-Neto et al., 2015, Gomes-Neto et al., 2013b, Jaggers, 2018, O’Brien et al., 2016, Poton et al., 2016, Yahiaoui et al., 2012). None have segregated supervised from unsupervised interventions in their reviews or analyzed the outcomes of these respective interventions separately. The purpose of our review was to investigate the impact of supervised PA interventions on functional capacity among adults living with HIV.

Methods

Search Strategy

Our review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (Moher, Liberati, Tetzlaff, & Altman, 2009). A total of five databases were systematically searched for all relevant literature, including PubMed, Cumulative Index to Nursing and Allied Heath Literature (CINAHL), PsycINFO, Embase, and Physical Education Index. First, search terms were created based on the population and interventions of interest in PubMed. Second, comparable search terms were identified in each subsequent database. For example, a combination of MeSH terms, MeSH headings, keywords and phrases, and Boolean operators were used to create the search terms used in PubMed. The resultant search terms in PubMed were as follows: (“HIV”[Mesh] OR HIV [Text Word] OR “HIV Long-Term Survivors”[Mesh] OR “HIV Long-Term Survivors” OR “HIV Infections”[Mesh] OR “HIV Infections” OR “HIV Seropositivity”[Mesh] OR “HIV Seropositivity” OR “HIV Seroprevalence”[Mesh] OR “HIV Seroprevalence”) AND (“Physical Activity”[Mesh] OR “Physical Activity” OR “Exercise”[Mesh] OR “Exercise” OR “Physical Exercise”[Mesh] OR “Physical Exercise” OR “Aerobic Exercise”[Mesh] OR “Aerobic Exercise” OR “Exercise Therapy”[Mesh] OR “Exercise Therapy” OR “Exercise Training”[Mesh] OR “Exercise Training” OR “Motor Activity” [Mesh] OR “Motor Activity”). All possible combinations of MeSH terms, MeSH headings, keywords, and phrases were used to identify articles for inclusion.

Eligibility Criteria

Study inclusion criteria included studies that (a) were randomized controlled trials that evaluated the effect of a supervised PA intervention as compared with usual care, (b) examined functional capacity as an outcome, (c) included adults ages 18 years and older living with HIV, and (d) were available as full text articles in English. Supervised PA was defined in our review as PA interventions that were supervised by a health care or allied health professional. Functional capacity was defined as per the American Heart Association definition, as the integrated efforts and health of an individual’s pulmonary, cardiovascular, and skeletal muscles to perform activities of daily living (Arena et al., 2007). Functional capacity outcomes in our review included strength measures, cardiovascular fitness measures, flexibility measures, and self-reported measures of functional capacity. We excluded poster sessions, presentations, protocols, letters, comments, editorials, correspondence, or gray literature (e.g., blogs, newsletters, videos).

Study Selection and Data Extraction

We used Covidence (Veritas Health Innovation, 2018), a web-based program designed to support the systematic review process, to facilitate screening at the title/abstract and full text level, as well as for data extraction, adjudication of disagreements, and confirmation of data. A hand search was conducted for additional studies meeting inclusion criteria. Two authors independently assessed eligibility, and a third consulted in instances of uncertainty/disagreement. The following information was extracted from the final studies: name of the first author, publication year, locations of the studies, study sample size, type of supervised PA interventions, frequency/duration of interventions, and duration of the studies. No publication date or restrictions were used in the search.

Quality Assessment

The quality appraisal was conducted using the Cochrane Collaboration Risk of Bias Tool (CCRBT; Higgins et al., 2011). The CCRBT describes six domains to assess possible avenues of bias: sequence generation; allocation concealment; blinding of participants, personnel, and outcome assessors; incomplete outcome data; selective outcome reporting; and, other sources of bias. Each domain is evaluated as yes (low risk of bias), no (high risk of bias), or unclear (uncertain risk of bias). Each study is then assigned an overall grade of high, moderate, or low risk of bias. Two authors independently appraised the quality of the studies, and discrepancies were reviewed until consensus was reached.

Results

Study Selection

The database literature search yielded 8,267 articles; 126 articles were identified from the hand search. One hundred ninety-five articles remained for full text screening after duplicates were removed and title and abstract screening was completed; 180 articles were excluded based on the full text review. Of the 180 excluded articles, full text for 28 abstracts were not found despite independent searches, the assistance of university librarians, and requests from other institutions via interlibrary loan. Fifteen studies remained and were included in this systematic review. Figure 1 summarizes the results of the search and study selection.

Figure 1.

Flow diagram of study selection.

Study Characteristics

The combined studies had 537 participants whose ages ranged between 18 and 73 years. The supervisors of the PA interventions included certified yoga instructors, exercise physiologists, nurse trainers, physical therapists, exercise specialists, physiotherapists, cardiologists, and personal trainers. Table 1 summarizes the characteristics of the included studies. The duration of the PA interventions ranged between 31 minutes and 2 hours, and the intensity ranged from low to vigorous. The length of the PA interventions ranged from 6 weeks to 6 months across studies.

Table 1.

Characteristics of the Physical Activity Interventions of Included Studies (n = 15)

Study and Sample Size	Country	Intervention	Frequency (per Week)	Session Duration	Study Length (Weeks)
Agarwal et al., 2015 (n = 24)	USA	Yoga & meditation	2	60 min	8
Agin et al., 2001 (n = 30)	USA	PRT	3	3 sets, 8-10 repetitions per muscle group	14
Baigis et al., 2002 (n = 99)	USA	Aerobics	3	40 min	15
Cade et al., 2010 (n = 50)	USA	Yoga	2-3	60 min	20
Dolan et al., 2006 (n = 38)	USA	Aerobics & PRT	3	120 min	16
Driscoll et al., 2004 (n = 25)	USA	Aerobics & PRT	3	Weeks 1-2: 20 min aerobics + PRT Weeks 3-12: 30 min aerobics + PRT PRT: 3 sets, 10 repetitions per muscle group	12
Dudgeon et al., 2012 (n = 26)	USA	Aerobics & PRT	2	Aerobics: 30 min PRT: 1 set, 12 repetitions per muscle group	6
Farinatti et al., 2010 (n = 27)	Brazil	Aerobics & PRT	3	90 min	12
Fillipas et al., 2006 (n = 35)	Australia	Aerobics & PRT	2	60 min	24
Galantino et al., 2005 (n = 38)	USA	Aerobics & T’ai Chi	4	60 min	8
McDermott et al., 2016 (n = 11)	Ireland	Aerobics	3	31-52 min	16
Ogalha et al., 2011 (n = 63)	Brazil	Aerobics & PRT	3	60 min	24
Perez-Moreno et al., 2007 (n = 19)	Spain	Aerobics & PRT	3	90 min	16
Strawford et al., 1999 (n = 22)	USA	PRT	3	60 min	8
Terry et al., 2006 (n = 30)	Brazil	Aerobics	3	60 min	12

Note. PRT = progressive resistance training.

Study Findings

Supervised PA interventions were classified into the following categories: (a) aerobics, (b) progressive resistance training (PRT), (c) combined aerobics and PRT, (d) yoga and meditation, and (e) t’ai chi. Each intervention in the review was assessed as a change in one or a combination of four functional capacity outcomes of interest: cardiovascular (e.g., VO₂max [maximal aerobic capacity]; BMI; the 6-Minute Walk Test), strength (e.g., the one-repetition maximum, 6-repetition maximum), flexibility and balance (i.e., Sit and Reach and Forward Reach tests), and self-report (e.g., SF-36 Physical Functioning and Role Limitations due to Physical Health subscale scores, exit interview questions, health diary). Table 2 describes the functional capacity outcomes of each study.

Table 2.

Functional Capacity Outcomes of Intervention Groups in the Included Studies (n = 15)

Type of Supervised PA	Study (Year)	Functional Capacity Measure	Functional Capacity Result
Aerobics	Baigis (2002)	VO2max Diary, DASI, MOS-HIV, SIP	Ø Ø
	Galantino (2005)	Sit-up Stair climbing	↑ ↑
	McDermott (2016)	VO2max, treadmill time, light/moderate/vigorous PA	Ø
	Terry (2006)	Peak HR, BMI, BP	Ø
Progressive Resistance Training	Agin (2001)	1-RM SF-36	↑ Ø
	Strawford (1999)	1-RM	↑
Aerobics + Progressive Resistance Training	Dolan (2006)	VO2max, Bike time, 6MWT Change in strength Exit interview questions	↑ ↑ ↑
	Driscoll (2004)	BP, Exercise time Change in strength 1& 2	↑ ↑, Ø
	Dudgeon (2012)	Change in strength	↑
	Farinatti (2010)	BMI Heart rate, 12-RM, Sit and reach	Ø ↑
	Fillipas (2006)	Heart rate HRQoL	↑ Ø
	Ogalha (2011)	BMI, Resting HR VO2max, MET Change in muscle mass, SF-36	Ø, ↑ ↑ ↑
	Perez-Moreno (2007)	Workload, Peak HR, Decline HR 6-RM	↑ ↑
Yoga + Meditation	Agarwal (2015)	SF-36	Ø
	Cade (2010)	SF-36	Ø
T’ai Chi	Galantino (2005)	Forward reach Sit and reach	↑

Note. Ø = no significant change in functional capacity; ↑ = significantly improved functional capacity; VO₂max = maximal aerobic capacity; PA = physical activity; BMI = body mass index; DASI = Duke Activity Status Index; MOS-HIV = Medical Outcomes Study, HIV Health Survey; SIP = Sickness Impact Profile; HR = heart rate; BP = blood pressure; 1-RM = 1 repetition maximum; 6-RM = 6 repetition maximum; 12-RM = 12 repetition maximum; SF-36 = Medical Outcomes Study, 36-item Short Form Health Survey; 6MWT = 6-Minute Walk Test; Change in strength 1: leg curl, leg extension, lateral pull down, arm curl, chest press; Change in strength 2: leg press; HRQoL = Health Related Quality of Life; MET = metabolic equivalent.

Aerobics interventions (n = 4)

Four studies employed aerobic interventions in their respective study samples (Baigis et al., 2002, Galantino et al., 2005, McDermott et al., 2016, Terry et al., 2006). Three out of four studies reported no significant differences between intervention and control group participants in cardiovascular and self-reported outcomes (Baigis et al., 2002, McDermott et al., 2016, Terry et al., 2006). However, the participants in the Galantino and colleagues (2005) study showed improvements in both cardiovascular (p < .05) and flexibility outcomes (p < .01).

Progressive resistance training interventions (n = 2)

Agin and colleagues (2001) and Strawford and colleagues (1999) administered PRT as the intervention in their studies. Both studies showed significant improvements in strength outcomes in the intervention group compared to those in the control group (p < .001 and p < .05, respectively).

Combined aerobic and PRT interventions (n = 7)

Seven studies employed combined aerobic and PRT interventions (Dolan et al., 2006, Driscoll et al., 2004, Dudgeon et al., 2012, Farinatti et al., 2010, Fillipas et al., 2006, Ogalha et al., 2011, Perez-Moreno et al., 2007). All strength, cardiovascular, and flexibility outcome measures associated with these interventions demonstrated a significant improvement (p ≤ .05) in the intervention group as compared to control group participants. However, BMI in the Farinatti and colleagues (2010) and Ogalha and colleagues (2011) studies and health-related quality of life in the Fillipas and colleagues (2006) study demonstrated no significant difference between intervention and control groups (p > .05).

Yoga and meditation interventions (n = 2)

One study administered a combined yoga and meditation intervention (Agarwal, Kumar, & Lewis, 2015) and another employed yoga alone (Cade et al., 2010). Both studies measured change in functional capacity via self-reported measures, the SF-36 Physical Functioning and Physical Limitations subscale scores. In both studies, changes in functional capacity were not significantly different from the control group at study end.

T’ai chi intervention (n = 1)

The t’ai chi group showed significant improvements in both flexibility (p < .01) and cardiovascular (p < .05) outcomes (Galantino et al., 2005).

Study Quality

Seven of 15 studies had a low risk of bias (Agin et al., 2001, Baigis et al., 2002, Dolan et al., 2006, Fillipas et al., 2006, Galantino et al., 2005, Perez-Moreno et al., 2007, Strawford et al., 1999). The remaining eight studies had an unclear risk (Agarwal et al., 2015, Cade et al., 2010, Farinatti et al., 2010, McDermott et al., 2016, Terry et al., 2006) or high risk of bias (Driscoll et al., 2004, Dudgeon et al., 2012, Ogalha et al., 2011). For sequence generation, allocation concealment, blinding of participants, personnel, and outcome assessors, the risk of bias was low or unclear risk for all studies. Most of the studies demonstrated low risk of bias for incomplete outcome data. The results of the assessment of study quality using CCRBT are reported in Figure 2, and the Cochrane risk of bias summary is reported in Figure 3.

Figure 2.

Risk of bias within studies.

Figure 3.

Risk of bias across studies.

Discussion

We examined the impact of supervised PA interventions on the functional capacity of adult PLWH. Overall, our findings supported supervised PA interventions as an effective intervention for increasing functional capacity among adult PLWH. Supervised combined aerobics and PRT interventions improved functional capacity outcomes across all of the following outcomes: cardiovascular, strength, flexibility and balance, and self-report. However, studies that assessed functional capacity via self-reported measures alone revealed inconsistent results.

We found that supervised combined aerobics and PRT was the most effective intervention, followed by supervised t’ai chi, PRT, aerobics, yoga, and combined yoga and meditation to increase functional capacity in this population. The participants in all seven studies involving supervised combined aerobics and PRT (Dolan et al., 2006, Driscoll et al., 2004, Dudgeon et al., 2012, Farinatti et al., 2010, Fillipas et al., 2006, Ogalha et al., 2011, Perez-Moreno et al., 2007) showed statistically significant improvements in cardiovascular, strength, and flexibility and balance outcomes. Improvements in cardiovascular health, strength, and flexibility have been associated with reduced cholesterol and risk of heart attack and stroke, increased muscle mass and strengthening of bones, and reduced stress in both PLWH and those without HIV infection (Lachman et al., 2018, U.S. Department of Veterans Affairs, 2015, Yahiaoui et al., 2012). These health benefits are of particular import to PLWH, as the incidence of heart attack, lipodystrophy, and wasting syndrome are high in this population (Deeks et al., 2013b, Farahani et al., 2017, Nguyen et al., 2016, Palmeira dos Santos et al., 2017, Schouten et al., 2014). While yoga is considered a form of PA and has been shown to effectively improve health-related outcomes in both healthy and diseased populations (Govindaraj et al., 2016, Ross and Thomas, 2010), in our review, yoga failed to demonstrate an improvement in functional capacity for PLWH.

Supervised combined aerobics and PRT interventions showed improvements in strength, cardiovascular, and flexibility and balance outcomes; however, cardiovascular outcomes associated with supervised aerobics alone showed no significant changes between groups in three out of four studies (Baigis et al., 2002, McDermott et al., 2016, Terry et al., 2006). In our review, when PRT was included with aerobics, cardiovascular outcomes demonstrated significant improvement in functional capacity. Improved cardiovascular outcomes in the Galantino and colleagues (2005) study may be attributed to three possible reasons: (a) the sample of male-only subjects, whereas the subjects in the other aerobics studies were comprised of male and female subjects; (b) the investigator was not blinded to the participants in the Galantino and colleagues (2005) study, and therefore, introduced a source of bias in the data collection or analysis; and (c) the cardiovascular outcome measures in the Galantino and colleagues (2005) study involved the simplest interactions with equipment, and thus, a reduced likelihood of equipment malfunction or failure, unlike the measures in the Baigis and colleagues (2002), McDermott and colleagues (2016), and Terry and colleagues (2006) studies. For instance, the latter studies used VO₂max, peak heart rate, and blood pressure as cardiovascular outcomes among others. Measuring VO₂max involves a mouth guard, nose clip, and treadmill. Peak heart rate measurements require electrode sensors and an electrocardiogram machine. Blood pressure was measured with an automated sphygmomanometer. Sit-up and stair-climbing tests in the Galantino and colleagues (2005) study involved counts of each within a specified time frame.

Of the studies that employed self-reported measures, (Agarwal et al., 2015, Agin et al., 2001, Baigis et al., 2002, Cade et al., 2010, Dolan et al., 2006, Fillipas et al., 2006, Ogalha et al., 2011), the findings were inconclusive. Self-reported improvement of functional capacity was more likely to be significant when supervised PRT was included in the PA intervention. This finding is supported by evidence that subjective assessments of PA are beneficial only to ranking participants as more or less physically active, but not quantifying PA in terms of time spent in each level of PA (Masse & de Niet, 2012). In other words, reporting bias may have affected participant recall for time spent in PA.

Our findings support evidence that PA interventions are effective at improving cardiovascular fitness and muscular strength in PLWH (Gomes-Neto et al., 2013a, O’Brien et al., 2016, O’Brien et al., 2017, Poton et al., 2016). While other systematic reviews focused on overall PA levels in PLWH (Schuelter-Trevisol et al., 2012, Vancampfort et al., 2016), benefits of PA and PA recommendations (Gomes-Neto et al., 2015, Gomes-Neto et al., 2013a, Gomes-Neto et al., 2013b, Kamitani et al., 2017, O’Brien et al., 2016), or facilitators and barriers of PA (Vancampfort et al., 2017a, Vancampfort et al., 2017b), our review stands apart because of the focus on the supervisory aspect of PA interventions and its potential to improve functional capacity for PLWH. Essentially, our review showed that supervision of PA increases functional capacity in adult PLWH and may be an enabling factor to support regular PA among PLWH.

HIV is now managed as a chronic illness for many PLWH in the United States and the developed world due to the success of ART (Deeks et al., 2013a, Samji et al., 2013). Research has shown that chronically ill persons are less likely to engage in PA (Mansfield et al., 2018, Miravitlles et al., 2014, Volaklis et al., 2018). Studies of supervised PA interventions in HIV-uninfected, chronically ill populations have suggested that incorporation of formal, structured programs may facilitate increased PA participation (Allen and Morey, 2010, Bousquet-Dion et al., 2018), and supervision may play a role in retaining participants in PA interventions (Akbaba et al., 2016, Dalager et al., 2015, Rossomanno et al., 2012). Increases in PA may be due, in part, to the reported improved morale and confidence to engage in PA when a clinician is supervising the activity (Bauman et al., 2012, Colak et al., 2017, Tully et al., 2010). Studies of people living with diabetes, heart failure, kidney failure, cancer, arthritis, and obesity have demonstrated empirical improvement in PA participation where supervised PA was the intervention under inquiry (Akbaba et al., 2016, Azad et al., 2012, Casla et al., 2014, Colak et al., 2017, Daul et al., 2004, Klempfner et al., 2015, Negri et al., 2010, Nicolai et al., 2009, Tully et al., 2010). Therefore, supervised PA interventions can serve as formal, structured programs to increase and maintain regular PA among PLWH.

Recommendations and Implications

Recommendations

The therapeutic alliance between a health professional and patient, an important component in patient engagement (Higgins, Larson, & Schnall, 2017), may increase regular PA in PLWH. For example, the benefit of the therapeutic alliance during a PA intervention was demonstrated in the Galantino and colleagues (2005) study. Those participants reported a newly formed “brotherhood,” working harder during the PA, and that the added presence of a leader facilitated their participation in the PA, where they would not have participated if a leader was not present. A recent systematic review and meta-analysis investigating dropout from PA interventions with PLWH revealed reduced dropout rates involving supervised PA interventions (p < .001) and interventions using qualified professionals (p < .001; Vancampfort et al., 2017a, Vancampfort et al., 2017b). Therefore, supervised PA interventions modeled after the general medicine therapeutic alliance, with the presence of a health care or allied health professional, may increase regular PA participation and ultimately sustained improvement in functional capacity in PLWH.

Nursing implications

Findings from our study present a unique opportunity for nurses who care for PLWH. Because supervised PA has been shown to improve outcomes in PLWH and nurses are the largest sector of health professionals, this provides an opportunity for nurses to supervise PA interventions for PLWH. Moreover, the presence of a health care professional, such as a nurse, during PA may encourage PLWH to participate in and work harder during the PA interventions, improve long-term adherence to PA, and improve motivation and self-efficacy to reach PA goals (Peddle-McIntyre et al., 2013, Trinh et al., 2014, Vancampfort et al., 2017a, Vancampfort et al., 2017b). Nurses can fill that gap by becoming the PA facilitator.

Additionally, our findings present an opportunity for nurse scientists to help fill gaps in the literature. Specifically, the current literature is comprised of studies with small sample sizes and high variability of intervention components (e.g., frequency, session duration, and length) and outcome measures. This variability makes comparison via meta-analysis and drawing succinct conclusions difficult. Further research by nurse scientists with more rigorous and robust research designs will enhance knowledge of the ability of PA interventions to improve health outcomes in PLWH.

Limitations

Inclusion of only English language studies was a limitation because there may be relevant studies published in African languages, given that the burden of the HIV epidemic persists in Africa (World Health Organization, 2017). In our review, none of the included studies took place in an African country. Therefore, the external validity of our findings is limited to the countries represented: Australia, Brazil, Ireland, Spain, and the United States. Additionally, 9 of the 15 studies had samples of 30 or less and all had fewer than 100 participants. Selection bias may be present as the study participants may represent PLWH who were already motivated to take ownership of their health and participate in PA. The small sample sizes in the individual studies also presented the potential for a type 1 error. Inclusion of randomized controlled trials and the inherent randomization, however, elevated the level of evidence presented here and countered the potential for selection bias and type 1 error. The number of articles included in this review may also be a limitation, as 15 articles may be considered a small number of articles to comprise a systematic review. However, PRISMA guidelines were followed and the search strategy and selection criteria were developed a priori. The resulting articles included for review reflected the dearth of studies on this topic. Our systematic review will, nonetheless, make an important contribution to the literature considering increased longevity in PLWH and the benefits of supervised PA interventions on health outcomes in this population.

The heterogeneity of studies, including the length of the interventions, duration of the study, follow-up assessment period, assessment instruments, and assessment modality, precluded a meta-analysis. Like other systematic reviews, our review demonstrated the need for consistent protocols and outcome measures across studies (Gomes-Neto et al., 2013a, Schuelter-Trevisol et al., 2012, Vancampfort et al., 2017a, Vancampfort et al., 2017b). At the individual study level, clinical significance may have been present where statistical significance was absent. For instance, at the end of 16 weeks in the McDermott and colleagues (2016) study, there was a trend toward improvement in cardiovascular fitness that may have revealed statistically significant improvements with a longer follow-up period.

Conclusions

The results of our review indicate that supervised PA interventions are associated with improved functional capacity in adult PLWH and may motivate PLWH to regularly participate in PA. Of all the interventions, the supervised aerobics and PRT interventions demonstrated improvements in strength, cardiovascular, and flexibility and balance outcomes. The inclusion of PRT with aerobics interventions demonstrated improved cardiovascular outcomes compared to aerobics alone. Furthermore, functional capacity assessed by self-report rather than objective measures may not reflect improvements in functional capacity, except where PRT is included in the intervention.

Future studies should consider PA interventions with and without supervision to quantify the benefit of these interventions in this population. A special focus should be placed on the long-term impact and outcomes of supervised PA for elderly PLWH as the average age of PLWH continues to rise.