Effectiveness of exercise interventions on mental health and health-related quality of life in women with polycystic ovary syndrome: a systematic review

Of the 11 included trials, nine were RCTs [22‐24, 31‐33, 35, 39, 41], one was a single arm study [34] and one was a case-control study [36]. Sample sizes ranged from 13 to 149 participants. The mean age of participants ranged from 26 to 33 years of age and the mean body mass index (BMI) ranged from 27.9 to 41.3kg/m². Participants from nine trials were diagnosed with PCOS according to the Rotterdam criteria [22‐24, 31, 33, 34, 36, 39, 41] and three trials according to the NIH criteria [32, 35]. Of the included trials, nine recruited only overweight and obese women with PCOS and all studies excluded women with chronic co-morbidities, such as diabetes and cardiovascular disease. One study included only women with a positive screen for depression symptoms [32]. Three studies excluded women who were taking medication treatment for depression [22, 31, 43], five studies did not exclude women who were taking anti-depressant medication [23, 24, 36, 39, 41] and two studies did not report on use of medication for mental health status [34, 35], as shown in Table 2.

Scores from the modified Downs and Black checklist varied greatly with scores ranging from 7 to 19 on a 21-point scale, with a lower score indicating a poorer methodological quality. The most common issues were poor reporting of aspects of the intervention or the characteristics of participants lost to follow-up. The full assessment of methodological quality is presented in Supplementary Table 3.

Aerobic exercise alone was delivered in eight out of the 11 trials [24, 31‐35, 39, 41]. One study included an aerobic exercise group, a resistance training group and a combined aerobic and a resistance training group [22] and the final two studies delivered resistance training only [23, 36]. Of the included studies, only four of the interventions had full supervision by an exercise specialist [23, 33, 36, 39], three had partial supervision [22, 31, 35] and four had no supervision [24, 32, 34, 41]. The duration of the interventions varied from 12 to 26 weeks, with the number of sessions ranging from two to five per week. Exercise intensity was only adequately reported in five of the 11 included studies, all of which reported using moderate-vigorous intensity exercise. Adherence, classified as the percent of sessions attended in regards to the expected attendance, was reported in only three of the included studies. Two of these reported an average adherence of 81% [33] and 95% [23] for supervised sessions, with one of these trials also reporting an average of 51% for unsupervised session adherence [23]. The third study reported that on average, only 38% of participants in the lifestyle only group reported meeting their weekly exercise goal [32]. No adverse events were reported as a result of any of the exercise interventions.

Six of the included trials had a primary outcome of mental health and/or health-related quality of life. The remaining five trials included mental health and/or health-related quality of life as secondary outcome.

Of the studies included, eight used the PCOSQ to assess PCOS related distress, seven of which reported statistically significant improvements in a minimum of two domains. The PCOSQ is a validated questionnaire which assessed the multidimensional concept of health-related quality of life, encompassing five domains; emotions, hair growth, body weight, infertility and menstrual problems. A higher overall score indicates better function, and a change of 0.5 is considered clinically relevant [44]. Six studies reported statistically significant improvements in emotions [22, 23, 31, 34, 35, 41] weight [22‐24, 31, 34, 35] and infertility domains [22‐24, 31, 34, 41], three reported improvements in the menstrual problems domain [22, 24, 31] and one reported improvements in the body hair domain [34]. The eighth study did not report a statistically significant impact of exercise on PCOS related distress, but did report minimally important clinically significant improvements (≥0.5 point increase), in all domains for the cognitive behavioural therapy plus lifestyle group, and clinically significant increases in the weight and infertility domains for the lifestyle only group [32]. In total, the weight domain had the most clinically significant improvements with six studies reporting improvements ranging from 0.7 to 1.75 point increase in scores. Clinically significant improvements were also commonly reported in the emotions domain, with four studies reporting increases in scores of between 0.5 and 0.8 points. A large proportion of interventions that utilised a range of exercise intensities, doses, types and durations, resulted in significant improvements in multiple domains of the PCOSQ, therefore suggesting that regardless of these factors, exercise can improve health-related quality of life in regards to PCOS symptom distress.

Six studies used SF-36 to determine the health-related quality of life of participants. The SF-36 measures eight dimensions of health; physical functioning, role limitations due to physical problems (role physical) , bodily pain, general health perception, vitality, social functioning, role limitations due to emotional problems (role emotional) and mental health [45]. Four studies reported statistically significant increases in the physical functioning domain [23, 37, 39, 43]. Three studies reported significant increases in general health [38, 39, 43], vitality [23, 38, 39] and mental health [23, 39, 43]. Two studies reported statistically significant improvements in role physical [39, 41], social functioning [23, 39] and role emotional domains [23, 39]. No studies reported statistically significant improvements in the bodily pain domain. Clinically meaningful improvements were most commonly reported for the role physical domain, with 4 studies showing clinically meaningful changes after an exercise intervention [37, 39, 41, 43], reporting increases in scores of 6.3 to 39.3. In regards to the SF-36, a change of score of 5 points or greater is considered clinically meaningful [46]. Improvements were also commonly reported for emotional and mental health domains with increases in scores of between 5.0 to 40.6, and 11.7 to 15.4 respectively. There were three studies that observed large improvements in multiple domains of the SF-36. Two of these studies used an aerobic exercise intervention of moderate to vigorous intensity, delivered three times per week for 16 weeks [39, 43], while the third study conducted a 12 week progressive resistance training program delivered twice per week [23]. In summary, many studies reported improvements in domains of the SF-36 as a result of various exercise interventions, however it appears that there were no common denominators in regards to exercise characteristics.

Eight studies measured the effect of an exercise intervention on depression symptoms, with five reporting significant reductions in depression scores. Two studies that reported improvements delivered an aerobic exercise intervention [32, 40], two delivered resistance training [23, 36] and the final study compared three interventions (diet only, diet and aerobic exercise and diet and combined exercise) [22], all of which resulted in improved depression scores. There were no obvious common denominators in regards to exercise characteristics between studies that did, and those that did not report improvements in symptoms of depression. One study that used the Depression Anxiety and Stress Scales (DASS-21) questionnaire reported significant changes in depression symptoms at post-intervention, in comparison to a no-intervention control group, after 12 weeks of supervised resistance training [23]. One study, reported significant decreases in depression symptoms following a combined diet and aerobic intervention, with average scores on the CES-D decreasing from 18.6 at baseline to 14.0 post-intervention [22]. A second study that used the CES-D included only women classified as possibly depressed (score of ≥16) at baseline, and reported large decreases in symptoms with average scores decreasing from 24 to 18 [32]. Lastly, both studies that used the HADS questionnaire reported significant improvements in depression scores from baseline, one of which utilised a resistance training intervention [36] and the second study utilised two different aerobic training programs, both of which reported significant improvements [39].

Six studies also examined symptoms of anxiety, three of which reported statistically significant within group reductions in symptoms after an exercise intervention [36, 38, 40]. A third study reported significant reductions in anxiety symptoms compared to a control group with scores reducing from 10.3 at baseline to 7.4 post-intervention [23]. Two studies reported the prevalence of anxiety at baseline and following an exercise intervention. One of these studies reported that 44.1% of participants had anxiety (according to the measures used) at baseline, which decreased to 23.2% after a 16 week resistance training intervention [36]. The second study reported a drop from 15.9 to 4.7% of women considered to have clinical anxiety after a 16 week aerobic exercise intervention [38].

In summary, exercise interventions reduced symptoms of depression and anxiety in half of the reviewed studies. Due to the large variety of training interventions, conclusions cannot be made regarding the impact of a specific type or intensity of an exercise intervention, compared to another.

This review builds on existing knowledge and provides preliminary data to support the inclusion of exercise to manage and improve mental health and health-related quality of life outcomes in women with PCOS. A strength of this review is that it follows the PRISMA guidelines, including double screening of articles, data extraction and quality appraisal of each publication. This review was limited by large variations and poor reporting of exercise characteristics in the included primary studies. This hindered us from being able to conduct a quantitative synthesis of results and limited our ability to form strong conclusions about the effectiveness of exercise, and particular exercise characteristics, on mental health and health-related quality of life in women with PCOS. Future studies should endeavour to adequately report all intervention characteristics, including frequency, intensity, type, format and session duration of exercise interventions and as well as reporting both adherence and compliance to the exercise intervention inform future research.