Outcomes of the effect of exercise on cognitive function in patients with depression: A meta-analysis

Abstract

BACKGROUND

Major depressive disorder is a common mental disorder, characterized by a high rate of suicide and recurrence, which is frequently accompanied by cognitive impairments, particularly in executive function, memory, attention, and information processing speed. As such, improving the cognitive function in patients with depression and enhancing their quality of life are urgent issues.

AIM

To perform a systematic review and meta-analysis of the effects of exercise on cognitive function in patients with depression.

METHODS

The PubMed, Web of Science, Cochrane Library, Embase, China National Knowledge Infrastructure, Wanfang Medical, VIP, and Biomedical Databases for randomized controlled trials (RCTs) databases were searched (from inception to October 19, 2023) for studies investigating improvements in cognitive function in patients with depression through exercise. Tools recommended by the Cochrane Handbook for RCT evaluation, and GRADEpro and Stata17 software, were employed for risk of bias assessment, evidence grading, forest plot construction, subgroup and sensitivity analyses, and assessment of publication bias.

RESULTS

Seventeen RCTs (1173 patients with depression) were included. Exercise had a small but significant positive effect on attention, with an effect size of 0.21, 95%CI: 0.07-0.34, P < 0.01. Specifically, aerobic exercise regimens of 30-60 minute/session, thrice a week, at moderate intensity, and sustained over 3-12 weeks, were associated with the most pronounced benefits (P < 0.05), with effect sizes for executive function, memory, and information processing speed of 0.11, 95%CI: -0.11-0.32, P = 0.34; 0.08, 95%CI: 0.00-0.16, P = 0.05; and 0.14, 95%CI: 0.04-0.25, P = 0.01, respectively. The evidence levels for attention, information processing speed, and memory were rated as 'low,’ whereas that for executive function was rated as 'very low’.

CONCLUSION

Exercise could improve attention and information-processing speed in patients with depression, although improvements in executive function and memory are not significant.

Key Words: Exercise; Depression; Cognitive function; Patients; Meta-analysis

Core Tip: Herein, we conducted a meta-analysis showed that exercise exerted an improvement effect on executive function in patients with depression. Indicators of cognitive function, including attention, memory, and information processing speed, were added, seeking to provide evidence-based support for clinical practice and a theoretical reference for researchers.

INTRODUCTION

Major depressive disorder is a common mental disorder, characterized by high rates of suicide and recurrence. Depression currently affects approximately 260 million individuals globally, with projections suggesting that depression may become the leading cause of disease burden by 2030. In China, depression in adults has a lifetime prevalence of 3.4%, with a rate of 20% among adolescents[1,2]. Individuals with depression typically experience a persistently low mood and lack of interest, with severe cases presenting with suicidal tendencies. The risk of suicide in patients with depression is 20 times higher than that in the general population[3], and is frequently accompanied by cognitive impairments, particularly in executive function, memory, attention, and information processing speed[4,5]. As such, improving cognitive function and enhancing quality of life are urgent issues in patients with depression.

Exercise, which has been recognized in the treatment guidelines of many countries, offers promise for ameliorating the cognitive impairments associated with depression[6,7]. Previous studies have supported the notion that regular exercise can mitigate cognitive impairment in depression, potentially through the modulation of neurotrophic factors in the hippocampus, increased catecholamine levels, and enhanced neurological arousal[8,9]. One meta-analysis by Jiang et al[10] including 12 randomized controlled trials (RCTs) showed that aerobic exercise selectively improved cognitive function, with moderate-intensity aerobic exercise lasting 4-8 weeks showing the best improvement in executive function. In a more recent meta-analysis including 14 RCTs, Ren et al[11] also indicated that exercise selectively improved executive function in patients with depression.

Despite the above research, a review of previous studies revealed that some have suggested that exercise does not significantly improve cognitive flexibility and inhibition within the executive function of patients with depression, although working memory is greatly improved. In a systematic review encompassing 8 RCTs, Brondino et al[12] posited that exercise did not significantly enhance overall cognitive function in patients with depression.

In this context, the present study aims to continue searching for evidence to further clarify the improvement effect of exercise on executive function in patients with depression. Indicators of cognitive function, including attention, memory, and information processing speed, were further evaluated in this study, with the aim of providing evidence-based support for clinical practice and a theoretical reference for researchers.

MATERIALS AND METHODS

Research methods

The PRISMA[13] statement guidelines were followed. The PICOS framework of this study is shown in Table 1.

Table 1 PICOS framework for the intervention effect of exercise on cognitive function in patients with depression.

PICOS	Content
Population	Patients with depression: Meet any of the diagnostic criteria of the International Classification of Disease and the Diagnostic and Statistical Manual of Mental Disorders
Intervention	Exercise or exercise in addition to the control group intervention
Comparison	Standard treatment, daily activities, and stretching exercises, etc.
Outcome	All indicators related to cognitive function
Study design	Randomized controlled trials

Literature retrieval strategy

Two researchers independently conducted searches across seven databases, including PubMed, Web of Science, the Cochrane Library, Embase, CNKI, Wanfang Medical Database, VIP Database, and various biomedical databases, to identify RCTs investigating the effect of exercise on cognitive function in patients with depression. The retrieval period was set from the establishment of each database until October 19, 2023. The search terms in Chinese included exercise, aerobic exercise, strength training, Tai chi, Baduanjin, cognitive function, executive function, and working memory. Only core journal articles were retrieved from Chinese databases. The English search terms included depression, depressive disorder, depressive symptoms, exercise, aerobic exercise, cognitive performance, executive function, inhibition, working memory, and RCTs. This primary search was supplemented by tracking relevant systematic reviews and references cited in the identified studies. The search combined subject terms with free words using Boolean operators "AND" and "OR,” and the strategies were finalized after repeated pre-checks. See Table 2 for the data retrieval strategies.

Table 2 Retrieval strategies for each database.

Database	Retrieval strategies
Cochrane and PubMed	#1“Exercise”[Mesh] OR “Aerobic exercise” [Title/Abstract] OR “Resistance exercise”[Title/Abstract] OR “High-intensity interval” [Title/Abstract] OR “Yoga” [Title/Abstract] OR “Dance” [Title/Abstract] OR “Taichi” [Title/Abstract] OR “Baduanjin” [Title/Abstract] OR “Wuqinxi” [Title/Abstract] OR “Yijinjing” [Title/Abstract] OR “Walking” [Title/Abstract] OR “Physical and mental exercise” [Title/Abstract]; #2“Depression”[Mesh] OR “Depressive disorder”[Title/Abstract] OR “Depressive symptom” [Title/Abstract] OR “Emotional depression” [Title/Abstract] OR “Depressive neurosis” [Title/Abstract] OR “Endogenous depression” [Title/Abstract] OR “Deurotic depression” [Title/Abstract] OR “Unipolar depression” [Title/Abstract]; #3“Cognition”[Mesh] OR“Cognitive performance”[Title/Abstract] OR“Executive function”[Title/Abstract] OR“Inhibition”[Title/Abstract] OR“Working memory”[Title/Abstract] OR“Shifiting”[Title/Abstract]; #4Randomized controlled trial[Publication Type] OR “Randomized” [Title/Abstract] OR “controlled”[Title/Abstract] OR “Trial” [Title/Abstract]; #5 #1 AND #2 AND #3 AND #4
Embase	#1 “Exercise”[exp] OR “Aerobic exercise”[ab,ti] OR “Resistance exercise”[ab,ti] OR “High-intensity interval” [ab,ti] OR “Yoga” [ab,ti] OR “Dance” [ab,ti] OR “Taichi” [ab,ti] OR “Baduanjin” [ab,ti] OR “Wuqinxi” [ab,ti] OR “Yijinjing” [ab,ti] OR “Walking” [ab,ti] OR “Physical and mental exercise” [ab,ti]; #2 “Depression”[exp] OR “Depressive disorder” [ab,ti] OR “Depressive symptom” [ab,ti] OR “Emotional depression” [ab,ti] OR “Depressive neurosis” [ab,ti] OR “Endogenous depression” [ab,ti] OR “Deurotic depression” [ab,ti] OR “Unipolar depression” [ab,ti]; #3 “Cognition” [exp] OR“Executive function” [ab,ti] OR “Cognitive performance” [ab,ti] OR “Cognitive” [ab,ti] OR “Working memory” [ab,ti] OR “Shifiting” [ab,ti] OR “Inhibition” [ab,ti]; #4 “Randomized controlled trial” [exp] OR “Randomized” [ab,ti] OR “Controlled” [ab,ti] OR “Trial” [ab,ti]; #5 #1 AND #2 AND #3 AND #4
Web of Science	#1 TS = ( “Exercise” OR “Aerobic exercise” OR “Resistance exercise” OR “High-intensity interval” OR “Yoga” OR “Dance” OR “Taichi” OR “Baduanjin” OR “Wuqinxi” OR “Yijinjing” OR “Walking” OR “Physical and mental exercise”); #2 TS = (“Depression” OR “Depressive disorder” OR “Depressive symptom” OR “Emotional depression” OR “Depressive neurosis” OR “Endogenous depression” OR “Deurotic depression” OR “Unipolar depression”); #3 TS = (“Cognition” OR “Cognitive performance” OR “Inhibition” OR “Working memory” OR “Shifiting” OR “Executive function”) #4 TS = ( “Randomized controlled trial” OR “Randomized” OR “Controlled” OR “Trial”); #5 #1 AND #2 AND #3 AND #4
CNKI	(Exercise + Physical education + Taijiquan + Baduanjin + Wuqinxi + Walking + Jogging + Strength Training + Aerobic Exercise) AND (Depression) AND (Memory + Working Memory + Cognition + Executive Function), core journal
Wanfang and China Biomedical	(Exercise OR Physical education OR Taichi OR Baduanjin OR Wuqinxi OR Walking OR Jogging OR Strength training OR Aerobic exercise) and (Depression) and (Memory OR Working memory OR Cognition OR Executive function), core journal

Inclusion and exclusion criteria

Inclusion criteria: (1) Study design: RCTs; (2) Language: Both Chinese and English; (3) Outcome indicators: Executive function, memory, attention, and information processing speed; (4) Patients with depression: Met any of the diagnostic criteria of the International Classification of Disease and the Diagnostic and Statistical Manual of Mental Disorders, without any other mental diseases; and (5) Intervention for the experimental group: Exercise alone or in addition to the control group intervention. We further included studies that focused on long-term exercise, defined as any "bodily movement produced by skeletal muscles with the expenditure of energy"[14], sustained for a minimum of three weeks. The intervention for the control group included no exercise or conventional treatment, including medication, occupational therapy, stretching exercises, health education, or a placebo.

Exclusion criteria: (1) Non-targeted intervention content; (2) No full text or specific data available; (3) Review articles, conference abstracts, etc.; (4) Non-targeted outcome indicators, and (5) Acute exercise intervention for the experimental group.

Literature screening and data extraction

Literature screening: Two researchers independently searched the seven databases and imported the retrieved literature into Endnote X9. After completing the searches, duplicates were removed from Endnote X9, and the titles and abstracts were screened. Finally, the full texts of the preliminarily screened studies were read according to the inclusion and exclusion criteria to determine the final inclusion.

Data extraction: Two researchers used a predeveloped Microsoft Excel spreadsheet for information and data extraction. The heterogeneity in data extraction between the two researchers was 97.02%. Unclear or missing data were clarified through email communication with the original authors. If discrepancies arose, a third researcher was consulted. The extracted data included basic information (author, year, country, age, sample size, and depression level), experimental characteristics (type, intensity, single exercise session duration, frequency, and duration of intervention), and outcome indicators.

Coding was performed based on exercise type, exercise intensity, exercise duration, and intervention content. Exercise types were coded as aerobic exercise, strength training, mind-body exercise, and others. Exercise intensity was coded as low, moderate, moderate-high, or not specified. Exercise duration was coded as 3-12 weeks or 13-16 weeks. The frequency was coded as 1, 2, 3, or 2-3 days/week. The single exercise session duration was coded as 30-60 minutes or > 60 minutes.

Quality assessment

Quality assessment of included studies: Two researchers independently assessed the study quality. If discrepancies arose, a third researcher was consulted. The Cochrane Handbook 5.1.0 recommended RCT risk of bias measurement tool was used for assessment, with five or more items indicating a low risk of bias, 3-4 items indicating a moderate risk of bias, and less than 3 items indicating a high risk of bias[15].

Quality assessment of outcome evidence: GRADEpro software was used for quality assessment of the outcome evidence, with results graded as follows: "High": Very confident in the result, further research is unlikely to change this result; "Moderate": Moderately confident in the result, further research may change the outcome; "Low": Limited confidence in the result, further research is very likely to change the outcome; "Very low": Almost no confidence in the result, further research is very likely to change the outcome.

Statistical analysis

Statistical analyses were conducted using Stata17 software. As all included outcome indicators were continuous variables measured using different tools, Hedge’s g and 95%CI were used as the effect size indicators. Owing to different measurement tools and the inconsistent direction of measurement units of outcome indicators, the means were multiplied by -1 according to the Cochrane Handbook 5.1.0, to ensure consistency in the direction of the measurement units. An effect size of 0-0.2 was considered as small, 0.2-0.49 as small to moderate, 0.5-0.79 as moderate, and greater than or equal to 0.8 as large[16]. Heterogeneity was quantified using the I² statistic with 25%, 50%, and 75% representing low, moderate, and high heterogeneity, respectively. If heterogeneity occurred, a random-effects model (REM) was used for analysis, along with a subgroup analysis to explore the sources of heterogeneity. Otherwise, a fixed effects model (FEM) was used.

RESULTS

Literature search results

A total of 5856 studies were retrieved through the initial search of the seven databases, with four additional studies[17-20] identified through citation tracking. Endnote X9 was used to remove 658 duplicate references. Subsequently, we conducted a preliminary screening of the remaining 5202 references through a review of the titles and abstracts, excluding 5098 irrelevant references. Finally, we performed full-text assessments of the remaining 104 references, from which 87 that did not meet our inclusion criteria were excluded. Finally, 17 studies were included in the meta-analysis[6,7,17-31]. A flowchart of this process is presented in Figure 1.

Figure 1 Flow chart of the literature search and study selection process.

Characteristics of the literature included

Seventeen studies, 15 in English and two in Chinese, originating from China, the United States, Germany, Switzerland, Lithuania, Denmark, and India, were included in this study. The total number of participants was 1173, with ages ranging from 18 to 80 years. The interventions for the experimental groups included aerobic exercise, strength training, and Tai chi, while the control groups received placebo, relaxation exercises, or stretching exercises. Of these studies, 15 reported the intensity of exercise, categorized as low, moderate, or high intensity[32], as shown in Table 3.

Table 3 Characteristics of the literature included.

Ref.	Country	Depression level	Sample size	Age	Intervention		Exercise intensity	Outcome indicators
					Experimental group	Control group
Buschert et al[6]	Germany	Moderate	18/20	47.27 ± 6.84; 47.47 ± 8.47	Aerobic exercise, 30 minutes, 2-3 times/week, 3-4 weeks	Occupational or art therapy	85% HRmax/moderate to high intensity	TAP
Hoffman et al[20]	United States	Mild-moderate	104/49	51.7 ± 7.6	Aerobic exercise, 3 times/week, 16 weeks	Placebo	70%-85% HRR/moderate to high intensity	Digit Span, Stroop, digit symbol subtest
Imboden et al[7]	Switzerland	Moderate-severe	22/20	41.3 ± 9.2; 38.3 ± 13.4	Aerobic exercise, 45 minutes, 3 times/week, 6 weeks	Stretching exercises	60%-75% HRmax/moderate intensity	Go-NoGo, TAP
Khatri et al[21]	United States	Mild-severe	42/42	56.73 ± 6.45	Aerobic exercise, 45 minutes, 3 times/week, 16 weeks	Sertraline	70%-85% HRR/moderate to high intensity	Digit Span, TMT-A/B, stroop, digit symbol
Krogh et al[22]	Denmark	Mild-moderate	47/48/42	41.98 ± 8.7; 38.1 ± 9.0; 36.7 ± 8.7	Aerobics, strength exercises, 90 minutes, 2 times/week, 16 weeks	Relaxation exercises	50%-75% R, 70%-89% HRmax/moderate to high intensity	Digit Span, Subtracting Serial Sevens, TMT-A/B, Digit Symbol
Lavretsky et al[19]	United States	Moderate-severe	36/37	69.1 ± 7; 72 ± 7.4	Tai chi, 120 minutes, 1 times/week, 10 weeks	health education	Low intensity	TMT-A
Olson et al[23]	United States	Moderate	15/15	21.0 ± 1.9; 21.2 ± 2.2	Aerobic exercise, 45 minutes, 3 times/week, 8 weeks	Stretching exercises	40%-65% HRR/moderate intensity	Flanker
Sharma et al[25]	India	Not specified	15/15	31.87 ± 8.78; 31.67 ± 8.46	Yoga in combination with standard treatment, 30 minutes, 3 times/week, 8 weeks	Standard treatment	Low intensity	LCT, TMT-A/B, Digit Span
Zhang et al[26]	China	Mild-severe	20/19	47.2 ± 6.99/54.16 ± 6.09	Taichi, 90 minutes, 2 times/week, 12 weeks	Daily life	Low intensity	Stroop
Krogh et al[18]	Denmark	Mild-moderate	56/59	39.7 ± 11.3; 43.4 ± 11.2	Aerobic exercise, 45 minutes, 3 times/week, 12 weeks	Stretching exercise	65%-80% VO2max/moderate to high intensity	TMT-A/B, troop, Buscher's SRT, RCFT, Digit Span, Subtracting Series Seven, Digit Symbol Test
Krogh et al[27]	Denmark	Mild-moderate	41/38	38.9 ± 11.7; 43.8 ± 12.2	Aerobic exercise, 45 minutes, 3 times/week, 12 weeks	Stretching exercise	80% HRmax/moderate to high intensity	RCFT
Brush et al[24]	United States	Mild-severe	35/31	20.26 ± 2.84; 20.19 ± 1.78	Aerobic exercise, 45 minutes, 3 times/week, 8 weeks	Stretching exercise	40%-65% HRR/moderate intensity	Flanker
Luttenberger et al[17]	Germany	Mild	22/25	42.71 ± 11.88; 44.96 ± 12.08	Indoor climbing, 180 minutes, 1 times/week, 8 weeks	Waiting group	Not specified	Attention Test d2-R
Chan et al[28]	China	Not specified	17/16	47.06 ± 9.54; 45.44 ± 8.25	Zen Martial Arts Medication, 90 minutes, 1 times/week, 10 weeks	Waiting group	Not specified	DVT
Kerr[29]	Lithuania	Moderate	39/13	54.1 ± 18.3; 55.1 ± 16.5	Aerobic exercise combined with medication treatment, strength training combined with medication treatment, 45 minutes, 3 times/week, 4 weeks	Medication treatment	Moderate intensity	Go-NoGo, Simple Reaction Time, Two Choice Reaction Time
Chen et al[30]	China	Remission phase	63/32	30.3 ± 7.5; 32.7 ± 6.5	Aerobic exercise, 30-60 minutes, 3 days/week, 16 weeks	Medication treatment	64%-76% HRmax/moderate intensity	TMT-A/B, CPT, Digit Span
Zheng et al[31]	China	Not specified	30/30	37.5 ± 9.12; 35.17 ± 5.93	Aerobic exercise combined with standard treatment, 60 minutes, 5 days/week, 4 weeks	Standard treatment	Moderate intensity	RBANS

TMT-A/B: The Trail Making Test (part A and B); TAP: Testbat terie zur Aufmerksamkeitsprüfung; Digit Span: Digit Span Subtest from the WAIS-R; Digit Symbol Subtest: Digit Symbol Subtest from the Wechsler Adult Intelligence Scale-Revised; LCT: Letter Cancellation Test; CPT: Continuous Performance Test; Buscher's SRT: Buschke Selective Reminding Test; RCFT: Rey’s Complex Figure Test; DVT: Digit Vigilance Test; RBANS: Repeatable Battery for the Assessment of Neuropsychological Status; HRmax: Maximal Heart Rate; HHR: Heart Rate Reserve; VO₂max: Maximal oxygen uptake; RM: Repetition maximum.

Quality assessment of the literature included

All 17 included studies detailed their randomization methodologies. Notably, one study[17] lacked allocation concealment, while one other study did not implement blinding of the participants and personnel. Concerns regarding incomplete outcome data reporting were identified in two studies[18,19], while one[21] exhibited potential for selective reporting. Additionally, two studies[25,26] presented various other risks of bias; eight studies had a low risk of bias and nine had a moderate risk, as shown in Figure 2.

Figure 2 Quality assessment of the included literature.

Results of the meta-analysis results

Effect of exercise on attention in patients with depression: Ten studies focused on attention enhancement in patients with depression (Figure 3A). The results of the heterogeneity test (I² = 46.65%) indicated a low heterogeneity among the studies. Therefore, a REM was used to combine effect sizes. The meta-analysis results showed that the combined effect size was 0.21, with a 95%CI: 0.07-0.34 (P < 0.01), indicating a statistically significant difference. This suggests that exercise has a moderate to small effect on attention improvement in patients with depression.

Figure 3 Forest-plot showing the effect of exercise on cognitive function in patients with depression. A: Attention; B: Memory; C: Executive function; D: Information processing speed. CI: Confidence interval.

Subgroup analysis of exercise type, single exercise session duration, exercise frequency, exercise intensity, and exercise program duration was conducted to explore the variables that might influence heterogeneity. As shown in Table 4, statistically significant results were found in the groups who underwent aerobic exercise, single exercise session duration of 30-60 minutes, exercise program duration of 3-12 weeks, exercise frequency of 3 days per week, and moderate intensity (all P < 0.05).

Table 4 Subgroup analysis of the effect of exercise on attention in depression patients.

Moderating variables	I2	n	Hedges’s, 95%CI	P value
Exercise type
Aerobic exercise	57.44	16	0.21, 0.03-0.40	0.02
Strength training	53.34	5	0.31, -0.05-0.68	0.09
Mind-body exercise and others	0.0	6	0.13, -0.14-0.40	0.36
Single exercise session duration
30-60 minutes	43.1	21	0.27, 0.11-0.42	0.00
> 60 minutes	0	6	0.03, -0.14-0.20	0.70
Exercise program duration
3-12 weeks	0	17	0.23, 0.06-0.40	0.01
13-16 weeks	67.77	10	0.18, -0.06-0.42	0.15
Exercise frequency
1 day/week	0.32	2	0.30, -0.14-0.73	0.18
2 days/week	0	4	-0.02, -0.20-0.17	0.87
3 days/week	49.48	19	0.27, 0.10-0.44	0.00
2-3 days/week	80.43	2	0.23, -0.91-1.37	0.69
Exercise intensity
Low	0	4	0.02, -0.33-0.37	0.92
Moderate	0	10	0.53, 0.373-0.68	0.00
Moderate-high	0	11	-0.03, -0.16-0.10	0.67
Not specified	0.32	2	0.30, -0.14-0.73	0.18

To further investigate the sources of heterogeneity, a sensitivity analysis was conducted to determine whether the heterogeneity was caused by individual studies. After sequentially excluding individual studies, the effect size showed no significant change, indicating stable results (Figure 4). Egger's test (t = 1.07, P > z = 0.19 > 0.05, suggested no publication bias in this study (Figure 5A).

Figure 4 Sensitivity analysis of the effect of exercise on attention in patients with depression. CI: Confidence interval.

Figure 5 Funnel-plot showing the effect of exercise on cognitive function in patients with depression. A: Attention; B: Memory; C: Executive function; D: Information processing speed. CI: Confidence interval.

Effect of exercise on memory in patients with depression: Eight studies examined the effect of exercise on memory in patients with depression, with low heterogeneity (I² = 24.72%) (Figure 3B). Therefore, a FEM was employed. According to the meta-analysis, the combined effect size was 0.08, 95%CI: 0.00-0.16, and P = 0.05, indicating that the improvement in memory was not statistically significant. Egger's test (t = 0.84, P > z = 0.53 > 0.05, indicated no publication bias in this study (Figure 5B).

Effect of exercise on executive function in patients with depression: The effect of exercise on executive function was reported in eight studies, demonstrating moderate heterogeneity (I² = 62.92%) (Figure 3C). The meta-analysis using an REM yielded a combined effect size of 0.11, 95%CI: -0.11-0.32, P = 0.34, suggesting that the effect of exercise on executive function in patients with depression was not statistically significant. Egger's test (t = 0.93, P > z = 0.35 > 0.05, suggested no publication bias in this study (Figure 5C).

Effect of exercise on information processing speed in patients with depression: Seven studies reported on the effect of exercise on improving the information processing speed in patients with depression. As shown in Figure 3D, the heterogeneity test results showed that I² = 0%, indicating no heterogeneity among the studies. Therefore, a FEM was applied to combine effect sizes. The results of the meta-analysis showed that the combined effect size was 0.14, with a 95%CI: 0.04-0.25, P = 0.01, indicating statistical significance, suggesting that exercise has a small effect on improving information-processing speed in patients with depression. Egger's test, t = -0.08, P > z = 0.97 > 0.05, indicated no publication bias in this study (Figure 5D).

GRADEpro quality assessment

The GRADEpro assessment indicated that the quality of evidence for the results on executive function was "very low", while the results on attention, memory, and information processing speed were "low," as shown in Table 5.

Table 5 Quality assessment of outcome level.

Outcome indicators	RCTs/item	Quality assessment of evidence					Number of events/sample size		Relative effect size	Quality
		Limitations	Inconsistency	Indirectness	Imprecision	Publication bias	Intervention group	Control group
Attention	22	Downgrade 1 level (A)	Downgrade 1 level (B)	No downgrade	No downgrade	No downgrade	877	877	0.21 (0.07, 0.34)	Low
Executive function	16	Downgrade 1 level (A)	Downgrade 1 level (C)	Downgrade 1 level (D)	No downgrade	No downgrade	521	464	0.11 (-0.11, 0.32)	Very low
Information processing speed	13	Downgrade 1 level (A)	Downgrade 1 level (B)	No downgrade	No downgrade	No downgrade	697	647	0.14 (0.04, 0.25)	Low
Memory	27	Downgrade1 level (A)	Downgrade 1 level (B)	No downgrade	No downgrade	No downgrade	1229	1124	0.08 (0.00, 0.16)	Low

RCT: Randomized controlled trial; A: Most studies did not perform allocation concealment and blinding assessment; B: Some studies did not differentiate the severity of depression in patients; C: High heterogeneity between studies; D: The assessment tools for executive function were not finely included.

DISCUSSION

Overall, the present investigation revealed that exercise moderately enhances attention and information processing speed in patients with depression, but fails to bolster executive function and memory. These results differ from those of some previous studies. For example, Brondino et al[12] detected no enhancements in information processing speed, attention, or working memory in patients with depression in eight studies, while Contreras-Osorio et al[33] identified improvements in working memory among individuals with mild to moderate depression, but observed no significant changes in cognitive flexibility or inhibitory capabilities. Ren et al[11] also reported positive effects on working memory, cognitive flexibility, and reasoning/planning, excluding inhibitory functions. Meanwhile, Jiang et al[10] noted that aerobic exercise boosted executive function, visual learning, and memory, but not the speed of processing information or patient levels of attention/alertness. Such variations in results could stem from the diversity in the exercise types and measurement tools assessed, as each tool harbors unique traits and sensitivities towards different cognitive domains. However, it should be noted that this study included only aerobic exercises. In contrast, our study included a variety of exercises other than aerobic exercises such as strength training, Taichi, and yoga. Moreover, existing research has shown that exercise type has specific benefits for cognitive function. We hypothesized that the lack of improvement in executive function and memory due to exercise may be related to the patients' adherence to a regular exercise regimen. Indeed, the World Health Organization recommends at least 150 minutes of moderate-intensity aerobic activity per week, or two or more days of physical activity per week. However, patients with depression may be more likely drop out or have low attendance rates due to a dislike of exercise, lack of time, or high exercise intensity[34], which could lead to minimal benefits from exercise. Contrary to previous studies, we found that exercise improved attention in patients with depression. However, prior research has indicated that aerobic exercise does not enhance attention in individuals with depression[35]. Some studies have further indicated that aerobic exercise does not improve attention in patients with schizophrenia[36]. Unlike these studies, our research included a variety of different exercise types, including aerobic exercise, yoga, and Tai chi. Yoga and Tai chi place greater emphasis on attention training as mind-body exercises[37].

Improvements in information processing speed and attention due to exercise may be linked to levels of Brain-Derived Neurotrophic Factor (BDNF). BDNF plays a crucial role in the development and survival of the nervous system[38], and is closely associated with cognitive function[39]. Indeed, research has indicated that serum levels of BDNF are reduced in patients with depression, while exercise can increase BDNF levels[40], thereby enhancing brain plasticity and improving cognitive functions such as attention.

This study has several limitations. Exercise has been shown to improve attention and information processing speed in patients with depression; however, due to limitations and inconsistencies, the level of evidence of these outcomes is considered "low." Exercise does not improve executive function, with a "very low" level of evidence attributable to its limitations, indirectness, and inconsistencies. Additionally, no improvement in memory was observed, resulting in a "low" evidence level due to limitations and inconsistencies. Although all included studies included RCTs, most lacked allocation concealment, assessor blinding, and researcher blinding, thereby increasing the risk of misleading results and influencing the reliability of outcomes. Furthermore, despite the study population consisting of patients with depression, the failure to distinguish between different levels of depression severity in most studies could be a key reason underlying the inconsistencies in findings. Additionally, executive function encompasses cognitive subcomponents, such as working memory, inhibition, and shifting. However, the included indicators were relatively singular, and did not directly or comprehensively assess the executive function in patients with depression. As such, future research should, based on the design of RCTs, strive to increase allocation concealment, researcher blinding, and assessment blinding to enhance methodological quality. Patients with depression should further be categorized more clearly to devise precise exercise prescriptions for patients with different levels of depression. Moreover, future studies should detail the categorization of different cognitive functions in patients with depression to obtain more scientifically robust results. Finally, we did not consider the patients' cultural backgrounds. Future research should incorporate cultural factors to enhance the reliability of results.

Overall, this study discovered that moderate-intensity aerobic exercises lasting 30-60 minutes, conducted three times a week, and spanning a period of 3-12 weeks, could be an effective regimen for preserving the attention of patients with depression. Depression generally manifests clinically as "psychomotor retardation," characterized by a slowdown in response times during psychological tests[41] which directly reflects a decrease in information-processing speed, which is potentially linked to attentional focus[42]. Patients with depression exhibit weaker functional connectivity in the right dorsolateral prefrontal cortex, which is associated with impaired response inhibition and cognitive flexibility[43]. In contrast to strength training and other exercises, long-term aerobic exercise boosts the number and efficiency of capillaries and mitochondria, thus augmenting the body’s energy supply, and improving blood oxygen levels and cerebral circulation. This, in turn, enhances brain neuroplasticity. The United States Department of Health and Human Services advocates 150 minutes of moderate exercise weekly to guard against cognitive decline, emphasizing that engaging in less than two hours of moderate-intensity or an hour of high-intensity exercise weekly over three decades correlates with accelerated cognitive deterioration[44]. According to the inverted U-shaped theory, neither low- nor high-intensity exercises can improve cognitive function, which explains the results of this study to some extent. Indeed, studies have also indicated that excessive or high-intensity exercise can cause exercise fatigue, disrupt neuronal order, and impair cognitive function in patients[45]. The "30-60" minute group showed improvements in attention, possibly because prolonged exercise could overactivate the prefrontal and supplementary motor areas, thus damaging cognitive performance, while long durations of exercise could increase fatigue and dehydration, negatively influencing cognitive performance[46]. The "3-12" week group showed improvement in attention, while the "13-16" week group did not, possibly because the population included in this study consisted of patients with depression, while previous research has suggested that long-term exercise, regardless of the duration per session, could cause psychological harm to patients with depression, potentially exacerbating attentional impairments[47].

In addition, this study revealed that exercise did not significantly improve memory in patients with depression. Age may act as a moderating variable affecting the degree to which exercise enhances memory in patients with depression, suggesting that patients with poor memory performance may benefit more from exercise. The age of the patients with depression included in this study ranged from 19/20 to 72 years. Compared to younger patients, older patients with depression may experience more significant memory impairment, thus benefiting more from exercise. Conversely, younger patients may see fewer benefits[48].

CONCLUSION

In conclusion, exercise can enhance attention and information processing speed in patients with depression, but results in no significant improvement in executive function and memory. Aerobic exercise, particularly of moderate intensity, lasting between 30-60 minutes, and conducted three times a week over a duration-3-12 weeks, demonstrated the best improvement effects on attention. As such, future clinical interventions should be tailored to the specific cognitive impairments of patients, while considering the impact of different exercise elements. Future research should employ more rigorous methodological designs and include larger multicenter RCT to verify the clinical efficacy of exercise on the cognitive function of patients with depression.