Chronic exercise effects on overall depression severity and distinct depressive symptoms in older adults: A protocol of a systematic and meta-analytic review.

This is a study protocol, not a completed study. The authors are planning to:

**Intervention:** Chronic (regular, repeated) exercise programs of any type (aerobic, resistance, mind-body like yoga/tai chi, balance/coordination exercises)

**Comparators:** Any control condition (no exercise, usual care, placebo, or alternative treatment)

**Primary outcome:** Change in overall depression severity from baseline to post-intervention

**Secondary outcomes:** Changes in specific depressive symptoms as defined by DSM-5 criteria, including:

- Sleep quality

- Fatigue/energy levels

- Anxiety

- Mood (sadness, depressed mood)

- Apathy/loss of interest

- Changes in weight/appetite

- Information processing speed

- Executive functions (planning, inhibition, cognitive flexibility)

The key innovation is that instead of treating depression as a single score, they will examine whether exercise affects different symptoms differently—for example, whether aerobic exercise improves sleep more than resistance training, or whether mind-body exercises improve mood more than cognition.

The planned review will include randomized controlled trials (RCTs) with:

**Sample:** Older adults with a mean age of at least 60 years

**Population:** Both clinically depressed and non-clinically depressed individuals (they will analyze these separately)

**Setting:** Any setting (community, clinical, residential care)

**No restrictions** on sex, gender, ethnicity, or baseline health status

**Exclusions:** Studies with mean age below 60, non-randomized designs, acute/single-session exercise studies

The final sample size will depend on how many eligible RCTs exist in the literature. Based on previous meta-analyses in this area, they might expect to find 20-50 eligible studies.

The authors will extract data from published studies using validated instruments. Common measures they expect to encounter include:

**Overall depression severity:** Geriatric Depression Scale (GDS, 0-15 or 0-30), Beck Depression Inventory (BDI-II, 0-63), Hamilton Depression Rating Scale (HAM-D, 0-52), Center for Epidemiologic Studies Depression Scale (CES-D, 0-60)

**Sleep quality:** Pittsburgh Sleep Quality Index (PSQI, 0-21, lower = better sleep)

**Fatigue:** Fatigue Severity Scale (FSS), Multidimensional Fatigue Inventory (MFI)

**Anxiety:** State-Trait Anxiety Inventory (STAI), Geriatric Anxiety Inventory (GAI)

**Cognition:** Trail Making Test (processing speed), Stroop Test (executive function), Digit Span (working memory), Wisconsin Card Sorting Test (cognitive flexibility)

**Apathy:** Apathy Evaluation Scale (AES), apathy subscale of neuropsychiatric inventories

They will also extract intervention characteristics: frequency (sessions/week), intensity (heart rate, RPE, % of 1RM), time (session duration in minutes), type (aerobic, resistance, mind-body, combined), cognitive demand of the exercise, social interaction during exercise, supervision level, use of behavior change techniques, and compliance/adherence rates.

### Study design

This is a **systematic review and meta-analysis protocol**—a detailed plan for how the authors will conduct a future review of existing RCTs. The protocol follows PRISMA-P guidelines (Preferred Reporting Items for Systematic Reviews and Meta-Analysis Protocols).

### Search strategy

They will search **six electronic databases**: Web of Science, Academic Search Complete, CINAHL, APA PsycInfo, SPORTDiscuss, and the Cochrane Library. Two independent reviewers will screen titles/abstracts, then full texts. A third reviewer resolves disagreements.

### Inclusion criteria (PICOS)

**Population:** Adults ≥60 years (mean age)

**Intervention:** Chronic exercise (≥2 weeks of structured, planned physical activity)

**Comparator:** Any control (no exercise, sham exercise, usual care, alternative treatment)

**Outcomes:** Depression severity (primary), specific depressive symptoms (secondary)

**Study design:** Randomized controlled trials only

### Data extraction and quality assessment

Two independent reviewers will extract data using a standardized form. They will assess risk of bias using the **Cochrane Risk of Bias tool (RoB 2)** , which evaluates: random sequence generation, allocation concealment, blinding of participants/personnel, blinding of outcome assessment, incomplete outcome data, selective reporting, and other biases.

### Statistical approach

**Meta-analysis:** They will use random-effects models (which assume true effects vary across studies) to pool effect sizes (standardized mean differences, Hedges' g)

**Subgroup analyses** to examine moderators:

- Intervention characteristics: frequency, intensity, duration, type, cognitive demand, social interaction, supervision, behavior change techniques, compliance, dropout rate, control group type

- Individual characteristics: age, sex, education, functional capacity, global cognition, population (clinical vs. non-clinical depression)

**Sensitivity analyses:** To test robustness of findings

**Publication bias assessment:** Funnel plots and Egger's test

**Quality of evidence:** GRADE approach (Grading of Recommendations, Assessment, Development and Evaluations)

### What this design can and cannot prove

**What it CAN prove:**

Whether chronic exercise, on average, reduces depression severity in older adults

Whether different types of exercise have different effects on different depressive symptoms

Which intervention characteristics (frequency, intensity, duration, type) are associated with larger effects

Whether individual characteristics (age, sex, cognition) moderate the effects

The overall quality and consistency of the evidence base

**What it CANNOT prove:**

Causality at the individual level (meta-analyses aggregate group-level data)

Mechanisms of action (why exercise works—this requires mechanistic studies)

Optimal individual prescription (subgroup analyses can suggest patterns but cannot provide personalized recommendations)

Long-term effects beyond the intervention period (most RCTs have short follow-ups)

Effects on rare subgroups (small sample sizes in subgroups limit statistical power)

Dose-response relationships in fine detail (studies use different doses, making precise dose-response curves difficult)

### Major methodological considerations

**Strengths of the planned review:**

Comprehensive search across 6 databases

Focus on specific symptoms, not just total scores—this is novel and clinically relevant

Extensive moderator analyses to explain heterogeneity

Pre-registered protocol (PROSPERO) to reduce reporting bias

Two independent reviewers at all stages

**Potential weaknesses (to be assessed when the review is completed):**

High heterogeneity across studies is likely (different populations, exercise protocols, outcome measures)

Many older adult exercise studies have small sample sizes

Blinding is difficult in exercise trials (participants know they're exercising)

Publication bias toward positive results

Limited data on specific symptoms (many studies only report total depression scores)

The review depends entirely on the quality of existing studies

**IMPORTANT: This is a protocol—no results are reported yet.** The authors have published their plan for a future review. The actual findings will be published later.

However, the protocol itself provides valuable context from the introduction:

**Global prevalence of geriatric depression:** 31.74% (average across studies from 1994-2020)

**Prevalence in those 85+:** Up to 50%

**Antidepressant efficacy in older adults:** "Modest and declines with age"

**Exercise as alternative:** "No serious side effects, accessible to everyone"

The authors cite previous evidence that chronic exercise benefits overall depression severity in older adults, but note that "the heterogeneity of the effects is high," which is why this detailed moderator analysis is needed.

Since no results are available, we can look at what the authors expect based on previous literature:

Previous meta-analyses of exercise for depression in older adults have found **moderate effect sizes** (Hedges' g ≈ 0.4-0.7), meaning exercise reduces depression by about 0.4-0.7 standard deviations compared to control

This translates roughly to a **3-5 point reduction on the Geriatric Depression Scale (30-point version)** or a **4-8 point reduction on the Beck Depression Inventory**

For comparison, antidepressant medication typically produces effect sizes of 0.3-0.5 in older adults

The authors expect that **different symptoms will respond differently**—for example, sleep and fatigue might show larger improvements than core mood symptoms, or vice versa

### Limitations acknowledged by the authors:

The review depends on the quality and completeness of published RCTs

Many exercise studies in older adults have small sample sizes

Heterogeneity in exercise protocols, populations, and outcome measures will make synthesis challenging

Publication bias may inflate effect estimates

The review cannot address mechanisms of action

### Additional limitations a critical reader would note:

**Protocol only:** This paper describes what the authors PLAN to do, not what they FOUND. The actual review may take 1-2 years to complete

**No results to evaluate:** We cannot assess the quality of their findings yet

**Broad inclusion criteria:** Including both clinically depressed and non-clinically depressed populations may muddy the results

**Age cutoff of 60:** This is relatively young for "older adults"—many geriatric studies use 65+. Results may differ for the "oldest old" (75+)

**No restriction on depression severity at baseline:** Exercise effects may differ for mild vs. severe depression

**Exercise definition:** "Chronic exercise" is defined as ≥2 weeks, which is very short for a "chronic" intervention. Most exercise studies run 8-16 weeks

**No plan for network meta-analysis:** They will compare exercise types indirectly but cannot rank them statistically

**Funding source:** COST Action (European research network)—no obvious industry bias, but the consortium includes exercise scientists who may have vested interest in positive results

**IMPORTANT: Since this is a protocol with no results, the takeaways below are based on the authors' stated goals and previous literature, not on this specific paper's findings.**

### For someone running their own n=1 experiment:

**What to test:**

Choose ONE type of exercise to test systematically:

- **Aerobic:** Brisk walking, cycling, swimming, or jogging at 60-80% of maximum heart rate

- **Resistance:** Bodyweight exercises, resistance bands, or weights (8-12 reps, 2-3 sets)

- **Mind-body:** Yoga, tai chi, or qigong (focus on breath-movement coordination)

- **Combined:** Mix of aerobic and resistance (e.g., 20 min walking + 20 min strength)

Test one type for 8-12 weeks before switching to another type

Consider adding a "cognitive demand" element (e.g., walking while doing mental arithmetic, learning new dance steps, tai chi with complex sequences)

**Minimum meaningful duration:**

**8 weeks minimum** for detectable changes in depression symptoms

**12-16 weeks optimal** for robust effects

**Exercise frequency:** 3-5 sessions per week

**Session duration:** 30-45 minutes per session

**Intensity:** Moderate (you can talk but not sing) to vigorous (you can only say a few words)

**What to measure (specific metrics):**

**Primary:** Geriatric Depression Scale (GDS-15 or GDS-30) or Beck Depression Inventory (BDI-II)—take weekly

**Sleep:** Pittsburgh Sleep Quality Index (PSQI) or simple sleep diary (bedtime, wake time, number of awakenings, sleep quality rating 1-10)

**Fatigue:** Single-item energy rating (1-10) each morning and evening

**Anxiety:** Generalized Anxiety Disorder scale (GAD-7) or single-item anxiety rating

**Mood:** Daily mood rating (1-10) at same time each day

**Cognition:** Simple processing speed test (e.g., Trail Making Test A) and executive function test (e.g., Trail Making Test B or Stroop test)—weekly

**Exercise adherence:** Log each session (type, duration, intensity, enjoyment rating 1-10)

**Key confounds to control for:**

**Medication changes:** Record any changes in antidepressants, sleep aids, or other medications

**Life events:** Note major stressors (bereavement, relocation, illness, financial problems)

**Seasonal effects:** Depression worsens in winter—run your experiment across seasons or control for daylight hours

**Social interaction:** Exercise classes provide social contact, which independently improves mood. If testing solo exercise, compare with a social activity control

**Sleep hygiene:** Changes in sleep schedule, caffeine, alcohol, or screen time before bed

**Diet:** Changes in eating patterns, especially omega-3 intake, vitamin D, and processed foods

**Other physical activity:** Track non-exercise movement (walking to shops, gardening, housework)

**Time of day:** Exercise at the same time each day to control for circadian effects

**Baseline depression severity:** More severe depression may require longer intervention or higher intensity

**What a positive result would look like:**

**Depression score:** Decrease of ≥3 points on GDS-15 (20% reduction) or ≥5 points on BDI-II (clinically meaningful change)

**Sleep:** PSQI decrease of ≥3 points (better sleep), or falling asleep 10-15 minutes faster, or 30+ minutes more total sleep

**Fatigue:** Energy rating increase of ≥2 points (on 1-10 scale)

**Mood:** Daily mood rating increase of ≥1.5 points (on 1-10 scale), sustained for 2+ weeks

**Cognition:** Trail Making Test B completion time decrease of ≥10 seconds (faster processing)

**Pattern:** Improvements should follow a dose-response pattern (more exercise = more improvement) and should reverse if you stop exercising for 2+ weeks

**Timing:** Most people see initial improvements in 4-6 weeks, with peak effects at 8-12 weeks

### Specific recommendations for older adults (60+):

1. **Start with walking**—lowest injury risk, easiest to maintain

2. **Add resistance training** 2x/week for muscle maintenance and metabolic benefits

3. **Include balance exercises** (tai chi, yoga, or simple standing on one foot) to reduce fall risk

4. **Exercise with a partner or group** for social accountability

5. **Monitor for overtraining**—older adults need longer recovery between sessions

6. **Get medical clearance** if you have cardiovascular disease, joint problems, or take multiple medications

7. **Track side effects**—exercise should not cause persistent pain, dizziness, or shortness of breath

### What to watch for (red flags):

Worsening depression despite exercise (may need professional help)

Exercise addiction or compulsive exercise (rare but possible)

Injury from improper form or overtraining

Exercise becoming a source of stress rather than relief

Using exercise to avoid addressing underlying psychological issues

**Bottom line for self-experimenters:** This protocol confirms that exercise is a promising treatment for late-life depression, but the "best" exercise depends on your specific symptoms. If you struggle with sleep, try aerobic exercise. If you struggle with motivation/apathy, try group exercise or mind-body exercise with social elements. If you struggle with cognition, try exercise that requires coordination and learning (dance, tai chi, complex resistance training). Run your experiment for at least 8 weeks, measure multiple symptoms (not just mood), and track potential confounds carefully.