The effects of physical activity on cortisol and sleep: A systematic review and meta-analysis.

This systematic review and meta-analysis investigated the relationship between physical activity (PA) and two key physiological markers: cortisol levels (a primary stress hormone) and various aspects of sleep quality.

The **intervention** was any form of physical activity, ranging from structured exercise programs (e.g., aerobic, resistance, yoga) to general increases in daily movement. The review considered studies that either compared an active group to a sedentary control group, or observed the effects of varying levels of PA in a population.

The **comparators** were typically sedentary control groups, usual care, or lower levels of physical activity.

The **outcome measures** were:

**Cortisol levels:**

* **Morning cortisol:** Cortisol levels measured shortly after waking.

* **Cortisol Awakening Response (CAR):** The sharp increase in cortisol levels that typically occurs within 30-45 minutes of waking. A healthy CAR is often characterized by a robust rise.

* **Diurnal cortisol:** The pattern of cortisol secretion throughout the day, often measured by the total area under the curve (AUC) or the slope of decline.

**Sleep quality:**

* **Overall sleep quality:** Often assessed by questionnaires.

* **Sleep duration:** Total time spent asleep.

* **Sleep latency:** Time taken to fall asleep.

* **Wake After Sleep Onset (WASO):** Time spent awake after initially falling asleep.

* **Sleep efficiency:** The percentage of time in bed spent asleep.

This meta-analysis included data from 31 studies involving a total of 2,233 adult participants. The studies encompassed a broad range of adult populations, including:

Healthy adults.

Older adults.

Individuals with specific conditions such as depression, anxiety, chronic fatigue syndrome, fibromyalgia, and spinal cord injury.

Participants were from various settings, including community-dwelling individuals and those in care homes.

The review specifically focused on studies involving human adults (aged 18 years or older) and excluded animal studies or studies focused on specific clinical populations where cortisol or sleep dysregulation was a primary diagnostic criterion (e.g., Cushing's syndrome, Addison's disease, severe sleep disorders like narcolepsy or sleep apnea, unless the PA intervention was specifically designed to address these). The broad inclusion criteria mean the findings are generally applicable to a diverse adult population, though specific effects might vary by age or health status.

The methods for measuring cortisol and sleep varied across the included studies, reflecting the diverse nature of research in this field.

**Cortisol measurements:**

**Salivary cortisol:** This was the most common method, involving collecting saliva samples at specific times (e.g., immediately upon waking, 30 minutes post-waking, before bed) to assess morning cortisol, CAR, and diurnal patterns. Salivary cortisol is non-invasive and reflects the biologically active free cortisol.

**Serum/plasma cortisol:** Blood samples were used in some studies, providing a measure of total cortisol (free and bound).

**Urinary cortisol:** Less common, but sometimes used for measuring cortisol over a longer period (e.g., 24-hour collection).

**Sleep measurements:**

**Self-reported questionnaires:** These were frequently used to assess subjective sleep quality. Common instruments included:

* **Pittsburgh Sleep Quality Index (PSQI):** A widely used 19-item self-report questionnaire assessing sleep quality over the past month, with scores ranging from 0-21 (lower scores indicate better sleep). A score >5 typically indicates poor sleep quality.

* **Richards-Campbell Sleep Questionnaire (RCSQ):** Used in some clinical settings.

* **Insomnia Severity Index (ISI):** Measures the severity of insomnia.

**Actigraphy:** Wrist-worn devices that objectively measure movement and provide estimates of sleep parameters such as:

* Total sleep time (TST).

* Sleep onset latency (SOL).

* Wake after sleep onset (WASO).

* Sleep efficiency (SE).

* Actigraphy is less invasive than polysomnography and suitable for home-based, long-term monitoring.

**Polysomnography (PSG):** The gold standard for objective sleep measurement, involving monitoring brain waves (EEG), eye movements (EOG), muscle activity (EMG), and other physiological parameters during sleep. While highly accurate, it's expensive and typically conducted in a lab setting, so it was less common for long-term PA interventions.

**Sleep diaries:** Self-reported daily logs of sleep and wake times, sleep quality, and other relevant factors.

The review synthesized data from studies using a combination of these objective and subjective measures, acknowledging that different measurement tools might capture different aspects of sleep and cortisol responses.

This study was a **systematic review and meta-analysis**, which is a high-level research design that synthesizes findings from multiple individual studies to provide a more comprehensive and statistically powerful conclusion than any single study could offer.

**How they ran the study:**

1. **Systematic Search:** The researchers conducted an exhaustive search across six major electronic databases (PubMed, Web of Science, Scopus, PsycINFO, CINAHL, and SPORTDiscus) from their inception up to February 2022. This systematic approach aimed to identify all relevant published literature, minimizing publication bias (where only studies with significant findings are published).

2. **Inclusion Criteria:** Studies were included if they met specific criteria (PICO framework):

* **Population (P):** Adults (≥18 years old).

* **Intervention (I):** Any form of physical activity (PA) intervention or exposure.

* **Comparison (C):** A control group (e.g., sedentary, usual care) or different levels of PA.

* **Outcomes (O):** Measured cortisol levels (e.g., morning cortisol, CAR, diurnal cortisol) and/or sleep parameters (e.g., sleep quality, duration, latency, WASO).

* **Study Design:** Randomized Controlled Trials (RCTs), quasi-experimental studies, and observational studies were all considered.

* **Language:** English language publications only.

3. **Exclusion Criteria:** Studies were excluded if they involved animals, were review articles, dissertations (ironically, given the source), or focused on specific clinical conditions (e.g., Cushing's syndrome, severe sleep disorders) where cortisol or sleep dysregulation was a primary diagnostic criterion.

4. **Study Selection:** Two independent reviewers screened titles and abstracts, then full texts, against the inclusion/exclusion criteria. Disagreements were resolved by consensus or a third reviewer, ensuring objectivity.

5. **Data Extraction:** For each included study, relevant data were extracted, including study characteristics (e.g., design, population, PA intervention details, duration), and outcome measures (mean changes, standard deviations, sample sizes for cortisol and sleep).

6. **Quality Assessment (Risk of Bias):**

* **Randomized Controlled Trials (RCTs):** Assessed using the Cochrane Risk of Bias tool, which evaluates domains like sequence generation, allocation concealment, blinding of participants/personnel, blinding of outcome assessors, incomplete outcome data, selective reporting, and other biases. This helps gauge the internal validity of the RCTs.

* **Observational Studies:** Assessed using the Newcastle-Ottawa Scale, which evaluates the quality of non-randomized studies based on selection of participants, comparability of groups, and ascertainment of exposure/outcome.

* This step is crucial for understanding the reliability of the evidence, as studies with higher risk of bias contribute less robust evidence.

7. **Statistical Analysis (Meta-analysis):**

* Data were pooled using Review Manager (RevMan 5.4) software.

* A **random-effects model** was used for all meta-analyses. This model assumes that the true effect size varies across studies (due to differences in populations, interventions, etc.) and provides a more conservative estimate of the overall effect, which is appropriate when heterogeneity is expected.

* **Heterogeneity** (variability between study results) was assessed using the I² statistic. I² values of 0-40% indicate low heterogeneity, 30-60% moderate, 50-90% substantial, and 75-100% considerable. High heterogeneity suggests that combining studies might be less appropriate or that subgroup analyses are needed.

* Results were presented as **standardized mean differences (SMD)** with 95% confidence intervals (CIs) for continuous outcomes. SMD allows for combining studies that used different scales to measure the same outcome. An SMD of 0.2 is considered a small