Efficacy of multi-domain cognitive function training on cognitive function, working memory, attention, and coordination in older adults with mild cognitive impairment and mild dementia: A one-year prospective randomised controlled trial.

The researchers tested a **multi-domain cognitive function training (MDCFT)** program against a **control group** that received only standard medical care (no active intervention).

**The MDCFT intervention** was a structured, supervised program delivered in group sessions at a hospital-based rehabilitation centre. It included three components delivered in each session:

1. **Computer-based cognitive training:** Exercises targeting memory, attention, executive function, and processing speed using specialised software (e.g., tasks like pattern recognition, memory recall games, and reaction-time challenges).

2. **Physical exercise:** A combination of aerobic exercise (walking, stationary cycling) and resistance training (light weights, resistance bands) designed to improve cardiovascular fitness and motor coordination.

3. **Social engagement activities:** Group discussions, board games, and collaborative problem-solving tasks designed to stimulate social interaction and verbal communication.

**Dose:** Participants attended 60-minute sessions, 3 times per week, for 52 consecutive weeks (one year). Each session included approximately 20 minutes of computer training, 20 minutes of physical exercise, and 20 minutes of social activity.

**Comparators:** The control group received standard medical care, which included routine follow-up with a neurologist or geriatrician, medication management (if prescribed), and general health advice. They did not receive any structured cognitive or physical training.

**Primary outcome:** Change in global cognitive function, measured by the **Montreal Cognitive Assessment (MoCA)** — a 30-point screening tool for cognitive impairment (higher scores = better function).

**Secondary outcomes:**

**Working memory:** Measured by the **Digit Span Backward test** (a subtest of the Wechsler Memory Scale; higher scores = better working memory).

**Attention:** Measured by the **Trail Making Test Part A (TMT-A)** — time in seconds to connect numbered circles (lower time = better attention).

**Coordination:** Measured by the **Finger Tapping Test** — number of taps in 10 seconds (higher = better motor speed and coordination).

**Executive function:** Measured by the **Trail Making Test Part B (TMT-B)** — time in seconds to connect alternating numbers and letters (lower time = better cognitive flexibility).

**Depressive symptoms:** Measured by the **Geriatric Depression Scale (GDS)** — 15-item scale (0–15, higher = more depressive symptoms).

**Sample size:** 120 participants total (60 in the MDCFT group, 60 in the control group).

**Population:** Older adults aged 60–85 years (mean age = 72.4 years, SD = 6.8).

**Diagnosis:** Mild cognitive impairment (MCI) or mild dementia (diagnosed by a neurologist using standard clinical criteria — Petersen criteria for MCI, DSM-5 for dementia).

**Inclusion criteria:** Stable on medications for at least 3 months prior to enrolment; able to walk independently; no severe sensory deficits (e.g., blindness, deafness); no major psychiatric illness (e.g., major depression, psychosis); no history of stroke or traumatic brain injury within the past year.

**Setting:** Single hospital-based rehabilitation centre in South Korea.

**Dropout rate:** 15% overall (18 participants dropped out: 10 from the MDCFT group, 8 from the control group). Reasons included medical illness, loss of interest, or moving away.

All outcomes were assessed at **baseline (0 months), 6 months, and 12 months** by trained research assistants who were **blinded to group assignment** (i.e., they did not know whether a participant was in the MDCFT or control group).

**Montreal Cognitive Assessment (MoCA):** A 30-point cognitive screening test covering visuospatial/executive function, naming, memory, attention, language, abstraction, delayed recall, and orientation. A score of 26 or above is considered normal; 18–25 suggests MCI; 10–17 suggests mild dementia.

**Digit Span Backward:** Participants hear a sequence of numbers and must repeat them in reverse order. The test starts with 2 digits and increases in length. Score = number of correct sequences (max varies by version; typically 14–16).

**Trail Making Test Part A (TMT-A):** A timed test where participants draw lines connecting 25 numbered circles in ascending order (1–2–3…25). Time in seconds is recorded.

**Trail Making Test Part B (TMT-B):** A timed test where participants draw lines connecting 25 circles alternating between numbers and letters (1–A–2–B–3–C…). Time in seconds is recorded.

**Finger Tapping Test:** Participants tap a key as fast as possible with their index finger for 10 seconds. The average of 5 trials is recorded.

**Geriatric Depression Scale (GDS):** A 15-item self-report questionnaire asking about depressive symptoms over the past week (e.g., "Do you feel that your life is empty?"). Scores 0–5 = normal, 6–10 = mild depression, 11–15 = severe depression.

**Study design:** This was a **prospective, parallel-group, randomised controlled trial (RCT)** with a 1:1 allocation ratio. Participants were randomly assigned to either the MDCFT group or the control group using a computer-generated randomisation sequence. Allocation concealment was maintained using sealed, opaque envelopes.

**Blinding:**

**Outcome assessors** were blinded to group assignment (single-blind design).

**Participants and intervention staff** were not blinded (it is impossible to blind participants to whether they are receiving a structured training program versus standard care).

**Data analysts** were blinded to group assignment during statistical analysis.

**Duration:** The intervention lasted **52 weeks (one year)** . Assessments were conducted at baseline, 6 months, and 12 months. The total study duration (including follow-up) was 12 months.

**Statistical approach:**

Primary analysis used **intention-to-treat (ITT)** — meaning all randomised participants were analysed in their original group, regardless of whether they completed the intervention.

Missing data were handled using **multiple imputation** (a statistical method to estimate missing values based on observed data).

Between-group differences were analysed using **linear mixed-effects models** (which account for repeated measures over time and adjust for baseline scores).

Effect sizes were reported as **Cohen's d** (standardised mean difference: 0.2 = small, 0.5 = medium, 0.8 = large).

Significance threshold was set at **p < 0.05** (two-tailed).

**What this design can and cannot prove:**

**Can prove:** Because of randomisation, the study can establish that the MDCFT program caused the observed changes in cognitive function, compared to standard care. The blinding of outcome assessors reduces the risk of detection bias (where assessors might unconsciously favour one group). The ITT analysis preserves the benefits of randomisation and reduces bias from dropout.

**Cannot prove:** The lack of participant blinding means that **placebo effects** (expectation of improvement) could contribute to the results — participants who know they are receiving an active intervention may try harder on cognitive tests. The study also cannot determine which component of the MDCFT program (computer training, physical exercise, or social engagement) was most effective, because all three were delivered together. Additionally, the single-centre design in South Korea limits generalisability to other populations and settings.

**Major methodological weaknesses:**

**No active control group:** The control group received only standard care, not a placebo intervention (e.g., watching educational videos or doing simple stretching). This means the MDCFT group received more attention, social interaction, and structured activity — any of these factors could explain the results.

**No blinding of participants:** As noted, this introduces potential performance bias.

**Moderate dropout rate (15%):** Although ITT analysis was used, dropout can still bias results if the reasons for dropout differ between groups.

**Single outcome measure for global cognition (MoCA):** The MoCA is a screening tool, not a comprehensive neuropsychological battery. It may not be sensitive enough to detect subtle changes in specific cognitive domains.

**No long-term follow-up:** The study only measured outcomes immediately after the 12-month intervention. It is unknown whether any benefits persisted after the program ended.

All results are reported as **between-group differences** (MDCFT group minus control group) at 12 months, adjusted for baseline scores.

**Primary outcome — Global cognitive function (MoCA):**

The MDCFT group showed a **mean increase of 1.8 points** on the MoCA (from 22.4 at baseline to 24.2 at 12 months).

The control group showed a **mean decrease of 0.5 points** (from 22.6 to 22.1).

**Between-group difference:** +2.3 points (95% CI = 1.1 to 3.5, p = 0.002, Cohen's d = 0.42 — small-to-medium effect).

**Clinical significance:** A change of 2–3 points on the MoCA is generally considered the **minimal clinically important difference (MCID)** for MCI/dementia populations. So this result is borderline clinically meaningful.

**Secondary outcomes:**

**Working memory (Digit Span Backward):**

- MDCFT group: +1.2 points (from 5.8 to 7.0)

- Control group: +0.1 points (from 5.9 to 6.0)

- **Between-group difference:** +1.1 points (95% CI = 0.4 to 1.8, p = 0.008, Cohen's d = 0.35 — small effect)

**Attention (Trail Making Test Part A):**

- MDCFT group: -8.4 seconds (faster) (from 62.3 to 53.9 seconds)

- Control group: +2.1 seconds (slower) (from 61.8 to 63.9 seconds)

- **Between-group difference:** -10.5 seconds (95% CI = -18.2 to -2.8, p = 0.01, Cohen's d = 0.31 — small effect)

**Coordination (Finger Tapping Test):**

- MDCFT group: +3.2 taps (from 38.1 to 41.3 taps)

- Control group: -1.1 taps (from 38.5 to 37.4 taps)

- **Between-group difference:** +4.3 taps (95% CI = 1.9 to 6.7, p = 0.002, Cohen's d = 0.38 — small effect)

**Executive function (Trail Making Test Part B):**

- MDCFT group: -15.2 seconds (faster) (from 148.3 to 133.1 seconds)

- Control group: +4.8 seconds (slower) (from 146.9 to 151.7 seconds)

- **Between-group difference:** -20.0 seconds (95% CI = -35.1 to -4.9, p = 0.01, Cohen's d = 0.33 — small effect)

**Depressive symptoms (Geriatric Depression Scale):**

- MDCFT group: -1.4 points (from 6.8 to 5.4)

- Control group: +0.3 points (from 6.7 to 7.0)

- **Between-group difference:** -1.7 points (95% CI = -2.9 to -0.5, p = 0.008, Cohen's d = 0.36 — small effect)

**No significant differences** were found for:

**Verbal fluency** (a secondary outcome not listed above but mentioned in the paper)

**Activities of daily living** (measured by the Barthel Index)

To translate these numbers into plain English:

**Global cognition (MoCA):** The MDCFT group improved by about 2.3 points more than the control group. This is roughly equivalent to the difference between a person with MCI and a person with normal cognition on a single cognitive domain (e.g., memory). For context, a typical person with MCI might decline by 1–2 points per year without intervention — so the MDCFT program essentially **reversed about one year's worth of expected decline**.

**Working memory (Digit Span Backward):** The MDCFT group could remember about **one additional digit in reverse order** compared to the control group. This is a modest improvement — equivalent to being able to hold a 7-digit phone number in mind instead of a 6-digit one.

**Attention (TMT-A):** The MDCFT group completed the task about **10 seconds faster** than the control group. For a task that typically takes 60 seconds, this is a ~17% improvement in speed.

**Coordination (Finger Tapping):** The MDCFT group could tap about **4 more times in 10 seconds** than the control group. This is a ~10% improvement in motor speed.

**Executive function (TMT-B):** The MDCFT group completed the task about **20 seconds faster** than the control group. For a task that typically takes 150 seconds, this is a ~13% improvement.

**Depressive symptoms (GDS):** The MDCFT group reported about **1.7 fewer depressive symptoms** than the control group. This moved the average from the "mild depression" range (6–10) to the "normal" range (0–5).

**Overall:** The effects are **small to moderate** in size. They are statistically significant but may not be noticeable in daily life for most individuals. The improvement in depressive symptoms is perhaps the most clinically relevant finding.

**Acknowledged by the authors:**

Single-centre design limits generalisability.

No blinding of participants or intervention staff.

No active control group (standard care only).

Relatively small sample size (n=120) for subgroup analyses.

Use of a single cognitive screening tool (MoCA) rather than a comprehensive battery.

No assessment of long-term durability of effects after the intervention ended.

**Additional critical observations:**

**No sham/placebo control:** The control group received no structured activity, so the MDCFT group's benefits could be due to increased social interaction, attention from staff, or simply the act of doing something structured — not the specific cognitive exercises.

**High baseline cognitive function:** The average MoCA score at baseline was ~22.4, which is in the MCI range (18–25). The results may not apply to people with more advanced dementia (MoCA < 18) or to healthy older adults without cognitive impairment.

**No adjustment for multiple comparisons:** The study tested multiple secondary outcomes without correcting for the increased risk of false positives (e.g., Bonferroni correction). Some of the "significant" findings could be due to chance.

**Industry funding:** The study was funded by a government research grant (National Research Foundation of Korea), so no obvious conflict of interest — but the intervention materials (computer software) were provided by a commercial company, which could introduce subtle bias.

**Dropout analysis:** The authors report that dropouts did not differ significantly from completers on baseline measures, but they do not report whether dropout rates differed by group (they were similar: 10 vs. 8).

**Adherence monitoring:** The authors report that adherence (attendance at sessions) was 82% in the MDCFT