The experimental group received a uniform exercise - cognitive intervention at a community health centre or a geriatric activity station. The intervention lasted for a total of 12 weeks at a frequency of once every fortnight, with each intervention lasting 60-90 minutes. At the end of the intervention, the participants were followed up by the researchers for a total of 12 weeks, at a frequency of once every fortnight.

The intervention consisted of physical-motor activities as well as integrated cognitive training.

Participants in the control group did not undergo any intervention and kept their old lifestyle unchanged. The control group was asked to avoid cognitive training and exercise training beyond the routine throughout the investigation.

The intervention included physical activity and integrated cognitive training. The training was conducted in three sessions. The first session was a warm-up and wake-up preparation, in which participants massaged facial acupoints in a seated position to light music. In the middle phase of aerobic exercise, participants raised their legs in a standing position and performed mental exercises to fast-paced music. Session 1: tapping the five fingers of each hand in turn; Section 2: The thumb of both hands in turn to tap the rest of the fingers; Third section: hands five fingers cross fist; Section 4: The five fingers of both hands are stretched out in turn, and then retracted in turn. The final session was a relaxation session in which participants performed inspiratory and expiratory exercises to light music. The integrated cognitive training consisted of 15 minutes of attention training, 15 minutes of memory training, and 15 minutes of executive function training.

The Shapiro-Wilk test was used to assess the normality of the data and found that the MMSE scores of the two groups of MCI participants were not normally distributed at all three time points, so we used medians and quartiles to describe the data. Generalized estimating equations were used to analyze the trends in MMSE scores at each time point for both groups. As seen in Table 4, the MMSE scores in the intervention group showed an increasing trend over time compared to the control group. There was a significant time effect (χ² = 27.259, p < 0.001) and a significant group effect (χ² = 21.267, p < 0.001) for the MMSE scores of the two groups. This indicates that the intervention used in this study was effective in improving the cognitive level of MCI patients in the intervention group. In order to visualize the trend in MMSE scores, a line graph was used to represent the change.

At different time points, the MMSE scores of the two study groups were not normally distributed, therefore, the between-group differences in the MMSE scores of the two study groups at the three time points were statistically analyzed using a two independent samples non-parametric rank sum test. The results showed a significant difference in MMSE scores between the two groups of participants at the end of the intervention (Z = 5.406, p < 0.001). At the end of follow-up, there was a significant difference in MMSE scores between the two groups of participants (Z = 5.403, p < 0.001).

As the MMSE scores of MCI older adults in both the intervention and control groups were not normally distributed at all three time points, a paired rank sum test was used to compare the change in MMSE scores between the two groups of participants before and after the intervention. The results showed that MMSE scores in the intervention group were significantly higher at the end of the intervention compared to the initial level (Z = -4.287, p < 0.001). MMSE scores in the intervention group were significantly higher at the end of follow-up compared to initial levels (Z = -4.251, p < 0.001).