NEJM sub-journal: Improve the cognition of people with mild cognitive impairment, crossword puzzles are very good!

*For medical professionals only

As the saying goes, "the fist does not leave the hand, the song does not leave the mouth", the more you practice, the better a certain ability will be, and the same is true for the function of the brain
.
In a study of 22 populations, "mentally draining" activities such as reading, playing chess, and crossword puzzles reduced the risk of dementia by 46 percent over a seven-year follow-up period [1].

What about playing computer games? As a way for us to relax and have fun, some computer games do need to "use our brains" to pass the level
.
Studies have shown that for people with normal cognitive function, playing some computer games with cognitive training functions can help maintain good cognitive performance [2].

However, for people with impaired cognitive function, the protective effect of playing computer games on cognitive function is unclear [3].

Recently, D.
Smith from the New York State Psychiatric Institute P.
Devanand's team, published research in NEJM Evidence, a sub-journal of the New England Journal of Medicine (NEJM)[4], compared the protective effects
of computer games and crossword puzzles on cognitive function in people with mild cognitive impairment (MCI).

The results showed that after 78 weeks of training respectively, the cognitive function of the subjects in the computer game group decreased slightly from baseline, while the crossword group showed significant improvement
.
In addition, the decrease in hippocampal volume and cortical thickness was also more pronounced
in the computer game group than in the crossword group.

This suggests that for people with impaired cognitive function, mental exercise may still be traditional and good
.
Technology changes lives, but people with impaired cognitive function may not understand cognitive training computer games well (especially older adults who have no computer exposure) and are therefore less likely to benefit
from them.

Screenshot of the first page of the paper

This clinical trial (NCT03205709) was conducted at Columbia University/New York State Psychiatric Institute and Duke University Medical Center from November 2017 to February 2020 and included people aged 55 to 95 years with MCI (early MCI and late MCI according to the Merriam-III logical memory delayed recall score), The Brief Mental Status Test Scale scores between
23 and 30.

Participants were randomized to a 1:1 ratio to either computer games or crossword puzzle groups and stratified
by research centre, age (≤ 70 years and > 70 years), and early or late MCI.
Their training is divided into two phases, the first is the first 12 weeks, completing an initial training session of 30 minutes four times a week, and the second is followed by 4 weeks of 30-minute intensive training
4 times a week, 30 minutes.

Each training session of the computer game group consists of 6 randomly selected modules out of 18 available modules, which cover memory tasks, matching tasks, spatial recognition tasks, and processing speed tasks
.
Over time, the difficulty of computer games increased
with the subject's gaming and cognitive performance.

The difficulty of the crossword group was set to medium (the equivalent of The New York Times Thursday's crossword puzzle) and the difficulty did not increase over time, and if the participants completed the crossword puzzle 15 minutes before the 30-minute session, a second crossword
was performed.

Crossword puzzle (Image source: Columbia University)

The primary endpoint of the study was change from baseline in scores on 11 Alzheimer's Assessment Scale-Cognitive Scales (ADAS-Cog, between 0-70, higher scores, more severe cognitive impairment) at 78 weeks
.
Secondary endpoints included change in composite z-score on neuropsychological tests from baseline at 78 weeks, a concise UC San Diego task-based ability to live ability test (UPSA, 0 to 100, higher scores, better functional performance), bystander-reported social functioning activity questionnaire (FAQ, 0-30 points, higher scores are associated with greater impairment in activities of daily living) Score
.

In this study, the subject's apolipoprotein Ee4 genotype, hippocampal volume calculated from baseline MRI, and baseline University of Pennsylvania olfactory recognition test score were used as covariates to control for the impact
of these factors on the primary endpoint analysis.
In addition, an exploratory analysis
was performed on changes in hippocampal volume and cortical thickness from baseline to week 78.

The study ultimately included 107 participants, 51 in the computer games group and 56 in the crossword group, with a mean age of 71.
2 years, of whom 58% were women
.
There were no significant differences in the demographics of participants between the two groups at baseline
.

At week 78, the mean ADAS-Cog score in the computer game group increased by 0.
4 points (9.
53 to 9.
93 points) from baseline, while the mean ADAS-Cog score in the crossword group decreased by nearly 1 point (9.
59 to 8.
61 points)
from baseline.

The smallest squares mean (LS mean, corrected mean after adjusting for imbalances between groups) showed a slight increase in cognitive impairment in the computer game group and improvement in the crossword group (LS mean difference between the two groups: -1.
44 points, P=0.
04).

。

Further analysis found that as early as the end of the initial training, that is, week 12, the difference in ADAS-Cog score change LS mean was already -1.
35 points
between the two groups.

The mean LS of change in ADAS-Cog scores in both groups

At the same time, the results also showed that by week 78, 25.
0% of subjects with ADAS-Cog scores improving by more than 2 points were in the computer game group and 37.
3%
in the crossword group.
In addition, in subjects with late MCI, the change in ADAS-Cog score was manifested as cognitive decline in the computer game group and improvement in cognitive function in the crossword group at week 12 (LS mean difference: -1.
92) and week 78 (LS mean difference: -2.
45), but this difference
was not observed in the early MCI group.

This suggests that crossword puzzles are better than computer games in improving cognition in participants with MCI, and that the better the cognitive function, the more pronounced
this difference is.

For secondary endpoints, there was no significant difference in composite neuropsychological test scores or UPSA scores between the two groups, while the corrected change in FAQ scores showed a greater decline in function in the computer game group than in the crossword group (LS mean difference, -1.
08 points).

The difference in FAQ scores in both groups was LS mean

The researchers also analysed the risk of progression to dementia in two groups of MCI participants, with 6 of 56 participants (10.
7%) in the crossword group being diagnosed with dementia during the follow-up period, compared with 8 of 51 participants in the computer game group (15.
7%) were diagnosed with dementia (odds ratio 0.
65).

Conversely, 17 of the 56 participants in the crossword group (30.
4%) returned to normal cognition from MCI, compared with 12 of the 51 participants in the game group (23.
5%) returned to normal cognition (odds ratio 1.
41).

This once again proves that crossword puzzles are better than computer games
in improving cognition in MCI subjects.

For hippocampal volume and cortical thickness, after adjusting for baseline levels, at week 78, the decrease in hippocampal volume in the computer play group was greater than in the crossword group (LS mean difference: 34.
07), and the decrease in cortical thickness was equally pronounced (LS mean difference: 0.
02).

Overall, this study demonstrates that crossword puzzles are superior to computer games
in improving cognition in people with MCI.

Although computer games are more entertaining, they are more complex than traditional mental activities such as crossword puzzles, which is not conducive to the training
of cognitive functions for patients with pre-existing cognitive impairment.
Therefore, for patients with different levels of cognitive impairment, appropriate mental exercise methods
should be selected.

References

1.
Valenzuela MJ, Sachdev P: Brain reserve and dementia: a systematic review.
Psychol Med 2006, 36(4):441-454.

2.
Preiss M, Shatil E, Cermakova R, Cimermanova D, Ram I: Personalized cognitive training in unipolar and bipolar disorder: a study of cognitive functioning.
Front Hum Neurosci 2013, 7:108.

3.
Hill NT, Mowszowski L, Naismith SL, Chadwick VL, Valenzuela M, Lampit A: Computerized Cognitive Training in Older Adults With Mild Cognitive Impairment or Dementia: A Systematic Review and Meta-Analysis.
Am J Psychiatry 2017, 174(4):329-340.

4.
Devanand DP, Goldberg TE, Qian M, Rushia SN, Sneed JR, Andrews HF, Nino I, Phillips J, Pence ST, Linares AR et al: Computerized Games versus Crosswords Training in Mild Cognitive Impairment.
NEJM Evidence, 0(0):EVIDoa2200121.

Responsible editorYing Yuyan