Table 3.
Moderators | TST | WASO | Number of awakenings | SE | SL | N1% | N2% | SWS% | REM% | REML | REMD | |
---|---|---|---|---|---|---|---|---|---|---|---|---|
Sex (male%) | No. of comparisons | 22 | 16 | 15 | 20 | 19 | 18 | 19 | 24 | 24 | 22 | 6 |
No. of AD/Controls | 634/650 | 540/541 | 383/415 | 579/594 | 578/600 | 494/485 | 541/529 | 634/649 | 658/701 | 599/636 | 119/155 | |
Point estimate | 0.477 | 0.793 | 1.354 | 1.651 | −1.290 | −3.561 | −1.189 | 0.378 | −0.557 | 0.202 | −1.077 | |
SE | 1.319 | 2.410 | 0.991 | 1.836 | 0.502 | 2.168 | 1.626 | 0.989 | 1.270 | 0.743 | 0.550 | |
P | 0.718 | 0.742 | 0.172 | 0.369 | 0.010 | 0.100 | 0.464 | 0.702 | 0.661 | 0.786 | 0.0503 | |
Mean age | No. of comparisons | 22 | 16 | 15 | 19 | 19 | 18 | 19 | 24 | 24 | 22 | 6 |
No. of AD/Controls | 636/651 | 540/541 | 383/415 | 571/585 | 580/601 | 496/486 | 543/530 | 636/650 | 660/702 | 601/637 | 121/156 | |
Point estimate | −0.019 | −0.071 | −0.023 | −0.001 | 0.048 | 0.041 | 0.004 | −0.002 | 0.028 | 0.004 | 0.036 | |
SE | 0.033 | 0.056 | 0.041 | 0.053 | 0.018 | 0.057 | 0.041 | 0.037 | 0.049 | 0.027 | 0.024 | |
P | 0.567 | 0.206 | 0.575 | 0.983 | 0.008 | 0.470 | 0.922 | 0.952 | 0.570 | 0.882 | 0.133 | |
Education (years) | No. of comparisons | 8 | 8 | 9 | 7 | 6 | 7 | 8 | 12 | 11 | 8 | – |
No. of AD/Controls | 322/318 | 322/318 | 289/310 | 278/273 | 287283 | 256/254 | 303/298 | 391/409 | 366/384 | 287/304 | – | |
Point estimate | 0.053 | 0.019 | 0.151 | 0.030 | −0.035 | −0.473 | −0.200 | 0.080 | 0.023 | 0.032 | – | |
SE | 0.088 | 0.079 | 0.086 | 0.098 | 0.038 | 0.059 | 0.119 | 0.054 | 0.078 | 0.046 | – | |
P | 0.542 | 0.811 | 0.079 | 0.762 | 0.359 | <0.001 | 0.094 | 0.138 | 0.768 | 0.492 | – | |
MMSE score | No. of comparisons | 21 | 16 | 15 | 18 | 18 | 16 | 17 | 22 | 23 | 21 | 6 |
No. of AD/Controls | 621/636 | 540/541 | 383/415 | 556/570 | 565/586 | 456/446 | 503/490 | 596/610 | 645/687 | 586/622 | 121/156 | |
Point estimate | 0.044 | 0.023 | −0.022 | −0.027 | 0.027 | −0.011 | 0.011 | 0.049 | 0.067 | −0.026 | 0.011 | |
SE | 0.045 | 0.057 | 0.025 | 0.067 | 0.021 | 0.080 | 0.054 | 0.028 | 0.037 | 0.022 | 0.030 | |
P | 0.330 | 0.681 | 0.370 | 0.693 | 0.202 | 0.888 | 0.842 | 0.075 | 0.071 | 0.229 | 0.714 | |
Medication free | ||||||||||||
Not reported | No. of comparisons | 1 | – | 3 | – | 1 | 1 | 1 | 4 | 4 | 4 | 1 |
No. of AD/Controls | 10/10 | – | 44/66 | – | 10/10 | 10/10 | 10/10 | 54/76 | 54/76 | 54/76 | 10/10 | |
SMD | −0.062 | – | 0.895*** | – | −0.240 | 0.601 | 0.277 | −1.100*** | −1.375*** | 0.713*** | −0.749 | |
Q | 0 | – | 0.010 | – | 0 | 0 | 0 | 3.198 | 10.498* | 2.654 | 0 | |
I2 | 0 | – | 0 | – | 0 | 0 | 0 | 6.184 | 71.424 | 0 | 0 | |
Yes | No. of comparisons | 22 | 16 | 12 | 20 | 19 | 18 | 19 | 21 | 21 | 19 | 6 |
No. of AD/Controls | 634/650 | 540/541 | 339/349 | 579/594 | 578/600 | 494/485 | 541/529 | 590/583 | 614/635 | 555/570 | 119/155 | |
SMD | −0.616*** | 0.739*** | 0.471** | −0.962*** | 0.471*** | 0.831** | 0.084 | −0.818*** | −0.654** | 0.290* | −0.245 | |
Q | 101.331*** | 112.309*** | 47.307*** | 174.611*** | 27.648 | 186.036*** | 118.443*** | 120.355*** | 216.640*** | 66.595*** | 5.472 | |
I2 | 79.216 | 86.644 | 76.748 | 89.119 | 34.895 | 90.862 | 84.803 | 83.383 | 90.768 | 72.971 | 8.633 | |
Between-group difference | Q | 1.404 | – | 2.415 | – | 2.431 | 0.199 | 0.162 | 1.196 | 2.737 | 3.531 | 1.094 |
P | 0.236 | – | 0.120 | – | 0.119 | 0.656 | 0.688 | 0.274 | 0.098 | 0.060 | 0.296 | |
Adaptation night | ||||||||||||
No | No. of comparisons | 6 | 4 | 3 | 5 | 3 | 4 | 5 | 6 | 6 | 4 | – |
No. of AD/Controls | 113/114 | 93/90 | 46/46 | 102/103 | 67/68 | 61/61 | 108/105 | 113/114 | 113/114 | 78/79 | – | |
SMD | −0.526** | 0.018 | 0.054 | −0.495** | 0.657*** | 0.870*** | −0.006 | −0.635** | −0.415 | 0.164 | – | |
Q | 7.131 | 0.034 | 3.998 | 1.017 | 0.777 | 4.303 | 15.965** | 12.638* | 17.452** | 5.657 | – | |
I2 | 29.880 | 0 | 49.979 | 0 | 0 | 30.280 | 74.946 | 60.438 | 71.350 | 46.972 | – | |
Yes | No. of comparisons | 17 | 12 | 12 | 15 | 17 | 15 | 15 | 19 | 19 | 19 | 7 |
No. of AD/Controls | 531/546 | 447/451 | 337/369 | 477/491 | 521/542 | 443/434 | 443/434 | 531/545 | 555/597 | 531/567 | 129/165 | |
SMD | −0.599*** | 0.974*** | 0.663*** | −1.091*** | 0.413*** | 0.789** | 0.119 | −0.925*** | −0.862*** | 0.385** | −0.286* | |
Q | 95.029*** | 93.838*** | 44.358*** | 168.870*** | 28.124* | 179.947*** | 98.831*** | 109.629*** | 218.276*** | 67.292*** | 6.644 | |
I2 | 83.163 | 88.278 | 75.202 | 91.710 | 43.110 | 92.220 | 85.834 | 83.581 | 91.754 | 73.251 | 9.698 | |
Between-group difference | Q | 0.092 | 12.830 | 3.045 | 4.037 | 1.491 | 0.051 | 0.121 | 0.976 | 1.524 | 0.603 | – |
P | 0.792 | <0.001 | 0.081 | 0.045 | 0.222 | 0.822 | 0.728 | 0.323 | 0.217 | 0.438 | – | |
PSG scoring methods | ||||||||||||
AASM | No. of comparisons | 6 | 5 | 1 | 5 | 4 | 5 | 5 | 5 | 5 | 6 | |
No. of AD/Controls | 182/158 | 165/138 | 11/11 | 171/147 | 151/132 | 162/143 | 162/143 | 162/143 | 162/143 | 182/158 | – | |
SMD | −1.108*** | 1.386*** | 0.774 | −1.996*** | 0.488** | 2.156*** | 0.456 | −1.699** | −2.170*** | 0.587* | – | |
Q | 26.862*** | 26.104*** | 0 | 59.169*** | 4.833 | 14.677** | 65.047*** | 49.543*** | 35.402*** | 18.508** | – | |
I2 | 81.386 | 84.677 | 0 | 93.240 | 37.924 | 72.747 | 93.851 | 91.926 | 88.701 | 72.984 | – | |
R&K | No. of comparisons | 17 | 11 | 14 | 15 | 16 | 14 | 15 | 20 | 20 | 17 | 7 |
No. of AD/Controls | 462/502 | 375/403 | 372/404 | 408/447 | 437478 | 342/352 | 389/396 | 482/516 | 506/568 | 427/488 | 129/165 | |
SMD | −0.398** | 0.435** | 0.539** | −0.597*** | 0.437*** | 0.301* | −0.044 | −0.650*** | −0.385** | 0.250* | −0.286* | |
Q | 41.806*** | 31.063** | 52.508*** | 32.323** | 25.181* | 25.200* | 44.029*** | 39.866** | 62.067*** | 39.630** | 6.644 | |
I2 | 61.728 | 67.808 | 75.242 | 56.687 | 40.430 | 48.412 | 68.203 | 52.340 | 69.388 | 59.626 | 9.698 | |
Between-group difference | Q | 5.190 | 6.250 | 0.249 | 5.964 | 0.074 | 34.444 | 0.843 | 4.129 | 14.221 | 1.719 | – |
P | 0.023 | 0.012 | 0.618 | 0.015 | 0.786 | <0.001 | 0.358 | 0.042 | <0.001 | 0.190 | – |
A meta-regression or subgroup analysis (depending on whether the potential moderators were continuous or categorical variables) was conducted to determine the potential factors that could moderate heterogeneity between studies. Of which, the effects of sex, age, education, and MMSE score were explored by using a meta-regression analysis, while the effects of whether taking medication impacts sleep (Yes vs. No), adaptation night (Yes vs. No), and PSG scoring methods (AASM vs. R&K) were explored by using subgroup analysis.
AASM American Academy of Sleep Medicine, AD Alzheimer’s disease, AHI apnea-hypopnea index, Q Cochran’s Q statistic, MMSE mini-mental state examination, PSG polysomnography, REM rapid eye movement sleep, REMD REM density, REML REM latency, R&K Rechtschaffen and Kales, SMD standardized mean difference, SWS slow-wave sleep, SE sleep efficiency, SL sleep latency, TST total sleep time, WASO wake time after sleep onset.
*p < 0.05, **p < 0 .01, ***p < 0.001.