Skip to main content

Table 3 Comparison of characteristics of studies that examined the effect of different levels of alcohol consumption

From: Long-term effects of alcohol consumption on cognitive function: a systematic review and dose-response analysis of evidence published between 2007 and 2018

 Study dates (years from T0; bold= ‘baseline’)Alcohol category* (bold=referent)Cognitive function (bold=selected result)
Study ID
(sample size at T0; % female)
Age (T0)T0T1T2T3T4T5TXAge final follow-upLength of follow-up from b/l0 to < 10 g/week≥ 10 g/week to < 10 g/day≥ 10 to < 20 g/day≥ 20 to < 30 g/day≥ 30 to < 40 g/day≥ 40 to < 50 g/day≥ 50 g/dayGlobal functionMCI diagnosisComplex attentionExecutive functionLearning and memoryLanguagePerceptual motor abilitySocial cognitionOutcome description (details of selected result)
Arntzen 2010 [44]
(5033; 56%)**
58 (mean)AC
(~7)
     ~65~7XX       X X   SCD: learning and memory (immediate and delayed recall)
Downer 2015 [45]
(664; 56%)**
42 (mean)AC
(~28)
C
(~34)
    75 (mean)~34XX X X X  XX   GCF: Average of Z-scores on 11 tests (incl. memory, executive function, language, complex attention)
Hassing 2018 [46]
(305; 56%) critical RoB
~56–66AC
(~24)
C
(~26)
C
(~28)
C
(~30)
C
(~32)
 83 (mean)~32 XXX   X   X X GCF: MMSE score (change over time)
Heffernan 2016 [47]
(821; 55%)**
70–90A, CC
(~2)
C
(~2)
    ~74–94~4X X  X   XXXXX SCD: learning and memory (delayed recall)
Hogenkamp 2014 [48]
(652; 100%)
70A, CC (~7)     77~7XXXX     XX    SCD: executive function (TMT-B)
Horvat 2015 [49]
(28,947; 55%)**
45–69A, CC (~4)     47–78~4XWMWWM  M  X XX  SCD: learning and memory (delayed recall)
Kesse-Guyot 2012 [50]
(3088; 46%)**
45–60AC (~13)     ~58–73~13XWM WM WMWMX  XXX  GCF: Average of T-scores on 6 tests (executive function; learning & memory; language)
Kitamura 2017 [51]
(1814; 60%)**
44–79AC (~3)     ~47–82~3X X X XX       GCF: cognitive impairment (MMSE <24)
Lang 2007 [52]
(13,333; 57%)
≥ 65AC
(~4)
     ≥ 69~4XX XX  X       GCF: Binary “poor function” (bottom quintile for sum of scores on 3 tests)
McGuire 2007 [53]
(2572; 66%) critical RoB
≥ 70 (mean 76)AA, C
(~2)
C
(~2)
    ~80~2XXX    X       GCF: Binary “low” or “high” (based on cut-off score on 2 tests)
Piumatti 2018 [54]
(13,342; 55%)
40–73A, CC
(~5)
     ~45–78~5continuous variable  X     SCD: complex attention (mean reaction time over 7 test trials)
Richard 2017 [55]
(1334; 54%)**
55–84AC
(~4)
C
(~4)
C
(~4)
C
(~4)
C
(~4)
C
(~4)
8514±8
(med-ian)
XWMWMWMX       GCF: Binary “impaired” or “healthy” based on cut-off on Z-scores (age, sex, education adjusted MMSE)
Sabia 2011 [56]
(4073; 0%)**
~45–55AA
(1)
A
(1)
A
(1)
A
(1)
A
(1)
A (T6-9), C (T10, 11)~55–651MMMMM    X     SCD: complex attention (digit symbol substitution test)
Sabia 2014 [57]
(7153; 29%)**
35–55AA
(~5)
A, C (~5)C (~5)C (~5)  55–80~10XW
M
WW
M
 M X  XX   GCF: Average of Z-scores on 4 tests (executive function; learning & memory)
Samieri 2013 [58]
(6174; 100%)
≥ 60AC(5.6)C(~2)C(~2)   ~79 (mean)~10XX X   X   X   GCF: Average of Z-scores on 5 tests (TICS, learning & memory, language)
Solfrizzi 2007 [59]
(1445; 44%)
65–84A, CC     ~68–873.5XX XX   X      MCI: hazard ratio for incident MCI (Petersen diagnostic criteria)
Stott 2008 [60]
(5,804; 52%)**
70–82A, CC
(~1)
C
(~1)
C
(~1)
C
(~1)
  ~73–853.2XXM    X X X   GCF: MMSE score
Wardzala 2018 [61]
(486; 75%)
≥ 80A, CC
(~1)
C
(~1)
C
(~1)
C
(~1)
C
(~1)
C
(~1)
~86–91~5–7XW
M
 W
M
   X XXXX  GCF: MMSE score (change over time)
  1. Abbreviations: GCF: Global cognitive function, SCD: Specific cognitive domain, MCI: Mild cognitive impairment, A: Measure of alcohol intake, C: Measure of cognition, M: Men, W: Women, X: Men and women. DSST: Digit symbol substitution test, DSCT: Digit symbol coding test, TMT: Trail making test, MMSE: Mini-Mental State Exam, TICS: Telephone Interview for Cognitive Status, RoB: Risk of bias
  2. *Based on mean or median alcohol consumption, or midpoint of specified category (if average consumption for group not reported). For the largest dose category, if an upper bound was not specified then the assigned dose value was calculated as the lower bound of the largest dose category plus the width of the previous (second-to-largest) category. Alcohol category: Bolded entry (X, M or W) was used as referent in study
  3. **Included in dose-response analysis for ‘females’ only, ‘males’ only, or ‘females and males’
  4. Follow up until age 86—number of follow-up measures depends on age at baseline