Skip to main content

Table 4 Studies reporting outcome of global cognitive functioning

From: Functional social support and cognitive function in middle- and older-aged adults: a systematic review of cross-sectional and cohort studies

Author (year)

Dimension of FSS

Coefficient (CI or P-value)

Interpretation

Study Design: Cross-Sectional

 Alpass et al., 2004 [27]

Satisfaction with FSS

0.034 (p-value not reported)

Satisfaction with social support is positively but not significantly associated with cognitive function

 Gow et al., 2013) [34]

Satisfaction with FSS

positive direction of association (p = 0.278)

Satisfaction with social support is positively but not significantly associated with cognitive function

 Keller-Cohen et al., 2006 [40]

Satisfaction with FSS

Quantitative data for this variable not reported

Satisfaction with social relationships did not predict performance on Composite Cognistat or BNT

 Bourgeois et al., 2020 [23]

Perceived / Subjective

1.72 (p = sig)

Perceived social support significantly positively associated with better outcome on MoCA

 Conroy et al., 2010 [29]

Perceived / Subjective

OR = 1.3 (p = 0.175)

Low perceived social support (+ widowed and lives alone) positively associated with possible cognitive impairment

 Ficker et al., 2002 [24]

Perceived / Subjective

3.589 (p < 0.001)

Cognitively impaired elders perceived their social support as significantly less adequate than did the cognitively intact participants

 Krueger et al., 2009 [43]

Perceived / Subjective

0.068 (p = 0.003)a

Small significant positive association between perceived support and global cognitive function

 Lee & Waite, 2018 [45]

Perceived / Subjective

Female—0.65 (p < 0.05)

Male – no association

Significant positive effect of social support on cognition only in female participants. No association in male participants

 Mehrabi & Béland, 2021 [46]

Perceived / Subjective

Partner—0.275 (0.028, 0.522)

Children – no association

Friends – no association

Extended family – no association

Low perceived social support from partner significantly positively associated cognitive impairment. No association between perceived support from children, friends, or extended family and cognitive function

 Oremus et al., 2019 [6]

Perceived / Subjective

 

Proportion of participants with low cognitive function greater among persons who reported low perceived social support (and vice versa)

 Saenz et al., 2020 [54]

Perceived / Subjective (from spouse)

Husbands: 0.02 (0.01,0.03)

Wives: 0.00 (-0.01, 0.01)

Perceived social support from wife significantly positively associated with the husband’s cognitive ability

 Yeh & Liu, 2003 [58]

Perceived / Subjective (from friends)

0.11 (p = 0.005)

Perceived positive support from friends is significantly and positively associated with cognitive function

 Zank & Leipold, 2001 [61]

Perceived / Subjective

R2 = 0.085 (p < 0.05)

Perceived social support positively and significantly associated with cognitive function

 Zhu et al., 2012 [63]

Perceived / Subjective

0.020 (p < 0.05)

Total perceived support positively and significantly associated with cognitive function

 Zullo et al., 2021 [66]

Perceived / Subjective

OR = 0.93 (0.70, 1.24)

Individuals with subjective cognitive decline scored higher on the MSPSS indicating greater perceived social support

 Deng & Liu, 2021 [30]

Emotional

Relatives/friends/neighbors: OR = 0.219 (0.154, 0.311)

Children: OR = 0.400 (0.293, 0.546)

Spouse: OR = 0.242 (0.160, 0.366)

Emotional support from relatives / friends / neighbors, children, or spouse significantly associated with a reduced risk of cognitive impairment

 Harling et al., 2020 [36]

Emotional

0.72 (0.63, 0.82)

Emotional support significantly associated with decreased risk of cognitive impairment

 Kim et al., 2019 [41]

Emotional

4.160 (p = 0.002)

Emotional support significantly positively associated with cognitive function

 Murayama et al., 2019 [48]

Emotional

Male: OR = 0.46 (0.24, 0.86) **

Female: OR = 0.59 (0.35, 0.99) **

Higher emotional support significantly associated with decreased risk of cognitive impairment

 Nakamura et al., 2019 [49]

Emotional

-0.02 (p = 0.04)

Higher emotional social support significantly associated with better cognitive scores

 Okabayashi et al., 2004 [50]

Emotional

Spouse: 0.02 (p < 0.05)

Children: 0.05 (p < 0.05)

Others: 0.01 (p < 0.05)

Emotional support from spouse, children, or others all significantly positively associated with cognitive function

 Pillemer & Holtzer, 2016 [51]

Emotional

1.620 (0.343, 2.897)

Emotional support positively associated with cognitive function

 Weng et al., 2020 [56]

Emotional

OR = 1.68 (1.37 to 2.06)

Insufficient emotional support significantly associated with increased reporting of subjective cognitive decline

 Deng & Liu, 2021 [30]

Instrumental

OR = 0.242 (0.630, 0.804)

Instrumental (financial) support significantly associated with decreased risk of cognitive impairment

 Harling et al., 2020 [36]

Instrumental

0.73 (0.64, 0.82)

Instrumental support significantly associated with decreased risk of cognitive impairment

 Millán-Calenti et al., 2013 [47]

Instrumental

OR = 1.04 (0.27, 4.0) b

Small positive association between instrumental support and cognitive function

 Murayama et al., 2019 [48]

Instrumental

Male: OR = 0.43 (0.22, 0.83) b

Female: OR = 0.62 (0.30, 1.28) b

Higher instrumental support associated with decreased risk of cognitive impairment. Significant association in males

 Nakamura et al., 2019 [49]

Instrumental

0.00 (p = 0.97)

No association between instrumental support and cognitive function

 Pillemer & Holtzer, 2016 [51]

Instrumental

-0.235 (-1.535, 1.066)

Tangible support has a small negative association with cognitive function

 Ge et al., 2017 [32]

Emotional + Instrumental

R2 = 0.11 (p < 0.001)

Emotional and instrumental support significantly positively associated with cognitive function

 Gow et al., 2007 [33]

Emotional + Instrumental

0.14 (p < 0.01)

Emotional and instrumental support significantly positively associated with IQ

 Holtzman et al., 2004 [38]

Emotional + Instrumental

0.25 (p < 0.0005)

Emotional and instrumental support significantly positively associated with cognitive function

 Pillemer & Holtzer. 2016 [51]

Positive Interaction

B = 1.8883 (0.595, 3.171)

Positive social interaction positively associated with cognitive function

 Pillemer & Holtzer, 2016 [51]

Affectionate

B = -0.093 (-1.369, 1.183)

Affectionate social interaction not associated with cognitive function

 Rashid et al., 2016 [53]

FSS

OR = 2.6 (1.2–5.4)

Increased risk of cognitive impairment among individuals with poor social support

 Jang et al., 2020 [39]

Family Solidarity

0.00

No association between family solidarity and cognitive function

Study Design: Cohort

 Hughes et al., 2008 [79]

Satisfaction with FSS

0.09 (p = 0.22)

Positive association between satisfaction with social support and cognitive function

 Bowling et al., 2016 [70]

Perceived / Subjective

Family: -0.01 (-0.30, 0.27)

Friend: 0.02 (-0.29, 0.32)

Small negative association between perceived support from family and cognitive function. Small positive association between perceived support from friends and cognitive function

 Chen & Zhou, 2020 [73]

Perceived / Subjective

OR = 2.09 (p < 0.001)

Social isolation significantly associated with cognitive impairment

 Eisele et al., 2012 [75]

Perceived / Subjective

F-ratio = 2.114

Positive association between perceived support and cognitive function

 Howrey et al., 2015 [78]

Perceived / Subjective

Rapid decline: 1.89 (p < 0.001)

Slow decline: 0.25

Stable: 0.35

In rapid decline group, social support significantly associated with increases in MMSE

 Hudetz et al., 2010 [26]

Perceived / Subjective

0.01 (p = 0.64)

Small positive association between perceived support and cognitive function

 Kats et al., 2016 [80]

Perceived / Subjective

African Americans: -0.01 (-0.14, 0.12); Caucasians: 0.01 (-0.05, 0.05)

Small negative association between perceived support and cognitive function in African American population. Small positive association between perceived support and cognitive function in Caucasian population

 Luo et al., 2021 [86]

Perceived / Subjective

b = 1.90 (p = 0.050)

Quality of social relationships significantly predicts cognitive function

 Moreno et al., 2022 [88]

Perceived / Subjective

0.066 (p < 0.001)

Significant positive association between perceived social support and cognitive function

 Pais et al., 2021 [91]

Perceived / Subjective (from friends)

HR = 0.77 (0.635, 0.933)

Perceived social support from friends significantly associated with a reduced risk of cognitive impairment

 Bedard & Taler, 2020 [69]

Emotional

OR = 0.97 (0.95, 0.99)

Emotional support had a small but significant protective effect against cognitive decline

 Chen & Chang, 2016 [72]

Emotional

Starting high and declining: 0.87 (0.71, 1.07)

Starting low and declining: 0.77 (0.60, 0.99)

Emotional social support had a significant protective effect in the starting low and declining group compared with the high-stable group. (Protective but not statistically significant effect in starting high and declining group)

 Ellwardt et al., 2013 [76]

Emotional

0.03 (intercept), 0.40 (slope), p = 0.06

Emotional support positively associated with cognitive function

 Holtzman (2004) [38]

Emotional

Continuous model: 0.15 (p < 0.005)

Categorical model: 0.18 (p < 0.004)

Emotional support was a significant predictor of MMSE scores

 Hughes et al., 2008 [79]

Emotional

-0.05 (p = 0.45)

Small negative association between emotional support and cognitive function

 Noguchi et al., 2019 [89]

Emotional

-0.42 (p = 0.462)

Emotional support negatively associated with cognitive function

 Pillemer et al., 2019 [17]

Emotional

Incident cognitive decline: HR = 1.43 (0.94,2.18)

Cognitive decline – males: HR = 1.62 (0.93,2.84)

Cognitive decline – females: HR = 1.39 (0.68,2.84)

Emotional support positively associated with cognitive decline

 Seeman et al., 2001 [96]

Emotional

1.26 (p = 0.07)

Emotional support positively associated with cognitive function

 Thomas & Umberson, 2018 [98]

Emotional (from children)

-0.004, p < 0.05

Support from children related to fewer cognitive limitations

 Bedard & Taler, 2020 [69]

Instrumental

OR = 0.98 (0.94, 1.02)

Instrumental support had a small protective effect against cognitive decline

 Dickinson et al., 2011 [74]

Instrumental

0.578 (p = 0.0333)

Instrumental support significantly positively associated with cognitive function

 Ellwardt et al., 2013 [76]

Instrumental

-0.01 (intercept), -0.02 (slope)

Small negative association between instrumental support and cognitive function

 Hughes et al., 2008 [79]

Instrumental

0.01 (p = 0.88)

Small positive association between instrumental support and cognitive function

 Noguchi et al., 2019 [89]

Instrumental

0.38 (p = 0.642)

Instrumental support positively associated with cognitive function

 Pillemer et al., 2019 [17]

Instrumental

Incident cognitive decline: HR = 1.75 (1.12,2.72)

Cognitive decline – males: HR = 1.91 (1.00,3.62)

Cognitive decline – females: HR = 1.78 (0.94,3.35)

Instrumental support positively associated with cognitive decline

 Seeman et al., 2001 [96]

Instrumental

-0.04 (p = 0.93)

Small negative association between instrumental support and cognitive function

 Yin et al., 2020 [100]

Instrumental (sick care)

HR = 0.795 (0.550, 1.148)

Instrumental support negatively associated with cognitive impairment

 Noguchi et al., 2019 [89]

Emotional + Instrumental

Co-residing family: 0.28, p = 0.813

Non-residing family and relatives: 0.51 (p = 0.283)

Neighbours and friends: 1.23, p = 0.006

Significant positive association between emotional and instrumental social support from neighbours and friends and MoCA-J scores. Negative association between emotional and instrumental support from co-residing family or non-residing family and relatives and cognitive function

  1. CI Confidence Interval, FSS Functional Social Support, HR Hazard Ratio, MoCA-J Japanese version of the Montreal Cognitive Assessment, OR Odds Ratio, RR Relative Risk
  2. aEffects merged using Borenstein (Borenstein et al., 2009)
  3. bInverse of point estimate and confidence limits taken to convert outcome to yes vs. no or high vs. low