Table 3.
Individual cognitive test performance by quintile of total quantity and variety in fruit and vegetable intake from the Boston Puerto Rican Health Study, 2004–9† (Mean values with their standard errors)
Mean‡ | SD‡ | Fruit and vegetable intake | P for trend§ | Fruit and vegetable variety | P for trend§ | |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
|
|||||||||||||||
Quintile 1 | Quintile 3 | Quintile 5 | Quintile 1 | Quintile 3 | Quintile 5 | |||||||||||
|
|
|
|
|
|
|||||||||||
Mean | SE | Mean | SE | Mean | SE | Mean | SE | Mean | SE | Mean | SE | |||||
Word-list immediate recall | 37·4 | 10·6 | −0·08 | 0·07 | −0·09 | 0·07 | −0·05 | 0·07 | 0·81 | −0·17 | 0·07 | −0·11 | 0·07 | 0·08* | 0·07 | 0·006 |
Word-list percentage recognition | 29·4 | 6·14 | 0·01 | 0·08 | −0·12 | 0·07 | −0·1 | 0·08 | 0·074 | −0·24 | 0·08 | −0·01 | 0·07 | 0·14*** | 0·07 | <0·001 |
Word-list percentage retention | 82·6 | 37·5 | −0·05 | 0·08 | −0·03 | 0·07 | −0·02 | 0·08 | 0·42 | −0·18 | 0·08 | −0·04 | 0·07 | 0·01 | 0·08 | 0·049 |
Stroop percentage | 34·0 | 18·0 | −0·15 | 0·08 | −0·06 | 0·08 | −0·1 | 0·08 | 0·79 | −0·06 | 0·08 | −0·13 | 0·07 | −0·07 | 0·08 | 0·55 |
Letter fluency | 23·7 | 10·3 | 0·06 | 0·07 | −0·08 | 0·07 | −0·05 | 0·07 | 0·23 | −0·22 | 0·07 | −0·09 | 0·07 | 0·18*** | 0·07 | <0·001 |
Digit span forward | 7·26 | 1·98 | 0·06 | 0·08 | 0·16 | 0·07 | −0·01 | 0·08 | 0·61 | −0·11 | 0·07 | 0·08 | 0·07 | 0·10 | 0·07 | 0·007 |
Digit span backward | 3·35 | 1·57 | −0·09 | 0·07 | 0·01 | 0·07 | 0·03 | 0·08 | 0·36 | −0·12 | 0·07 | −0·03 | 0·07 | 0·09 | 0·07 | 0·039 |
Clock drawing | 2·07 | 1·02 | −0·07 | 0·07 | −0·13 | 0·07 | 0·1 | 0·08 | 0·041 | −0·1 | 0·07 | −0·18 | 0·07 | 0·14 | 0·07 | 0·009 |
Figure-copying weighted | 10·1 | 7·6 | 0·03 | 0·07 | 0·04 | 0·07 | 0·15 | 0·07 | 0·28 | 0·05 | 0·07 | −0·01 | 0·07 | 0·16 | 0·07 | 0·073 |
Mean values were significantly different compared with the lowest quintile after Tukey’s adjustment for multiple comparisons:
P<0·05,
P<0·001.
Results of the omitted quintiles generally follow the pattern of those of the shown quintiles, but are not presented for simplicity; adjusted means (with their standard errors) of cognitive performance scores (standardised with a mean of 0 and a standard deviation of 1) according to quintiles of total quantity and variety in fruit and vegetable intake were calculated using general linear models (PROC GLM in SAS 9.2), respectively, after adjustment for each other (original continuous values) and age (years), sex, educational attainment (none or primary school, secondary school, high school, or college or higher), household income below the threshold (yes/no), acculturation score (continuous), BMI (kg/m2), total energy intake (kJ/d), smoking status (never, past smoker or current smoker), alcohol use (never, moderate or high), physical activity score (continuous), supplement use (yes/no), taking more than five types of medications within the past 12 months (yes/no), activities of daily living scores (continuous), hypertension (yes/no), diabetes (yes/no) and fruit and vegetable variety for fruit and vegetable intake and vice versa.
Sample mean and standard deviation of the corresponding original cognitive test score.
P values for trend were calculated using total quantity and variety in fruit and vegetable intake with PROC GLM in SAS 9.2, respectively.