NEIGHBORHOOD CONTEXT AND AGE 5 IQ SCORES: ONLINE SUPPLEMENT
Assessing Neighborhood Social Context and Individual Exposures to Polycyclic Aromatic Hydrocarbon Exposure as Determinants of Age 5 Cognitive Test Scores
Online Supplement
This online supplement provides results from several sensitivity analyses. Sensitivity analyses were considered (1) adjusting for urinary PAH metabolites collected at age 3, (2) using age 3 addresses for estimation of neighborhood characteristics, and (3) using inverse probability weighting to account for loss to follow-up.
Post-natal assessment of PAH exposure was assessed through urinary metabolites at age 3 years. Nine specific PAH metabolite concentrations were summed, and the resulting variable log-transformed. Specific gravity of each sample was included as a covariate to account for differences in sample dilution. Associations of prenatal PAH and prenatal neighborhood characteristics with age 5 cognitive test scores were similar in models adjusting for age 3 urinary PAH metabolites (Table S1).
Geocoding and geographic analyses were performed for the home address reported at the time of the year 3 visit in parallel with the methods used for prenatal addresses reported at the time of recruitment. More than half of the cohort participants (54%) had moved between recruitment and the year 3 visit. Characteristics of a 1-km network buffer around each year 3 home address were used to predict age 5 cognitive test scores, yielding associations which were somewhat attenuated (Table S2). Low neighborhood English proficiency remained a statistically significant predictor of lower total and performance scores.
Potential bias due to incomplete follow-up was addressed through inverse probability weighting as a sensitivity analysis. Inverse probability weights were estimated for each outcome using logistic regression models predicting successful follow-up and completion of the WPPSI-R test at age 5. The logistic model was evaluated using a Hosmer-Lemeshow test and visual inspection across deciles, both of which suggested good fit to the data. Inverse probability weights ranged from 1.1 to 6.3. Associations of prenatal PAH exposure with cognitive test scores were attenuated following inverse probability weighting (Table S3). Associations of low neighborhood English proficiency with lower scores remained statistically significant.
NEIGHBORHOOD CONTEXT AND AGE 5 IQ SCORES: ONLINE SUPPLEMENT
Table S1. Sensitivity analysis adjusting for postnatal PAH exposure
Model 1PAH only / Model 2
+ Household variables / Model 3
+ neighborhood poverty / Model 4
+ neighborhood education / Model 5
+ neighborhood low English proficiency / Model 6
+ neighborhood inadequate plumbing
TOTAL AGE 5 WPPSI SCORE
High prenatal PAH / β / -4.40 / -4.07 / -4.03 / -3.81 / -3.50 / -4.08
95% CI / (-7.43 to -1.37) / (-7.11 to -1.02) / (-6.78 to -1.28) / (-6.91 to -0.71) / (-6.91 to -0.10) / (-7.09 to -1.06)
Prenatal Questionnaire in English / β / 0.88 / 0.60 / 1.00 / 1.20 / 0.68
95% CI / (-1.97 to 3.73) / (-2.37 to 3.57) / (-1.90 to 3.89) / (-1.67 to 4.08) / (-2.47 to 3.83)
English Primarily Spoken in the Home / β / 6.18 / 5.97 / 5.93 / 6.13 / 6.41
95% CI / (-5.99 to 18.36) / (-6.61 to 18.54) / (-6.38 to 18.24) / (-5.73 to 17.98) / (-5.67 to 18.48)
Both English and Spanish in the Home / β / 7.39 / 7.49 / 7.23 / 7.12 / 7.47
95% CI / ( 2.82 to 11.95) / ( 2.51 to 12.47) / ( 2.51 to 11.95) / ( 2.44 to 11.79) / ( 3.08 to 11.86)
Building Dilapidation Index / β / 0.37 / 0.26 / 0.34 / 0.34 / 0.36
95% CI / (-0.57 to 1.31) / (-0.63 to 1.16) / (-0.56 to 1.24) / (-0.57 to 1.24) / (-0.59 to 1.32)
Neighborhood Percent Poverty / β / -1.51
95% CI / (-3.51 to 0.48)
Neighborhood Percent < High School Education / β / -0.91
95% CI / (-2.37 to 0.56)
Neighborhood Percent Low English Proficiency / β / -1.67
95% CI / (-3.06 to -0.28)
Neighborhood Percent Inadequate Plumbing / β / 0.53
95% CI / (-0.77 to 1.83)
VERBAL SCORE
High prenatal PAH / β / -4.14 / -4.02 / -4.02 / -4.24 / -4.04 / -4.04
95% CI / (-6.43 to -1.86) / (-6.57 to -1.47) / (-6.61 to -1.44) / (-6.84 to -1.65) / (-6.82 to -1.26) / (-6.58 to -1.51)
Prenatal Questionnaire in English / β / 1.09 / 1.13 / 0.99 / 1.08 / 0.57
95% CI / (-1.28 to 3.46) / (-1.11 to 3.37) / (-1.38 to 3.35) / (-1.37 to 3.53) / (-1.96 to 3.10)
English Primarily Spoken in the Home / β / 7.84 / 7.87 / 8.06 / 7.84 / 8.42
95% CI / (-1.12 to 16.80) / (-1.07 to 16.81) / (-0.90 to 17.02) / (-1.16 to 16.84) / (-0.37 to 17.21)
Both English and Spanish in the Home / β / 8.57 / 8.56 / 8.70 / 8.58 / 8.79
95% CI / ( 3.49 to 13.65) / ( 3.52 to 13.59) / ( 3.86 to 13.55) / ( 3.57 to 13.59) / ( 3.85 to 13.74)
Building Dilapidation Index / β / 0.79 / 0.80 / 0.82 / 0.79 / 0.77
95% CI / (-0.12 to 1.70) / (-0.12 to 1.73) / (-0.09 to 1.72) / (-0.12 to 1.71) / (-0.16 to 1.71)
Neighborhood Percent Poverty / β / 0.21
95% CI / (-1.66 to 2.08)
Neighborhood Percent < High School Education / β / 0.79
95% CI / (-0.41 to 1.98)
Neighborhood Percent Low English Proficiency / β / 0.06
95% CI / (-1.44 to 1.57)
Neighborhood Percent Inadequate Plumbing / β / 1.40
95% CI / ( 0.26 to 2.53)
PERFORMANCE SCORE
High prenatal PAH / β / -3.29 / -2.75 / -2.69 / -2.09 / -1.79 / -2.75
95% CI / (-6.97 to 0.40) / (-6.29 to 0.79) / (-5.56 to 0.19) / (-5.66 to 1.49) / (-5.83 to 2.25) / (-6.28 to 0.79)
Prenatal Questionnaire in English / β / -0.29 / -0.80 / 0.01 / 0.26 / -0.15
95% CI / (-4.61 to 4.03) / (-5.69 to 4.09) / (-3.85 to 3.88) / (-3.54 to 4.06) / (-4.35 to 4.05)
English Primarily Spoken in the Home / β / 2.77 / 2.38 / 2.12 / 2.68 / 2.61
95% CI / (-9.87 to 15.42) / (-10.86 to 15.61) / (-10.69 to 14.92) / (-9.35 to 14.70) / (-9.99 to 15.22)
Both English and Spanish in the Home / β / 4.45 / 4.64 / 4.06 / 3.99 / 4.39
95% CI / (-1.95 to 10.85) / (-2.05 to 11.33) / (-2.75 to 10.86) / (-2.54 to 10.53) / (-2.05 to 10.83)
Building Dilapidation Index / β / -0.22 / -0.42 / -0.31 / -0.28 / -0.22
95% CI / (-1.38 to 0.94) / (-1.39 to 0.56) / (-1.35 to 0.73) / (-1.38 to 0.82) / (-1.40 to 0.96)
Neighborhood Percent Poverty / β / -2.77
95% CI / (-5.05 to -0.49)
Neighborhood Percent < High School Education / β / -2.32
95% CI / (-3.90 to -0.75)
Neighborhood Percent Low English Proficiency / β / -2.85
95% CI / (-4.07 to -1.64)
Neighborhood Percent Inadequate Plumbing / β / -0.39
95% CI / (-1.91 to 1.14)
Note: PAH indicates Polycyclic Aromatic Hydrocarbons, for which high values have defined as above the median of 2.26 ng/m3; the building dilapidation index has a range from 0 to 4 based on the number of reported problems (e.g., holes in walls, leaking pipes); neighborhood characteristics (percent poverty, percent not completing high school, percent low English language proficiency, percent inadequate plumbing) have been estimated for a 1-km network buffer surrounding the prenatal home address; each neighborhood characteristic has been rescaled to a z-score such that the coefficients shown are for a 1 standard deviation increase; all regression models account for clustering within community districts and adjust the child’s sex, ethnicity, maternal education, maternal IQ score, environmental tobacco smoke exposure in the home, quality of the caretaking environment, home English language exposure, and postnatal PAH urine metabolites (log-transformed PAH, specific gravity); bold face indicates statistical significance (p<0.05); N = 205
Table S2. Sensitivity analysis adjusting measuring neighborhood characteristics based on year 3 home address
Model 1PAH only / Model 2
+ Household variables / Model 3
+ neighborhood poverty / Model 4
+ neighborhood education / Model 5
+ neighborhood low English proficiency / Model 6
+ neighborhood inadequate plumbing
TOTAL AGE 5 WPPSI SCORE
High prenatal PAH / β / -3.45 / -3.42 / -3.55 / -3.53 / -3.46 / -3.54
95% CI / (-6.63 to -0.27) / (-7.17 to 0.32) / (-6.62 to -0.47) / (-6.76 to -0.30) / (-6.75 to -0.17) / (-6.67 to -0.40)
Prenatal Questionnaire in English / β / 1.91 / 2.66 / 2.72 / 2.40 / 2.72
95% CI / (-1.79 to 5.60) / (-1.54 to 6.86) / (-1.24 to 6.67) / (-1.24 to 6.04) / (-1.46 to 6.90)
English Primarily Spoken in the Home / β / 7.57 / 8.29 / 8.28 / 9.45 / 8.27
95% CI / (-5.73 to 20.87) / (-4.72 to 21.29) / (-4.68 to 21.24) / (-2.60 to 21.51) / (-4.85 to 21.39)
Both English and Spanish in the Home / β / 3.93 / 3.68 / 3.59 / 4.09 / 3.61
95% CI / (-1.92 to 9.78) / (-2.32 to 9.67) / (-2.74 to 9.93) / (-1.87 to 10.06) / (-2.66 to 9.89)
Building Dilapidation Index / β / 0.04 / -0.00 / -0.00 / 0.07 / -0.00
95% CI / (-0.81 to 0.89) / (-0.90 to 0.90) / (-0.88 to 0.87) / (-0.86 to 1.01) / (-0.89 to 0.89)
Neighborhood Percent Poverty / β / 0.15
95% CI / (-1.16 to 1.45)
Neighborhood Percent < High School Education / β / -0.09
95% CI / (-1.34 to 1.17)
Neighborhood Percent Low English Proficiency / β / -1.88
95% CI / (-3.35 to -0.41)
Neighborhood Percent Inadequate Plumbing / β / 0.02
95% CI / (-0.98 to 1.01)
VERBAL SCORE
High prenatal PAH / β / -3.90 / -4.08 / -4.14 / -4.14 / -4.10 / -4.15
95% CI / (-6.98 to -0.81) / (-7.83 to -0.33) / (-7.32 to -0.97) / (-7.43 to -0.84) / (-7.35 to -0.84) / (-7.32 to -0.98)
Prenatal Questionnaire in English / β / 1.85 / 2.47 / 2.53 / 2.36 / 2.48
95% CI / (-0.72 to 4.42) / (-0.76 to 5.70) / (-0.22 to 5.27) / (-0.29 to 5.02) / (-0.52 to 5.48)
English Primarily Spoken in the Home / β / 7.94 / 8.64 / 8.62 / 9.23 / 8.63
95% CI / (-4.37 to 20.26) / (-3.43 to 20.70) / (-3.30 to 20.54) / (-1.92 to 20.39) / (-3.56 to 20.83)
Both English and Spanish in the Home / β / 3.86 / 3.85 / 3.79 / 4.04 / 3.83
95% CI / (-2.30 to 10.02) / (-2.45 to 10.16) / (-2.66 to 10.23) / (-2.13 to 10.21) / (-2.55 to 10.21)
Building Dilapidation Index / β / 0.50 / 0.45 / 0.45 / 0.49 / 0.45
95% CI / (-0.37 to 1.38) / (-0.46 to 1.35) / (-0.43 to 1.33) / (-0.45 to 1.42) / (-0.44 to 1.34)
Neighborhood Percent Poverty / β / 0.15
95% CI / (-1.09 to 1.39)
Neighborhood Percent < High School Education / β / 0.00
95% CI / (-1.44 to 1.45)
Neighborhood Percent Low English Proficiency / β / -0.98
95% CI / (-2.34 to 0.39)
Neighborhood Percent Inadequate Plumbing / β / 0.24
95% CI / (-0.83 to 1.31)
PERFORMANCE SCORE
High prenatal PAH / β / -1.67 / -1.36 / -1.53 / -1.53 / -1.43 / -1.52
95% CI / (-4.89 to 1.55) / (-5.02 to 2.29) / (-4.72 to 1.66) / (-4.82 to 1.76) / (-4.88 to 2.03) / (-4.81 to 1.77)
Prenatal Questionnaire in English / β / 1.45 / 2.06 / 2.19 / 1.83 / 2.15
95% CI / (-3.62 to 6.51) / (-3.12 to 7.23) / (-3.13 to 7.50) / (-3.05 to 6.70) / (-3.27 to 7.58)
English Primarily Spoken in the Home / β / 5.30 / 5.96 / 5.91 / 7.20 / 5.93
95% CI / (-6.37 to 16.96) / (-5.37 to 17.29) / (-5.65 to 17.47) / (-3.48 to 17.87) / (-5.63 to 17.50)
Both English and Spanish in the Home / β / 2.94 / 2.56 / 2.47 / 2.95 / 2.45
95% CI / (-3.19 to 9.08) / (-3.79 to 8.91) / (-4.33 to 9.26) / (-3.38 to 9.28) / (-4.28 to 9.18)
Building Dilapidation Index / β / -0.57 / -0.58 / -0.57 / -0.49 / -0.57
95% CI / (-1.56 to 0.43) / (-1.63 to 0.48) / (-1.59 to 0.45) / (-1.57 to 0.58) / (-1.61 to 0.46)
Neighborhood Percent Poverty / β / 0.28
95% CI / (-1.28 to 1.83)
Neighborhood Percent < High School Education / β / 0.18
95% CI / (-0.73 to 1.09)
Neighborhood Percent Low English Proficiency / β / -2.01
95% CI / (-3.36 to -0.66)
Neighborhood Percent Inadequate Plumbing / β / 0.08
95% CI / (-0.68 to 0.85)
Note: PAH indicates Polycyclic Aromatic Hydrocarbons, for which high values have defined as above the median of 2.26 ng/m3; the building dilapidation index has a range from 0 to 4 based on the number of reported problems (e.g., holes in walls, leaking pipes); neighborhood characteristics (percent poverty, percent not completing high school, percent low English language proficiency, percent inadequate plumbing) have been estimated for a 1-km network buffer surrounding the year 3 home address; each neighborhood characteristic has been rescaled to a z-score such that the coefficients shown are for a 1 standard deviation increase; all regression models account for clustering within community districts and adjust the child’s sex, ethnicity, maternal education, maternal IQ score, environmental tobacco smoke exposure in the home, quality of the caretaking environment, and home English language exposure; bold face indicates statistical significance (p<0.05); N = 265
Table S3. Sensitivity analysis using inverse probability weights to account for loss to follow-up
Model 1PAH only / Model 2
+ Household variables / Model 3
+ neighborhood poverty / Model 4
+ neighborhood education / Model 5
+ neighborhood low English proficiency / Model 6
+ neighborhood inadequate plumbing
TOTAL AGE 5 WPPSI SCORE
High prenatal PAH / β / -2.37 / -2.46 / -2.47 / -2.17 / -1.90 / -2.55
95% CI / (-6.52 to 1.79) / (-6.77 to 1.86) / (-6.79 to 1.86) / (-6.40 to 2.07) / (-6.50 to 2.70) / (-6.73 to 1.63)
Prenatal Questionnaire in English / β / 2.83 / 2.66 / 2.87 / 2.98 / 2.46
95% CI / (-0.34 to 6.01) / (-0.33 to 5.66) / (-0.27 to 6.02) / (-0.22 to 6.17) / (-0.82 to 5.74)
English Primarily Spoken in the Home / β / 5.21 / 5.20 / 5.29 / 5.54 / 5.32
95% CI / (-7.45 to 17.86) / (-7.66 to 18.06) / (-7.70 to 18.27) / (-7.04 to 18.13) / (-7.08 to 17.71)
Both English and Spanish in the Home / β / 3.21 / 3.29 / 3.22 / 3.47 / 3.17
95% CI / (-2.25 to 8.66) / (-2.74 to 9.32) / (-2.66 to 9.09) / (-2.16 to 9.10) / (-1.73 to 8.06)
Building Dilapidation Index / β / 0.30 / 0.27 / 0.27 / 0.27 / 0.26
95% CI / (-0.73 to 1.34) / (-0.80 to 1.35) / (-0.77 to 1.32) / (-0.71 to 1.25) / (-0.77 to 1.29)
Neighborhood Percent Poverty / β / -0.67
95% CI / (-3.16 to 1.82)
Neighborhood Percent < High School Education / β / -1.15
95% CI / (-2.63 to 0.33)
Neighborhood Percent Low English Proficiency / β / -2.30
95% CI / (-3.04 to -1.55)
Neighborhood Percent Inadequate Plumbing / β / 1.04
95% CI / (-0.50 to 2.58)
VERBAL SCORE
High prenatal PAH / β / -2.94 / -3.17 / -3.15 / -3.18 / -2.74 / -3.32
95% CI / (-7.38 to 1.51) / (-7.67 to 1.34) / (-7.65 to 1.35) / (-7.54 to 1.18) / (-7.73 to 2.25) / (-7.68 to 1.04)
Prenatal Questionnaire in English / β / 2.32 / 2.55 / 2.32 / 2.43 / 1.72
95% CI / ( 0.18 to 4.46) / ( 0.67 to 4.44) / ( 0.17 to 4.47) / ( 0.27 to 4.59) / (-0.48 to 3.92)
English Primarily Spoken in the Home / β / 5.42 / 5.43 / 5.41 / 5.68 / 5.60
95% CI / (-7.72 to 18.55) / (-7.46 to 18.32) / (-7.78 to 18.61) / (-7.26 to 18.61) / (-7.17 to 18.37)
Both English and Spanish in the Home / β / 4.35 / 4.23 / 4.35 / 4.55 / 4.29
95% CI / (-3.05 to 11.74) / (-2.99 to 11.45) / (-3.05 to 11.74) / (-2.95 to 12.05) / (-2.27 to 10.84)
Building Dilapidation Index / β / 0.76 / 0.80 / 0.76 / 0.73 / 0.69
95% CI / (-0.36 to 1.88) / (-0.30 to 1.90) / (-0.37 to 1.90) / (-0.34 to 1.81) / (-0.39 to 1.76)
Neighborhood Percent Poverty / β / 0.91
95% CI / (-1.15 to 2.97)
Neighborhood Percent < High School Education / β / 0.06
95% CI / (-1.26 to 1.39)
Neighborhood Percent Low English Proficiency / β / -1.76
95% CI / (-3.30 to -0.21)
Neighborhood Percent Inadequate Plumbing / β / 1.68
95% CI / (-0.15 to 3.50)
PERFORMANCE SCORE
High prenatal PAH / β / -0.67 / -0.55 / -0.58 / -0.02 / 0.01 / -0.58
95% CI / (-4.07 to 2.73) / (-4.19 to 3.10) / (-4.01 to 2.86) / (-3.61 to 3.57) / (-3.67 to 3.68) / (-4.16 to 3.01)
Prenatal Questionnaire in English / β / 2.75 / 2.17 / 2.82 / 2.89 / 2.63
95% CI / (-1.47 to 6.96) / (-2.16 to 6.51) / (-1.28 to 6.91) / (-1.33 to 7.11) / (-1.61 to 6.88)
English Primarily Spoken in the Home / β / 3.54 / 3.52 / 3.68 / 3.87 / 3.57
95% CI / (-6.00 to 13.07) / (-6.67 to 13.70) / (-6.36 to 13.72) / (-5.73 to 13.47) / (-5.86 to 13.00)
Both English and Spanish in the Home / β / 1.01 / 1.29 / 1.03 / 1.27 / 0.99
95% CI / (-3.42 to 5.43) / (-4.05 to 6.63) / (-3.93 to 5.99) / (-3.17 to 5.71) / (-3.32 to 5.31)
Building Dilapidation Index / β / -0.39 / -0.49 / -0.45 / -0.43 / -0.41
95% CI / (-1.49 to 0.70) / (-1.58 to 0.60) / (-1.48 to 0.59) / (-1.46 to 0.60) / (-1.52 to 0.70)
Neighborhood Percent Poverty / β / -2.23
95% CI / (-4.85 to 0.39)
Neighborhood Percent < High School Education / β / -2.11
95% CI / (-3.86 to -0.36)
Neighborhood Percent Low English Proficiency / β / -2.28
95% CI / (-3.93 to -0.63)
Neighborhood Percent Inadequate Plumbing / β / 0.32
95% CI / (-0.89 to 1.52)
Note: PAH indicates Polycyclic Aromatic Hydrocarbons, for which high values have defined as above the median of 2.26 ng/m3; the building dilapidation index has a range from 0 to 4 based on the number of reported problems (e.g., holes in walls, leaking pipes); neighborhood characteristics (percent poverty, percent not completing high school, percent low English language proficiency, percent inadequate plumbing) have been estimated for a 1-km network buffer surrounding the prenatal home address; each neighborhood characteristic has been rescaled to a z-score such that the coefficients shown are for a 1 standard deviation increase; all regression models use inverse probability weights and account for clustering within community districts and adjust the child’s sex, ethnicity, maternal education, maternal IQ score, environmental tobacco smoke exposure in the home, quality of the caretaking environment, and home English language exposure; bold face indicates statistical significance (p<0.05); N = 277