Table 1.
Polygenic scores for educational attainment are associated with criminal offending in two birth cohorts.
| E-Risk cohort (N=1,857)a | Dunedin cohort (N=898)a | |||
|---|---|---|---|---|
|
| ||||
| Bivariate | Multivariate | Bivariate | Multivariate | |
|
| ||||
| IRR (95%CI) | IRR (95%CI) | IRR (95%CI) | IRR (95%CI) | |
| Panel A. Effect of participants’ polygenic scores on their official records of offending | ||||
|
| ||||
| Low polygenic score for educational attainmentb | 1.33 (1.13, 1.55) | – | 1.21 (1.09, 1.34) | – |
|
| ||||
| Panel B. Polygenic effect on offending, accounting for a criminogenic family environment | ||||
|
| ||||
| Low polygenic score for educational attainmentb | 1.34 (1.14, 1.57) | 1.20 (1.01, 1.42) | 1.23 (1.11, 1.36) | 1.17 (1.05, 1.30) |
| Socioeconomic deprivation | 2.26 (1.88, 2.71) | 1.96 (1.58, 2.44) | 1.40 (1.25, 1.57) | 1.35 (1.19, 1.52) |
| Parental antisocial behavior | 1.49 (1.34, 1.66) | 1.18 (1.03, 1.35) | 1.16 (1.05, 1.27) | 1.07 (.96, 1.18) |
|
| ||||
| Panel C. Polygenic effect on offending, accounting for leaving school with poor qualifications | ||||
|
| ||||
| Low polygenic score for educational attainmentb | 1.33 (1.13, 1.56) | 1.19 (1.02, 1.39) | 1.21 (1.09, 1.34) | 1.14 (1.03, 1.26) |
| Leaving school with poor qualifications | 4.27 (3.27, 5.59) | 4.02 (3.05, 5.28) | 2.83 (2.31, 3.46) | 2.71 (2.21, 3.32) |
|
| ||||
| Panel D. Polygenic effect on offending, accounting for early-emerging psychological and behavioral risk factors | ||||
|
| ||||
| Low polygenic score for educational attainmentb | 1.30 (1.11, 1.53) | 1.22 (1.05, 1.42) | 1.21 (1.09, 1.35) | 1.14 (1.02, 1.27) |
| Low cognitive ability | 1.58 (1.35, 1.85) | 1.11 (.93, 1.32) | 1.27 (1.14, 1.43) | .99 (.86, 1.13) |
| Low self-control | 2.06 (1.83, 2.32) | 1.80 (1.58, 2.04) | 1.54 (1.43, 1.67) | 1.48 (1.34, 1.64) |
| Academic difficulties in primary school | 1.68 (1.47, 1.93) | 1.17 (1.00, 1.37) | 1.30 (1.16, 1.45) | 1.02 (.89, 1.17) |
| Truancy | 3.92 (2.65, 5.80) | 1.98 (1.38, 2.83) | 2.11 (1.61, 2.76) | 1.25 (.91, 1.72) |
a
Crime records data were obtained for 93% (1,857/1,999) and 98% (898/918) of the E-Risk and Dunedin study samples, respectively.
b
The polygenic score was reverse-coded in these analyses, so that a higher score indicates a greater genetic risk for low educational attainment.
IRR = Incidence-rate ratio
CI = Confidence interval.