Table 5.
Distal outcome scores at age 3 years, and family history and diagnostic group, by latent class
| Class | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| (a) Normative | (b) High attentional control | (c) Low attentional control | (d) Plateaued attention development | (e) Low focus, high shifting | Omnibus test with full dataset [excluding infants with missing data] | |||||||
| Sample | 1 | 2 | 1 | 2 | 1 | 2 | 1 | 2 | 1 | 2 | 1 | 2 |
| Mean SRS raw total (SE) | 22.54 (1.24) | 25.96 (0.93) | 51.52 (9.61) | 38.74 (5.17) | 30.68 (4.36) | 34.99 (3.19)a | 94.58 (7.60)a,b,c | 72.67 (8.69)a,b c,e†, | – | 42.71 (12.23) | Χ2(3) = 151.83, p < .001* |
χ2 (4) = 41.84, p < .001* [χ2 (4) = 43.76, p < .001*] |
| Mean CBCL-ADHD raw total (SE) | 4.55 (0.40) | 2.61 (0.26) | 3.63 (0.75) | 4.34 (1.37) | 3.22 (0.63) | 6.64 (0.89)a,b | 8.18 (0.97)a,b,c | 5.44 (1.21)a,b | – | 4.96 (1.21) | χ2 (3) = 17.54, p = .001* |
χ2 (4) = 28.07, p = .001* [χ2 (4) = 16.72, p = .002*] |
| Vineland ABC score (SE) | 99.65 (1.32) | 94.42 (1.04) | 103.11 (3.12) | 97.58 (3.33) | 96.33 (3.37) | 88.48 (1.91)a‡,b | 79.35 (4.61)a,b,c | 82.62 (4.89)a‡,b‡ | – | 89.06 (3.46) | χ2 (3) = 23.33, p < .001* |
χ2 (4) = 13.84, p = .008* [χ2 (4) = 9.52, p = .049] |
| TL | 138 (68.3%) | 114 (75.5%) | 18 (22.0%) | 19 (12.6%) | 7 (8.5%) | 10 (6.6%) | 1 (1.2%) | 7 (4.6%) | – | 1 (0.7%)1 | ||
| EL-all¶ | 138 (65.1%) | 183 (70.1%) | 25 (11.8%) | 17 (6.5%) | 36 (17.0%) | 31 (11.9%) | 13 (6.1%) | 16 (6.1%) | – | 14 (5.4%)1 | ||
| EL-no ASD | 83 (66.4%) | 84 (69.4%) | 15 (12.0%) | 8 (6.6%) | 22 (17.6%) | 17 (14.0%) | 5 (4.0%) | 4 (3.3%)2 | – | 8 (6.6%) | ||
| EL-ASD | 51 (63.0%) | 22 (55.0%) | 9 (11.1%) | 2 (5.0%) | 13 (16.0%) | 6 (15.0%) | 8 (9.9%) | 5 (12.5%)2 | – | 5 (12.5%) | ||
Superscripts indicate which groups have lower SRS and CBCL-ADHD scores, or higher Vineland ABC scores, based on significant (p < .05) chi-square tests
*Significant after a Benjamani–Hochberg correction for 3 family-wise tests, with a false discovery rate of 5%
†Only when infants with missing data were excluded.
‡Only when the full dataset was used (i.e. not excluding infants with missing data)
§Proportions in each class are calculated within outcome group (i.e. for each row). Values are based on most-likely class estimates
¶Comprises EL-ASD, EL-No ASD and EL-Outcome not known
1Significant association between likelihood group and membership of the low focus, high shifting class: p = .025 (exploratory analysis)
2Significant association between ASD diagnosis and membership of the plateaued attention development class: p = .043 (exploratory analysis)