Table 3.

Results from four separate logistic regression analyses for each scoring method on clinical CBCL scores and treatment status, taking the identification by the PCH into account (n = 707); adjusted odds ratios (OR) and 95% Confidence Intervals (95% CI), for identification by PCH and the SDQ scoring method

	Clinical CBCL total clinical score	Clinical CBCL Internalising problems score	Clinical CBCL Externalising problems score	Being treated for psychosocial problems
	Adjusted OR (95% CI)	Adjusted OR (95% CI)	Adjusted OR (95% CI)	Adjusted OR (95% CI)
Elevated SDQ TDS cut-off ≥ 14
PCH detects problems yes (vs no)*	2.59 (1.23–5.47)	3.58 (2.01–6.40)	1.92 (0.85–4.32) ns	7.20 (3.46–14.99)
Elevated SDQ score yes (vs no)	49.93 (23.67–105.36)	12.92 (6.97–23.92)	31.69 (13.96–71.98)	5.54 (2.78–11.06)
Parent-defined difficulties
PCH detects problems yes (vs no)*	3.57 (1.85–6.89)	3.56 (1.99–6.35)	2.79 (1.30–5.97)	6.72 (3.20–14.11)
Elevated SDQ score yes (vs no)	16.33 (6.20–24.45)	12.52 (6.52–24.03)	10.44 (4.85–22.48)	7.02 (3.44–14.34)
High sub-scale score/impairment
PCH detects problems yes (vs no)*	3.47 (1.76–6.85)	3.96 (1.80–6.29)	2.55 (1.17–5.57)	7.72 (3.72–16.00)
Elevated SDQ score yes (vs no)	26.76 (12.53–57.17)	13.79 (6.74–28.21)	19.64 (8.81–43.81)	5.68 (2.66–12.11)
Combination of these three scores$
PCH detects problems yes (vs no)*	1.96 (0.95–4.06) ns	2.63 (1.43–4.85)	1.73 (0.78–3.83) ns	5.75 (2.70–12.25)
Elevated SDQ score yes (vs no)	36.04 (16.66–77.96)	15.17 (8.26–27.88)	21.29 (9.24–49.06)	6.75 (3.40–13.41)

* adjusted odds ratios: adjusted for the scoring method.

$ Using the adapted Clinical SDQ cut-off; ≥ 14.