Determinants of participation in a longitudinal two-stage study of the health consequences of the Chornobyl nuclear power plant accident

Table 1 Factors associated with children's participation at follow-up

	Participants (N = 526)^b	Non-participants (N = 73)^b	Adjusted for group OR (95% CI)	Multivariable model^c OR (95% CI)
Attended clinic at time 1	485 (92.2)	61 (83.6)	2.3 (1.2–4.7) *	--^c
Mother participated at follow-up	477 (92.3)	16 (22.5)	41.6 (21.7–79.7) ***	46.9 (23.9–92.1) ***
Background characteristics ^a
Evacuee status	265 (50.4)	34 (46.6)	--	--^c
Female child	273 (51.9)	37 (50.7)	1.1 (0.6–1.7)	--^c
Child in utero at time of accident	158 (30.0)	25 (34.2)	0.8 (0.5–1.4)	--^c
Standard of living, mean ± SD^c	3.8 ± 1.6	3.6 ± 1.7	1.1 (0.9–1.4)	--^c
University graduate (either parent)	163 (31.0)	23 (31.5)	1.0 (0.6–1.8)	--^c
Child well-being ^a
No medical checkup in past year	343 (65.2)	44 (60.3)	1.4 (0.8–2.5)	--^c
≥ 2 colds in past year	375 (71.6)	56 (76.7)	0.7 (0.4–1.33)	--^c
P-CSI (mother report), mean ± SD^d	18.6 ± 14.4	17.6 ± 13.7	1.1 (0.8–1.4)	--^c
Childhood behavioral problems	106 (20.2)	14 (19.2)	1.1 (0.6–2.0)	--^c
CSI (child self-report), mean ± SD^d	16.5 ± 16.0	16.3 ± 18.7	1.0 (0.8–1.3)	--^c
Days absent from school
None	170 (34.1)	24 (35.3)	1.0	--^c
1–10 days	227 (45.6)	32 (47.1)	1.0 (0.6–1.8)	--^c
> 10 days	101 (20.3)	12 (17.6)	1.2 (0.6–2.5)	--^c
		linear trend	χ²(1) = 0.2
Mother's Chornobyl risk perception ^a
Child's health perceived as very affected by Chornobyl	231 (43.9)	32 (43.8)	1.0 (0.6–1.6)	--^c
Distrust of authorities, mean ± SD^d	3.1 ± 0.6	3.2 ± 0.6	0.8 (0.7–1.1)	--^c

^a All variables are based on maternal reports except the child-rated CSI and the school record information on absences. No interactions between these variables and evacuee status were found to be statistically significant.
^b Values are numbers (percentage, calculated from total column) except when noted.
^c The final multivariable model was obtained following a stepwise backwards likelihood-ratio elimination procedure.
^d Continuous variables were standardized in logistic regression models so that the ORs may be interpreted as the odds of participation for a one standard deviation increase in the predictor.
* p < 0.05; ** p < 0.01; *** p < 0.001

ISSN: 1471-2288