Table 1

Risk of developing cerebral oedema in relation to age, sex, season, and whether diabetes was newly diagnosed or already known

					Logistic regression analysis
					Unadjusted analysis			Adjusted analysis^1-150
Classification factor	Levels of factor	Total episodes of DKA	No. episodes of DKA affected by cerebral odema	Risk (per 1000)	Odds ratios	95% CI	p value for the likelihood ratio test of homogeneity across factor levels		Odds ratios	95% CI	p value for the likelihood ratio test of homogeneity across factor levels
Sex	Female	1707	9	5.3	1^1-151				1^1-151
	Male	1233	11	8.9	1.7	(0.7, 4.1)	0.239		1.5	(0.6, 3,6)	0.391

Age (y)	0–3	303	4	13.2	2.9	(0.7, 5.5)			1.5	(0.5, 4.4)
	4–7	387	2	5.2	1.1	(0.2, 5.6)	0.373^1-152		0.7	(0.1, 3.9)	0.738^1-152
	8–11	933	8	8.6	1.9	(0.8, 10.4)	0.175^1-153		1.5	(0.4, 5.9)	0.791^1-153
	12–15	1317	6	4.6	1^1-151				1^1-151

Old or newly diagnosed		Old	1845	7	3.8	1^1-151				1^1-151
Old or newly diagnosed		New	1095	13	11.9	3.2	(1.3, 7.9)	0.012		2.9	(1.1, 8.0)	0.039

Season		Spring	661		5		7.6		1^1-151					1^1-151
Season		(Mar–May)	661		5		7.6		1^1-151					1^1-151
		Summer	666		3		4.5		0.6		(0.1, 2.5)		0.824^1-152	0.6	(0.1, 2.3)	0.786^1-152
		(Jun–Aug)	666		3		4.5		0.6		(0.1, 2.5)		0.824^1-152	0.6	(0.1, 2.3)	0.786^1-152
		Autumn	766		5		6.5		0.9		(0.2, 3.0)			0.9	(0.2, 3.0)
		(Sep–Nov)	766		5		6.5		0.9		(0.2, 3.0)			0.9	(0.2, 3.0)
		Winter	847		7		8.3		1.1		(0.3, 3.5)			1.1	(0.3, 3.4)
		(Dec–Feb)	847		7		8.3		1.1		(0.3, 3.5)			1.1	(0.3, 3.4)
Total		2940	20	6.8

↵1-150 Adjusted for each of the other factors in the table.
↵1-151 Reference or baseline category.
↵1-152 Alternative hypothesis heterogeneity in odds ratios across different levels of factor.
↵1-153 Alternative hypothesis linear trend in odds ratios across different levels of factor.