Associations of metabolic heterogeneity of obesity with the risk of dementia in middle-aged adults: three prospective studies

Table 3 Multivariable-adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) for incident dementia according to intragroup comparisons of transitions of metabolic status

BMI-metabolic phenotype transitions		ARIC			FOS			Pooled^b
Baseline category	Follow-up category	Cases/N	HR (95% CI)^a	P-value	Cases/N	HR (95% CI)^a	P-value	HR (95% CI)^a	P-value
MNNO	MNNO	537/3,088	1 (Reference)		44/950	1 (Reference)		1 (Reference)
MNNO	MANO	232/1,202	1.10 (0.94, 1.29)	0.216	28/373	1.14 (0.71, 1.85)	0.543	1.10 (0.95, 1.28)	0.197
MANO	MANO	479/2,032	1 (Reference)		69/463	1 (Reference)		1 (Reference)
MANO	MNNO	98/493	0.79 (0.63, 0.98)	0.030	12/87	0.79 (0.42, 1.45)	0.442	0.79 (0.64, 0.97)	0.026
MNO	MNO	85/390	1 (Reference)		1/72	1 (Reference)		1 (Reference)
MNO	MAO	76/391	0.92 (0.67, 1.25)	0.593	8/104	7.43 (0.93, 59.65)	0.059	2.01 (0.28, 14.56)	0.490
MAO	MAO	325/1,404	1 (Reference)		32/267	1 (Reference)		1 (Reference)
MAO	MNO	34/174	0.81 (0.57, 1.15)	0.231	2/26	1.79 (0.42, 7.56)	0.428	0.87 (0.56, 1.37)	0.553

Abbreviations: UKB UK Biobank, ARIC Atherosclerosis Risk in Communities, FOS Framingham Offspring Study, BMI Body mass index, MNNO Metabolically normal non-obesity, MANO Metabolically abnormal non-obesity, MNO Metabolically normal obesity, MAO Metabolically abnormal obesity
^aThe Cox proportional hazard models using age as the time variable were adjusted for sex (male or female), race (White or non-White), education (below college or college and above), household income (< 25 000, 25 000–49,999, and ≥ 50,000 US $/year), smoking status (current, former, or never smokers), and alcohol intake (units per week)
^bRandom-effects model was used; I² = 0% for the estimates from the two cohorts

ISSN: 1758-9193