Table 3.

Relative risk of high allostatic load based on perceived baseline and long-term weight discrimination in the MIDUS study

	High allostatic loada			Continuous allostatic loadb			High allostatic load (excluding BMI and WHR)
	RR	95 % CI	p value	β	SE	p value	RR	95 % CI	p value
Baseline perceived weight discriminationc
Age- and sex-adjusted	2.60	1.60, 4.23	0.0001	0.15	0.04	0.0005	1.92	1.14, 3.23	0.01
Multivariable-adjusted model 1d	2.42	1.44, 4.04	0.0008	0.14	0.04	0.001	1.78	1.02, 3.11	0.04
Multivariable-adjusted model 2e	2.31	1.38, 3.84	0.001	0.13	0.04	0.002	1.71	0.99, 2.97	0.05
Multivariable-adjusted model 3f	2.23	1.28, 3.87	0.004	0.13	0.04	0.005	1.61	0.90, 2.87	0.11
Multivariable-adjusted model 4g	2.07	1.21, 3.55	0.008	0.11	0.04	0.01	1.55	0.87, 2.75	0.13
Long-term perceived weight discriminationc,h
Age- and sex-adjusted	2.50	1.72, 3.63	<0.0001	0.21	0.04	<0.0001	1.87	1.25, 2.79	0.002
Multivariable-adjusted model 1d	2.47	1.65, 3.69	<0.0001	0.21	0.04	<0.0001	1.79	1.15, 2.78	0.01
Multivariable-adjusted model 2e	2.37	1.58, 3.56	<0.0001	0.20	0.04	<0.0001	1.73	1.11, 2.69	0.02
Multivariable-adjusted model 3f	2.27	1.45, 3.56	0.0003	0.21	0.05	<0.0001	1.66	1.03, 2.69	0.04
Multivariable-adjusted model 4g	2.16	1.39, 3.36	0.0007	0.19	0.05	0.0001	1.62	1.01, 2.62	0.047

^aHigh allostatic load was defined as greater than or equal to three dysregulated systems, and low allostatic load was defined as less than three. Allostatic load was measured at follow-up, and medication usage was included in the definition

^bFamily status was added to the continuous models as a random effect

^cOnly individuals who reported discrimination “sometimes” or “often” were coded as having experienced discrimination

^dModel 1 includes age, sex, race (white, black, other), household income (>$100,000/year), smoking status (never, former, current), educational attainment (< high school, high school graduate, some college, college graduate or more)

^eModel 2 includes covariates in model 1 plus engagement in regular exercise at least 20 min three times per week

^fModel 3 includes covariates in models 1–2 plus perceived race, sex, and age discrimination at baseline

^gModel 4 includes covariates in models 1–3 plus baseline BMI

^hLong-term weight discrimination was computed as the average value of perceived discrimination at baseline and at 10-year follow-up for those who had both measures. For individuals with no baseline measure, but with a measure at 10 years, long-term discrimination was computed as their reported discrimination at 10 years