Skip to main content
. Author manuscript; available in PMC: 2016 Jan 31.
Published in final edited form as: J Behav Med. 2014 Aug 9;38(1):153–159. doi: 10.1007/s10865-014-9587-0

Table 2.

Regression coefficients from unadjusted and adjusted support-HbA1c models (N = 766)

HbA1c B [95% CI]
Demographic covariatesa
    Age −0.03 [−0.05, −0.02]
    Genderb −0.38 [−0.73, −0.04]
    Hispanic Backgroundc
        Cuban −0.20 [−0.80, 0.39]
        Puerto Rican 0.05 [−0.47, 0.56]
        Dominican −0.26 [−0.86, 0.34]
        Central American 0.01 [−0.74, 0.75]
        South American −0.24 [−1.03, 0.54]
        Other/Mixed origin −0.47 [−1.40, 0.46]
    Nativity/immigration statusd 0.42 [0.01, 0.83]
    Language of Interview −0.73 [−1.33, −0.14]
    Education 0.13 [−0.08, 0.33]
    Income 0.03 [−0.06, 0.11]
    Health insurance coveragee −0.38 [−0.87, 0.11]
    Glucose-lowering medicatione 0.49 [−0.08, 1.07]
Unadjusted Modelsf
    Structural Supportg 0.15 [0.04; 0.26]
    Functional Supporth 0.22 [0.02; 0.42]
Adjusted Modelsf
    Structural Supportg 0.08 [−0.03; 0.18]
    Functional Supporth 0.18 [0.12; 0.36]
Adjusted Conjoint Modeli
    Structural Supportg 0.05 [−0.06, 0.16]
    Functional Supporth 0.16 [−0.02, 0.34]

Notes. Analyses account for design effects and sample weights. HbA1c = Glycosylated hemoglobin.

a

Coefficients derived from model excluding support variables.

b

Gender: 0 = male, 1 = female.

c

Hispanic background variables were dummy coded with Mexican as reference group.

d

Nativity/immigration status: 1= Immigrated <10 years ago; 2 = Immigrated ≥10 years ago 3 = Born in the US mainland.

e

Health insurance coverage and medication: 0 = no, 1 = yes.

f

Support indicators evaluated in separate models.

g

Coefficients reflect difference in HbA1c per each additional social role.

h

Coefficients reflect difference in HbA1c per one SD increase in perceived support.

i

Both support indicators entered in single (conjoint) model.