. Author manuscript; available in PMC: 2020 Feb 1.

Published in final edited form as: Health Psychol. 2019 Feb;38(2):122–132. doi: 10.1037/hea0000710

Table 4.

Linear mixed models with nested fixed effects examining predictors of HbA1c measurements during the subsequent 12-month period.

	12-Month HbA1c^a

	n = 301

	B	SE	p

Model 1
Age	−.004	.006	.522
Race^b	−.199	.155	.201
Insulin use	.418	.125	<.001
Prior glycemic control	.555	.039	<.001
Time	.001	.010	.997
Health literacy	−.020	.122	.874
Model 2
Relation to MHS^c	.132	.140	.348
MHS co-residence^d	−.123	.130	.347
Respect for MHS	.021	.064	.742
Support satisfaction	−.053	.071	.456
Model 3
Autonomy support	−.257	.085	.003

Note. MHS = Main Health Supporter. Results of each model reported while controlling for variables entered in the previous model. Bold typeface indicates significant associations using the False Discovery Rate corrected critical p value (p <.014).

713 HbA1c measurement over the 12-months following the survey clustered within 301 participants

White = 1 vs. other race = 0.

Spouse or partner = 1 vs. other family member or friend = 0.

MHS lives inside of participant’s home = 1 vs. MHS lives outside of the participant’s home = 0.