Skip to main content

Table 4 Unadjusted and adjusted beta coefficients (95% confidence intervals (95% CI)) for association between egocentric social network characteristics, all depression and psychiatric symptoms subscale (PHQ 4) (n = 87)

From: Self-reported depression and social support are associated with egocentric network characteristics of HIV-infected women of color

Depression symptoms (Complete PHQ 9)
 Unadjusted modelsAdjusted modelsa
Network characteristicCoefficient95% CICoefficient95% CI
Functional network characteristics
Advice tie count−0.62−1.40, 0.17− 0.57− 1.36, 0.22
Emotional tie count− 0.65− 1.46, 0.15− 0.58− 1.41, 0.24
Financial tie count− 1.15− 2.14, − 0.15−1.17−2.15, − 0.18
Transportation tie count− 0.51− 1.49, 0.47−0.59− 1.57, 0.38
Structural network characteristics
Family tie count−0.69−1.60, 0.23−0.72− 1.66, 0.22
Non-family tie count−0.47−1.04, 0.09− 0.53−1.08, 0.03
Alter knows ego’s HIV status−0.63−1.40, 0.15− 0.59− 1.37, 0.19
Alter’s positive HIV status−2.32−4.57, − 0.07− 2.25−4.49, − 0.01
Total network size− 0.31− 0.89, 0.27− 0.30−0.69, 0.28
< Daily talk - primary2.180.95, 5.301.59−1.48, 4.65
< Daily talk - secondary−1.62−4.47, 1.23−1.08−3.88, 1.73
Psychiatric symptoms subscale (Estimated from 4 PHQ 9 questions)
 Unadjusted modelsAdjusted models
Coefficient95% CICoefficient95% CI
Functional network characteristics
Advice tie count−0.39− 0.81, 0.03− 0.35− 0.78, 0.08
Emotional tie count− 0.41− 0.84, − 0.03−0.36− 0.81, 0.09
Financial tie count−0.67−1.20, − 0.13−0.65− 1.19, − 0.11
Transportation tie count−0.42− 0.95, 0.10−0.44− 0.97, 0.09
Structural network characteristics
Family tie count−0.53− 1.01, − 0.04−0.53− 1.04, − 0.01
Non-family tie count0.14− 0.45, 0.17−0.14− 0.45, 0.17
Alter knows ego’s HIV status−0.40− 0.81, − 0.02−0.36− 0.79, 0.07
Alter’s positive HIV status−0.64−1.88, 0.59− 0.62− 1.87, 0.64
Total network size− 0.22− 0.54, 0.08− 0.22−0.54, 0.10
< Daily talk - primary1.46−0.21, 3.121.19−0.49, 2.87
< Daily talk - secondary−0.92−2.44, 0.60− 0.72− 2.27, 0.83
  1. In social network terms, “ego” is the person responding to the survey and alter is the person who has been named or identified by the ego; a unadjusted models include only the independent and dependent variables; models adjusted for ego’s continuous age, categorical education, dichotomous insurance status