Table 3.

Associations of Social Determinant of Health Risk Factors and Clusters with HbA1c and Blood Pressure Values by Linear Regression Models

OutcomeFactors/Clusters*Coefficient95% CIP value
Diabetes model (total n = 2193 and n = 1906 with complete data)
HbA1cIntercept8.7558.0769.434<0.001
BMI−0.007−0.0170.0030.184
Age−0.024−0.031−0.017<0.001
Female (ref: male)−0.166−0.330−0.0020.047
Social background score0.1140.0490.1780.001
Social insecurities score0.1660.0840.247<0.001
Insurance/employment score0.1240.0190.2290.020
Federal poverty level score−0.034−0.4160.3490.864
Social isolation score−0.314−0.553−0.0750.010
Housing status score−0.113−0.4250.1990.479
Blood pressure model (total n = 3865 and n = 3338 with complete data)
Systolic blood pressureIntercept118.445113.955122.935<0.001
BMI0.1750.1120.237<0.001
Age0.049−0.0020.0990.058
Female−1.844−2.899−0.7890.001
Social background score0.4980.0730.9240.022
Social insecurities score1.0260.4861.567<0.001
Insurance/employment score−0.138−0.8080.5320.687
Federal poverty level−0.097−2.5652.3710.939
Social isolation0.284−1.2771.8460.721
Housing status0.001−1.9971.9980.999
Diastolic blood pressureIntercept87.07184.17089.972<0.001
BMI0.036−0.0040.0770.080
Age−0.222−0.254−0.189<0.001
Female (ref: male)−1.566−2.247−0.884<0.001
Social background score−0.030−0.3050.2450.832
Social insecurities score0.7360.3871.085<0.001
Insurance/employment score−0.460−0.893−0.0270.038
Federal poverty level0.023−1.5721.6170.978
Social isolation−0.635−1.6440.3740.217
Housing status−0.119−1.4091.1720.857
  • * The actual effect of each composite cluster is the estimated coefficient multiplied with the actual number of positive SDoH factors in that composite cluster.

  • The P values < 0.05 are bolded.

  • Abbreviations: SDoH, social determinants of health; BMI, body mass index; CI, confidence interval.