Table 1—

Multivariate models for probability of absenteeism and actual hours absent in the past 4 weeks

Probability of absenteeism (any hours absent)
Actual hours absent
Male (n = 110)Female (n = 108)Male (n = 110)Female (n = 108)
A1C (%)
7.0–7.99−0.04 (−0.37 to 0.28)−0.08 (−0.43 to 0.27)1.40 (−3.81 to 6.62)2.86 (−2.38 to 8.11)
8.0–8.990.32 (0.04 to 0.61)*−0.04 (−0.38 to 0.31)5.96 (−0.24 to 12.16)2.04 (−2.52 to 6.61)
9.0–9.990.24 (−0.10 to 0.58)0.27 (−0.09 to 0.64)3.80 (−3.49 to 11.09)7.90 (0.96 to 14.84)*
≥10.00.35 (0.08 to 0.62)*0.62 (0.45 to 0.79)4.73 (−1.58 to 11.03)10.26 (1.92 to 18.60)*
LDL cholesterol ≥100 (mg/dl)0.26 (0.01 to 0.52)*0.13 (−0.20 to 0.45)3.80 (−0.65 to 8.24)0.38 (−3.03 to 3.79)
SBP ≥130 or DBP ≥80 (mmHg)−0.04 (−0.30 to 0.22)−0.18 (−0.44 to 0.07)0.59 (−2.81 to 3.99)−1.85 (−4.62 to 0.92)
• Partial derivatives of probability of any absenteeism from a probit model (marginal effects) and partial derivatives of hours absent from a tobit model (marginal effects: unconditional expected values) with respect to independent variables are reported with SEs of marginal effects in parentheses. The derivatives are computed as the difference in probabilities and the difference in hours at the observed censoring rate as the dummy variable takes on the values 0 and 1, with the other variables at the sample means. All models include the following covariates: age, race/ethnicity, education, marital status, Charlson comorbidity score, years since diagnosis, insulin usage, BMI, occupational categories, and usual weekly hours worked.

• * P < 0.05;

• P < 0.10;

• P < 0.01. DBP, diastolic blood pressure; SBP, systolic blood pressure.