Skip to main content

Table 4 Fuzzy RDD estimates of the impact of a T2DM diagnosis on depression—heterogeneity analysis

From: Sick and depressed? The causal impact of a diabetes diagnosis on depression

 

(1)

(2)

(3)

 

All

Men

Women

Panel A: Full-Sample:

T2DM Diagnosis [0,1]

0.016** (0.005)

-0.001 (0.006)

0.032*** (0.008)

 Observations

34,319

16,483

17,836

 Proportion of Treated

0.67

0.69

0.66

Panel B: Weight Loss:

T2DM Diagnosis [0,1]

-0.032 (0.022)

-0.083* (0.037)

0.001 (0.025)

 Observations

6990

3111

3879

 Proportion of Treated

0.84

0.85

0.83

Panel C: No Weight Loss:

T2DM Diagnosis [0,1]

0.032* (0.015)

0.015 (0.016)

0.058* (0.025)

 Observations

6862

3242

3620

 Proportion of Treated

0.84

0.85

0.83

Panel D: Healthy Weight:

T2DM Diagnosis [0,1]

-0.002 (0.014)

-0.010 (0.013)

0.018 (0.026)

 Observations

2690

1347

1343

 Proportion of Treated

0.69

0.74

0.64

Panel E: Overweight:

T2DM Diagnosis [0,1]

0.006 (0.008)

-0.002 (0.011)

0.014 (0.016)

 Observations

10,053

5619

4434

 Proportion of Treated

0.73

0.74

0.72

Panel F: Obese:

T2DM Diagnosis [0,1]

0.024*** (0.007)

0.003 (0.008)

0.041*** (0.010)

 Observations

21,539

9502

12,037

 Proportion of Treated

0.65

0.66

0.64

Panel G: Age ≥ 60:

T2DM Diagnosis [0,1]

0.013* (0.006)

-0.003 (0.008)

0.025** (0.008)

 Observations

23,161

10,612

12,549

 Proportion of Treated

0.72

0.73

0.71

Panel H: Age < 60:

T2DM Diagnosis [0,1]

0.027** (0.009)

0.003 (0.011)

0.055*** (0.015)

 Observations

11,158

5871

5287

 Proportion of Treated

0.57

0.62

0.51

  1. The Table reports RDD estimates of the impact of a T2DM diagnosis on depression. Although not shown in the Table, estimates are conditional on a set of covariates, time, areas and GP fixed effects. Robust standard errors are clustered on the running variable. * p < 0.05, ** p < 0.01, *** p < 0.001