Skip to main content
. 2023 Aug 20;50(6):926–935. doi: 10.1007/s10488-023-01290-x

Table 2.

Coefficients (standard errors) estimating the rate of prescriptions of antidepressants per quarter among patients in the Inland Empire Health Program (n = 18,833) as a function of provider characteristics, time, whether the provider ever received the “Train the Trainer” program, and the interaction of time and provider receipt of the TNT program

Antidepressants (1) (2) (3)
Time

0.003

(0.033)

0.030

(0.033)

0.030

(0.019)

TNT training

− 0.265***

(0.004)

− 0.087***

(0.004)

− 0.087

(0.056)

Time*TNT training

0.154***

(0.033)

0.158***

(0.033)

0.158*

(0.092)

Provider Characteristic
 Male gender (ref: female)

0.010

(0.004)

0.010

(0.022)

Specialty
 Family Practice (Ref.)
 General Practice

− 0.143***

(0.076)

− 0.143**

(0.072)

 Internal Medicine

− 0.010**

(0.005)

− 0.010

(0.031)

 Nurse

0.144***

(0.007)

0.144**

(0.066)

 Clinician Assistant

0.331***

(0.012)

0.331***

(0.030)

Age (yrs)
 30–39 (ref.)
 40–49

− 0.013***

(0.004)

− 0.013

(0.022)

 50–59

0.109***

(0.006)

0.109***

(0.034)

 60–69

0.050***

(0.006)

0.050*

(0.028)

 70–93

0.053***

(0.011)

0.053

(0.076)

 Quarterly trend

0.000

(0.000)

0.000

(0.002)

 Seasonality
 Jan.–Mar.(ref.)
 Apr.–Jun.

0.011***

(0.004)

0.011***

(0.003)

 Jul.–Sept.

0.007*

(0.004)

0.007*

(0.004)

 Oct.–Dec.

− 0.002

(0.004)

− 0.002

(0.004)

 Constant

0.216***

(0.015)

0.212***

(0.005)

0.212***

(0.023)

 R2 0.004 0.012 0.012

*p < 0.1; **p < 0.05; ***p < 0.01

Model 1 (left column) represents the base model;

Model 2 controls for provider characteristics, seasonality, and time trends.

Both Models 1 and 2 treat all patient observations as statistically independent.

Model 3 was clustered by providers using the “cluster” standard error estimation routine in STATA.