Table 4

Changes in Quality under peer scrutiny and encouragement

Dependent Variable
Whether a Clinician Provides a Particular ItemProtocol Adherence
4
123
SCR (peer scrutiny)0.031***
(0.005)
0.027***
(0.010)
0.035**
(0.013)
0.03**
(0.012)
POST SCR (post scrutiny)0.016**
(0.006)
0.015
(0.011)
0.02
(0.014)
0.022
(0.014)
ENC (encouragement)0.059***
(0.005)
0.055***
(0.009)
0.058***
(0.011)
0.08***
(0.010)
ENC · TRK (encouragement tracked)0.044***
(0.007)
0.041***
(0.006)
0.073***
(0.010)
Clinician effects (yes)
    Item effectsyesno
    Patient effectsG,AG,AG,AG,A,MS
    Patient order−0.001
(0.001)
0.0000
(0.001)
−0.0010
(0.001)
0.0000
(0.001)
N41,00741,00741,0071,855
  • Notes: Marginal effects reported. Significant at 1 percent (***), 5 percent (**), and 10 percent (*). (1) Logit regression with dummies for item effect, controlling for gender (G) and age (A) of patient; (2) Logit regression with random effects at the unique patient level, with dummies for item effect, controlling for gender (G) and age (A) of patient; (3) Fixed-effect linear regression with fixed effects for each unique item, controlling for gender (G) and age (A) of patient, clustered at the unique patient level; (4) Fixed-effect linear regression of patient average, controlling for gender (G), age (A), and major symptom (MS).