Table 1.

Association of Clinician Response With Subsequent Length of Visit

Type of Clinician ResponseSubsequent Length of Visit, β (95% CI), Minutes
Unadjusted ModelaModel 1bModel 2c
Broad response
Explicit vs nonexplicit−1.20 (−2.30 to −0.10)−0.98 (−2.19 to 0.23)−1.37 (−2.89 to 0.15)
Provides vs reduces space2.95 (1.55 to 4.35)1.75 (0.20 to 3.29)1.27 (−0.63 to 3.18)
Specific response
Provides space
   Empathy1.53 (−0.99 to 4.05)1.05 (−1.77 to 3.86)1.38 (−2.26 to 5.01)
   Explicit focus on affect−0.47 (−3.35 to 2.41)−3.06 (−6.22 to 0.10)−4.11 (−8.00 to −0.21)
   Acknowledgment−2.02 (−3.38 to −0.65)−1.17 (−2.69 to 0.34)−1.14 (−3.06 to 0.78)
   Exploring1.37 (0.06 to 2.68)1.03 (−0.42 to 2.48)0.77 (−1.00 to 2.55)
   Neutral/passive1.56 (0.43 to 2.68)0.76 (−0.47 to 2.00)0.68 (−0.90 to 2.26)
Reduces space
   Gives information/advice−3.51 (−5.20 to −1.82)−1.89 (−3.74 to −0.04)−1.64 (−3.96 to 0.68)
   Blocking−1.75 (−3.98 to 0.48)−0.62 (−3.06 to 1.82)−0.03 (−2.85 to 2.79)
  • Note: Values in bold are statistically significant (P <.05).

  • aRandom intercept univariate multilevel linear regression models (account for clustering of emotional expressions within patient encounters and of patient encounters within clinicians).

  • bRandom intercept multilevel linear regression models (account for clustering of emotional expressions within clinicians) with adjustment for the timestamp of the emotional expression.

  • cRandom intercept multilevel linear regression models (account for clustering of emotional expressions within clinicians) with adjustment for the timestamp of the emotional expression; clinician age, sex, and race/ethnicity; and patient age, sex, and race/ethnicity.