Physicians’ and Patients’ Interruptions in Clinical Practice: A Quantitative Analysis

Article Figures & Data

Download figure
Open in new tab
Download powerpoint
Figure 1.
Predicted probability for intrusive interruption according to speaker role and speaker gender.
Download figure
Open in new tab
Download powerpoint
Figure 2.
Predicted probability for intrusive interruption according to consultation phase and speaker role.

Table 1.

Characteristics of Physicians and Patients

Characteristic	Physicians			Patients
Characteristic	Total (N = 17)	Women (n = 10)	Men (n = 7)	Total (N = 84)	Women (n = 42)	Men (n = 42)
Age, mean (SD), y	44.3 (11.3)	43.8 (10.7)	44.8 (11.9)	55.1 (14.9)	53.9 (16.0)	56.2 (13.9)
Time in practice, mean (SD) [range], y	15.0 (11.5) [2-43]	…	…	…	…	…
Medically unexplained symptoms, No. (%)	…	…	…	14 (16.7)	8 (19.0)	6 (14.3)
First visit, No. (%)	…	…	…	38 (45.2)	17 (40.5)	21 (50.0)

Table 2.

Characteristics of Consultations, Overall and by Dyad Composition

Characteristic	Total Consultations (N = 84)	By Physician-Patient Gender Dyad
Characteristic	Total Consultations (N = 84)	Woman-Woman (n = 21)	Woman-Man (n = 21)	Man-Woman (n = 21)	Man-Man (n = 21)
Patient age, mean (SD), y	55.1 (14.9)	53.3 (14.8)	56.5 (13.3)	54.5 (17.4)	56.0 (14.8)
Consultation length, mean (SD), sec	870.1 (371.2)	856.7 (370.4)	718.8 (263.0)	966.4 (335.2)	938.4 (462.7)
Phase length,^{^a} mean (SD), sec	682.5 (335.1)	599.8 (288.7)	600.9 (240.9)	801.5 (357.0)	728.0 (407.6)

↵a Combined length of problem presentation phase and diagnosis and/or treatment plan phase.

Table 3.

Predictors of Intrusive vs Cooperative Interruption Type

Predictor	β (SE)	OR (95% CI)
Random intercept: patient identification	0.34 (0.58)	…
Fixed effects^{^a}
Speaker role: patient vs physician	1.15 (0.28)	3.17 (1.83-5.50)^{^b}
Speaker gender: man vs woman	0.51 (0.24)	1.67 (1.05-2.65)^{^c}
Consultation phase: diagnosis and/or treatment plan vs problem presentation	0.81 (0.22)	2.24 (1.46-3.43)^{^b}
Gender dyad: same vs mixed	−0.25 (0.19)	0.78 (0.53-1.13)
Speaker role by speaker gender^{^d}	−0.90 (0.33)	0.41 (0.22-0.77)^{^e}
Speaker role by consultation phase^{^d}	−0.83 (0.28)	0.44 (0.25-0.75)^{^e}
Covariate: patient age, per year	0.19 (0.07)	1.21 (1.06-1.38)^{^e}

OR = odds ratio.
Note: Results of the generalized logistic mixed-effects regression model.
↵a Reference category is mentioned last (eg, physician).
↵b P <.001.
↵c P <.05.
↵d For interpretation of the interaction terms, see the results of the stratified analyses in the text.
↵e P <.01.

Table 4.

Interruption Type by Consultation Phase, Speaker Role, and Speaker Gender (N = 2,298 Interruptions)

Consultation Phase	Speaker Role	Speaker Gender	Cooperative	Intrusive
Consultation Phase	Speaker Role	Speaker Gender	Interruption Type, No. (%)^{^a}
Problem presentation	Physician	Woman	227 (95.4)	11 (4.6)
	Physician	Man	201 (89.3)	24 (10.7)
	Patient	Woman	204 (83.3)	41 (16.7)
	Patient	Man	150 (88.2)	20 (11.8)
Diagnosis and/or treatment plan	Physician	Woman	198 (86.1)	32 (13.9)
	Physician	Man	272 (82.4)	58 (17.6)
	Patient	Woman	409 (84.5)	75 (15.5)
	Patient	Man	333 (88.6)	43 (11.4)