Table 3.

The Roter Interactional Analysis System (RIAS): Frequency of Four Code Groupings According to Physician Specialty*

Communication BehaviorUnadjustedControlled for Encounter Duration OnlyExpanded Model
I. Data gathering
 a. Closed-ended biomedical questions23.116.5.0221.418.6.0522.218.4.01
 b. Open-ended biomedical questions2.
 c. Closed-ended psychosocial questions8.
 d. Open-ended psychosocial questions1.61.5.721.51.6.542.32.4.79
 e. Bids for clarification1.
 Total count37.428.2.0236.030.1.0235.829.6.02
II. Patient education and counseling
 a. Provides biomedical information32.037.6.1932.837.3.0431.734.9.22
 b. Provides psychosocial information3.
 c. Counsels biomedical14.213.8.7814.113.9.8714.713.3.23
 d. Counsels psychosocial7.28.0.567.48.3.395.25.5.81
 Total count57.567.4.1859.766.4.0857.361.8.25
III. Rapport building
 a. Personal remark7.88.6.698.08.3.809.910.6.31
 b. Laughter2.83.5.302.93.4.303.75.1.03
 c. Approval3.
 d. Empathy0.
 e. Concern2.12.3.492.12.4.302.93.1.35
 f. Reassure3.43.9.383.
 g. Legitimate0.60.7.460.70.8.300.91.0.54
 Total count20.924.3.3418.621.0.2627.630.9.14
IV. Partnership building
 a. Paraphrase4.
 b. Partnering0.
 c. Asks for opinion0.
 d. Asks for understanding3.33.0.583.13.1.963.33.0.31
 e. Asks for reassurance0.00.0.620.00.0.690.00.1.90
 Total count8.
  • IM, internal medicine; FP, family practice.

  • * Frequency analysis conducted using two-tailed significance testing and generalized estimation equations to control for variable number of office visits per physician.

  • Controlled for duration of encounter, experimental group, generalized health questionnaire (GHQ) score, being an established patient, physician age, patient income, patient ethnicity, and patient age.

  • P < .05.