Terrence Steyer, M.D.
Chair, Department of Family Medicine
L.A. Anderson and R.F. Dedrick (1990)
The TPS was developed to measure interpersonal trust in his physician. An initial 25 items were created from patient interviews and other instruments. Testing of these items and elimination of redundant or inappropriate items resulted in an 11-item, self-administered questionnaire measuring three dimensions of trust: physician dependability, confidence in physician knowledge and skills, and confidentiality and reliability of information received from the physician. Items are answered in a 5-point Likert format.
Internal reliability is excellent (Cronbach alpha = 0.90), although no factor analysis could be conducted to identify the three dimensions of trust. Validity of the measure, by comparison with other trust scales, showed moderate correlation with these other measures. In further measures of reliability and validity, Thom et al. (1999) tested a slightly modified version of the TPS where one question and the response labels had been slightly re-worded. They report that the TPS has high internal consistency (Cronbach a=0.89) and good one-month test-retest correlation (?=0.77). TPS score correlated well several measures of the patient’s preferences regarding the physician’s role. In six-month followup surveys, TPS score was significantly correlated with continuity of care, adherence to prescribed medication, and overall satisfaction with care.
This instrument has not been specifically tested with elderly populations. Thom et al. (1999) report that older patients (>55) had significantly more trust in their physicians than younger patients.
Reliability and validity measurements specifically in minority populations have not been conducted.
In a study to assess the relative strength of the association between physician behaviors and patient trust, patients (N=414) enrolled from 20 community-based family practices rated 18 physician behaviors and completed the Trust in Physician Scale immediately after their visits (Thom et al. 2001). Trust was also measured at 1 and 6 months after the visit. The association between physician behaviors and trust was examined in regard to patient sex, age, and length of relationship with the physician. All behaviors were significantly associated with trust (P<.0001), with Pearson correlation coefficients (r) ranging from 0.46 to 0.64. Being comforting and caring, demonstrating competency, encouraging and answering questions, and explaining were associated with trust among all groups.
Meit et al. (1997) used the TPS to measure whether dressing patients in hospital gowns had an effect on patients' trust in their physician. No significant effects for gowning status were found with respect to patients' trust in their physician or duration of clinic visit.
The TPS shows excellent internal reliability and moderate to good external validity. The instrument is a rapid and stable assessment of patient trust in their physicians. The instrument has been shown effective in addressing issues regarding racial and ethnic differences in patient-physician trust.
The eleven questions of the TPS are published in Anderson and Dedrick (1990). The slightly modified questions are published in Thom et al. (1999).
Anderson, L.A. and R.F. Dedrick (1990) Development of the trust in physician scale: a measure to assess interpersonal trust in patient-physician relationships. Psychological Reports 67, 1091-1100.
Meit S.S., D. Williams, F.C. Mencken, V. Yasek (1997) Gowning: effects on patient satisfaction. J. Fam. Pract. 45, 397-401.
Thom, D.H., K.M. Ribisl, A.L. Stewart, D.A. Luke, and The Stanford Trust Study Physicians (1999) Further validation and reliability testing of the trust in physician scale. Medical Care 37(5) 510-517.
Thom, D.H., Stanford Trust Study Physicians (2001) Physician behaviors that predict patient trust. J. Fam. Pract.,50(4):329-30.