Intended for healthcare professionals

Primary Care

Length of patient's monologue, rate of completion, and relation to other components of the clinical encounter: observational intervention study in primary care

BMJ 2004; 328 doi: (Published 26 February 2004) Cite this as: BMJ 2004;328:501

This article has a correction. Please see:

  1. Israel Rabinowitz, family physician1,
  2. Rachel Luzzati, family physician1,
  3. Ada Tamir, statistician2,
  4. Shmuel Reis, family physician (reis{at}
  1. 1Clalit Health Services and Department of Family Medicine, B Rappaport Faculty of Medicine, Technion-Israel Institute of Technology, Haifa, Israel
  2. 2Department of Community Health and Epidemiology, B Rappaport Faculty of Medicine
  1. Correspondence to: S Reis, Departments of Medical Education and Family Medicine, B Rappaport Faculty of Medicine, Technion-Israel Institute of Technology, POB 9649, Bat-Galim, 31096 Haifa, Israel
  • Accepted 13 October 2003


The patient's opening statement in a consultation (the patient's monologue) is an important part of history taking, and doctors are encouraged not to interrupt the patient—but they often do,1 2 probably because they think that the patient's monologue is time consuming. When uninterrupted, patients conclude their monologue in less than 30 seconds in primary care and about 90 seconds in consultant settings.15

We assessed encounters in primary care that included a new clinical problem, recording the length and rate of completion of patients' monologues before and after instructing doctors not to interrupt.

Methods and results

We recorded consecutive encounters between eight family physicians and their patients on two days in six family clinics in northern Israel. All doctors were videotaped on both days. They had been told that the study focused on the doctor-patient interaction. Patients were given this explanation via a written notice on the door of the consulting room and also orally by the doctor when required. At the start of the second day the doctors were handed a written note that said: “When the patient starts speaking, please do not interrupt him or her until you are satisfied that he or she has finished.”

All practices had stable lists, and patients were seen by their regular doctors. The eight doctors were a convenience sample (five men; mean age 39.7 (range 35 to 44) years); all had completed the residency programme in family medicine. The sex and age of patients seen on days 1 and 2 was similar.

In total, 235 consultations (omitting two refusals) were recorded; 21 were excluded due to foreign languages, office procedures, and technical difficulties. Of 214 (91%) encounters we viewed, 112 (52%) involved a new clinical problem. We examined these for length of patient's monologue, whether the monologue was completed, performance and length of physical examination, ordering of accessory tests (or referrals to specialists), prescriptions, and total encounter time. Statistical analysis used χ2 and t tests, with significance level of 0.05. As patients are nested within physician, we used linear and logistic regression as well.

Monologues averaged 26 seconds on day 1 and 28 seconds on day 2 (table). After the intervention, twice as many monologues were completed, and six doctors accounted for this increase (90/112 (80%) encounters). A physical examination was performed in 88% of encounters; it averaged a minute and a half. Tests or referrals were requested in a third, a diagnosis was given in almost all, and prescriptions were issued in half the encounters. These figures did not change significantly after the intervention, nor did the length of the consultation.

Characteristics of consultations before and after doctors were instructed not to interrupt the patient's opening statement

View this table:


Allowing patients to complete their monologue requires little time and does not disrupt the other components of the clinical encounter. In consultations with a new clinical problem (that is, those aiming to reach a diagnosis), the number of completed monologues doubled when doctors were told not to interrupt.

The difference in monologue length between day 1 and day 2 is better represented by the median (15 and 21 seconds respectively) than by the mean (26 and 28), because the mean is affected by a number of relatively lengthy monologues. A similar difference was reported by Marvel.2

Different languages and cultures seem to have no effect on average length of monologue (Slovenia, 28 seconds3; United States, 23 seconds;2 Israel 27 seconds). Lengthier monologues have been reported in specialist settings (Switzerland, 90 seconds5).

The significant increase in the proportion of completed monologues is compatible with the observation that completed monologues are just marginally longer than interrupted ones.2 This is probably due to the natural brevity of patients' monologues.


Contributors: RL and IR wrote the protocol, collected and analysed data. AT gave statistical advice and supervised the analysis. IR wrote the first draft of the paper. All authors contributed revisions of drafts of the paper. SR supervised the whole process, wrote the final draft, and will act as guarantor


  • Funding No external funding.

  • Competing interests None declared.

  • Ethical approval: Helsinki Committee (IRB) of the Emek Medical Center, Afoula, Israel


View Abstract