The relationship between the patient and the psychiatrist is at the centre of treatment and predictive of adherence and outcomes. Some evidence suggests that the relationship can be quickly formed and that patients' initial appraisal of psychiatrists may already be important.Reference Martin, Garske and Davis1-Reference Gale, Hannah, Swain, Gray, Coverdale and Oud3 The introduction provides patients with a first impression of the psychiatrist. Ideally, the introduction should instil trust and facilitate a positive attitude by the patients towards the psychiatrist. How should this be achieved and how should psychiatrists introduce themselves? There is no consistent evidence on what type of introduction patients prefer. Although it is a matter of proper professional conduct to introduce oneself by stating ones name and the reason for the consultation, it is unclear whether patients at that stage would like to receive more information about what is likely to happen in the consultation or whether such information might be perceived as rather overloading and confusing.Reference Priebe, Dimic, Wildgrube, Jankovic, Cushing and McCabe4-Reference Johansson and Eklund9 In addition to the question as to how much information should be provided, there is the uncertainty about whether psychiatrists should disclose personal information. Some evidence suggests that patients with mental illness appreciate when psychiatrists disclose personal information. They may value the ‘human touch’ of a psychiatrist who shares a personal difficulty with them.Reference Priebe, Dimic, Wildgrube, Jankovic, Cushing and McCabe4,Reference Beach, Roter, Larson, Levinson, Ford and Frankel10-Reference Hanson12 However, this has not been studied in the context of the introduction when the patient has been referred for a consultation by other services and meets the psychiatrist for the first time.
Studies on patients' views of clinical communication and therapeutic relationships have so far employed naturalistic designs. Such designs cannot control for various confounding factors in the complex situation of real treatment, capture only the naturalistically occurring variance and do not allow conclusions on causal relationships. Experimental designs are required to assess patients' preferences of different styles of introductions. Against this background we conducted an experimental study to assess how patients appraise different types of introductions when they meet the psychiatrist for the first time having been referred to a secondary psychiatric service by their primary care physician. In particular, we tested whether patients prefer brief introductions in which psychiatrists just state who they are and what the reason for the consultation is, longer ones in which psychiatrists additionally provide information on what they are going to do in the consultation and what the outcome might be, or even longer introductions in which psychiatrists disclose a personal difficulty. Moreover, we explored whether these preferences vary depending on patients' characteristics.
Method
The study was a cross-sectional, exploratory, non-clinical experiment with a balanced incomplete block design. Video clips with consultant psychiatrists introducing themselves in different styles were shown to in- and out-patients with mental disorders, and patients' appraisal of the psychiatrist, based on viewing the video clips, was assessed.
In total, 12 psychiatrists working in East London were filmed. Consultant psychiatrists were purposively selected using gender, age and ethnic background as selection criteria and aiming at a spread of each of those criteria. All the psychiatrists worked outside services for adult general psychiatry in the London Borough of Newham where the patients were recruited to avoid any possible confusion of patients and any potential impact on a later actual treatment of the patient in Newham services.
Each psychiatrist was filmed with three different types of introduction. All introductions were designed for the first consultation, when a patient has been referred by a primary care physician to the psychiatrist, but they varied in the level of information provided and personal disclosure.
(a) Brief introduction (A). The psychiatrist stated only their real name (to maximise genuineness and credibility) and added that they are a consultant psychiatrist and that the consultation happened because the patient had been referred to them by the general practitioner.
(b) Introduction with more information (B). The same as in Introduction A plus telling patients that they will now do an assessment of about 30 min length following which they will explain what pharmacological and/or psychotherapeutic treatment they think would be appropriate and available.
(c) Introduction with information and disclosure of personal information (C). The same as in Introduction B plus an apology for being late for the appointment because of an event in their personal life. Possible explanations for being late were (i) that their daughter had fallen sick and they had to arrange for the child to be picked up from school, (ii) that their mother was very ill and needed some urgent advice on the telephone, (iii) that their partner had lost the keys and could not get into the house so that they had to arrange for a cab to take the house keys to the partner, and (iv) that they had to go and see the general practitioner adding that doctors sometimes have health problems like everyone else.
Psychiatrists were entitled to amend the specific wording if they felt that a different terminology would be more appropriate for their personal style (for example using ‘medication’ instead of ‘pharmacological treatment’) as long as content and length of the introduction remained unaltered.
In total, 36 clips were video recorded, one of each type by each psychiatrist. To detect a one-point difference in rating scores between two different introduction types with 80% power at the 5% significance level, assuming the standard deviation of the difference in scores is two (equivalent to a medium effect size of >0.5), would require 63 patients. Each patient was shown six video clips, two of each introduction A, B and C. Assuming an intracluster (i.e. between psychiatrist) correlation coefficient of 0.03 and a cluster size of 636/12431 gives a design effect of 1.9, thus a target sample size of 1.9634120 patients.
Video clips were randomly selected. The randomisation procedure was stratified to ensure that each patient watched six clips with six different psychiatrists and that the explanations of psychiatrists for being late in Introduction C varied for each patient (watching the same explanation twice might have compromised the credibility of the presentation).
Inclusion criteria for patients were: current in- or out-patient in secondary mental health services in East London; between 18 and 65 years of age; no organic mental disorder; sufficient command of the English language to understand the video clips and fill in rating scales in English; and capable of giving informed consent. Patients with too high a current symptom level to participate were excluded.
After watching each clip patients were asked three questions:
(a) Do you believe this is a good doctor?
(b) Would you have trust in this doctor?
(c) Would you like this doctor to be your psychiatrist?
Patients rated the answers to each question on a four-point Likert-type scale (with the categories: definitely no; probably no; probably yes; definitely yes). We also obtained information on patients' age, gender, school-leaving age, employment, living situation, length of illness and clinical diagnosis according to ICD-10.13
Procedure
Patients were recruited from a range of in- and out-patient settings in general adult psychiatry in the London Borough of Newham and interviewed between August 2011 and September 2012. Patients were screened for eligibility by their psychiatrists and asked whether they would consent to be approached by a researcher. All research interviews were conducted by trained psychiatrists, and patients were reimbursed for their time with £10. All patients provided written informed consent. The study was approved by the National Research Ethics Service, Committee London - East (ref. 09/H0701/11).
Analysis
Descriptive statistics were used to present characteristics of the sample. The three ratings of patients for each psychiatrist were summed and treated as a continuous outcome (range 0-9). The internal consistency of the three items was calculated using Cronbach's alpha.
Linear regression models were fitted with a random effect for psychiatrist (since there were between 47 and 67 ratings per psychiatrist). Aside from the type of introduction, we included either diagnostic group (F2 v. others) or a sociodemographic characteristic (gender; age group, ⩽40 v. >40 years; ethnic category, White, Asian, Black; educational level, i.e. school-leaving age of ⩽16 v. >16 years; living situation, with partner or family v. alone; employment, any v. none) and tested the interaction between introduction type and the characteristic, the significance of which was assessed using a Wald test. We used these models to predict adjusted mean scores and 95% confidence intervals. We also carried out subgroup analyses within the characteristic categories defined above, but did not test the differences within subgroups for statistical significance as the study had not been powered for this. All statistical analyses were carried out using Stata version 10.1 on Windows.
Results
Characteristics of psychiatrists and patients
Twelve psychiatrists (eight women, four men) were videotaped. Three of them were between 30 and 40 years of age, five between 40 and 50, and four between 50 and 60. Seven were of White origin, three of Asian and two of Black African ethnic origin. Each psychiatrist's Introduction A was rated by 11-25 patients, Introduction B by 15-25 patients and Introduction C by 13-23 patients.
In total, 120 patients participated in the study, resulting in a total of 720 ratings. Of these patients, 33 (27.5%) were recruited from in-patient wards and 87 (72.5%) from different out-patient clinics. Out of the patients who were deemed eligible by the clinicians and subsequently approached by the researchers, all patients consented to participate in the study. Sixty-six patients (55%) were male. The mean age was 40.9 years (s.d. = 12.1) and the mean length of illness 10.6 years (s.d. = 9.9). With respect to ethnicity, 52 (43.3%) were White, 31 (25.8%) Asian, 22 (18.3%) Black African, 11 (9.2%) African-Caribbean and 4 (3.3%) of mixed ethnic background. The school-leaving age was 16 years or lower for 84 patients (70.0%) and more than 16 years for 36 (30.0%). Ninety-seven patients (80.8%) were unemployed; 3 patients (2.5%) were in voluntary and 20 (16.7%) in regular employment. A total of 46 patients (38.3%) lived alone and 74 patients (61.7%) lived with a partner or family. At the time of the interview, 51 patients were out-patients, 42 patients in day care settings, 20 on conventional psychiatric wards, and 7 on a psychiatric intensive care unit.
The main clinical diagnoses according to ICD-10 were schizophrenia (35 patients, 29.2%), schizoaffective disorder (13, 10.8%), bipolar affective disorder (22, 18.3%), depression (37, 30.8%), anxiety disorder (9, 7.5%) and personality disorder (4, 3.4%).
Patients' ratings
Cronbach's alpha for the internal consistency of the three items was very high at 0.92. The mean sum score for the whole sample and all introductions was 5.2 (s.d. = 2.7). For Introduction A, the mean score was 4.95 (95% CI 4.5-5.4), for Introduction B, with more information, 5.73 (95% CI 5.3-6.2) and for Introduction C, with personal disclosure, 4.96 (95% CI 4.4-5.5). Type of introduction was strongly related to rating score (χ2 = 12.9, d.f. = 2, P = 0.002). The intraclass correlation coefficient of the rating score on psychiatrist was 0.03 (95% CI 0.00-0.06), implying that 3% of the variability in score was as a result of differences between psychiatrists.
Differences in subgroups
Table 1 shows the ratings for the subgroups with different gender, age, ethnicity, school-leaving age, living situation, employment and diagnostic category. The ratings were consistent across all subgroups with similar predicted mean scores and confidence intervals after adjusting for patient characteristics. Introduction B (with more information but no personal disclosure) received the most positive ratings in all subgroups. The other two introductions were consistently rated very similarly, and less favourably than Introduction B. The P-values comparing B with A, and B with C were all <0.05 after adjusting for each of the characteristics in Table 1, whereas the P-value comparing C with A was large for each characteristic. Some degree of overlap in confidence intervals for the introduction types does not preclude overall statistical significance. There was no significant interaction effect between the type of introduction and any of the patient characteristics.
Discussion
Main findings
This experimental study produced a clear result: patients prefer psychiatrists who introduce themselves in the first consultation not only by stating their name and explaining that the patient has been referred by a general practitioner, but also providing some information about what they intend to do, how long the consultation is likely to take and what a potential outcome might be. The preference for more information is not an effect of the length of the introduction in the sense that patients would simply prefer more elaborate and longer introductions. The introduction with personal disclosure was the longest one and did not receive higher ratings than the brief introduction. At the same time, patients are more critical when psychiatrists disclose a personal difficulty. This preference for an introduction with information about what is going to happen and without personal disclosure does not depend on patient characteristics. The findings were remarkably consistent across subgroups of patients with different ages, gender, ethnicity, school education, employment status and clinical diagnoses.
Ratings (predicted adjusted mean scores and confidence intervals) of the three different introductionsFootnote a in different subgroups of patients
| Mean (95% CI) | ||||
|---|---|---|---|---|
| Patient characteristics | n | Introduction A | Introduction B | Introduction C |
| Gender | ||||
| Male | 66 | 4.8 (4.4-5.2) | 5.6 (5.1-6.0) | 4.8 (4.4-5.3) |
| Female | 54 | 5.1 (4.7-5.6) | 5.9 (5.4-6.4) | 5.2 (4.7-5.6) |
| Age, years | ||||
| 18-40 | 62 | 4.8 (4.3-5.3) | 5.6 (5.1-6.0) | 4.8 (4.4-5.3) |
| 41-65 | 58 | 5.1 (4.7-5.6) | 5.9 (5.4-6.3) | 5.1 (4.7-5.6) |
| Ethnicity | ||||
| White | 52 | 5.0 (4.5-5.5) | 5.8 (5.3-6.3) | 5.0 (4.5-5.5) |
| Asian | 31 | 4.7 (4.1-5.2) | 5.4 (4.9-5.9) | 4.6 (4.1-5.2) |
| Black | 37 | 5.0 (4.5-5.5) | 5.8 (5.2-6.3) | 5.0 (4.5-5.5) |
| School-leaving age | ||||
| ⩽16 years | 84 | 5.0 (4.6-5.5) | 5.8 (5.4-6.2) | 5.1 (4.6-5.5) |
| >16 years | 36 | 4.7 (4.3-5.2) | 5.5 (5.0-6.0) | 4.8 (4.3-5.3) |
| Living situation | ||||
| With family/partner | 74 | 4.9 (4.5-5.4) | 5.7 (5.2-6.2) | 5.0 (4.5-5.4) |
| Alone | 46 | 5.0 (4.5-5.4) | 5.7 (5.3-6.2) | 5.0 (4.5-5.4) |
| Employment | ||||
| None | 97 | 5.0 (4.6-5.4) | 5.7 (5.3-6.1) | 5.0 (4.6-5.4) |
| Any | 23 | 4.9 (4.3-5.4) | 5.6 (5.1-6.2) | 4.9 (4.4-5.5) |
| Diagnosis | ||||
| Psychosis | 48 | 5.1 (4.6-5.5) | 5.8 (5.3-6.3) | 5.1 (4.6-5.5) |
| Other disorders | 72 | 4.9 (4.4-5.3) | 5.6 (5.2-6.1) | 4.9 (4.5-5.4) |
a. A: brief introduction with name and reason for consultation; B: same as A plus information about what is going to happen; C: same as B plus disclosure of personal information.
Strengths and limitations
All psychiatrists were rated with each of the introductions, so that the findings are independent of psychiatrist's sociodemographic characteristics such as gender, age and ethnicity and also independent of their personal styles, dress and accents. All psychiatrists were new to the patients and the findings were not influenced by potential experiences that the patients may have had with one or more of the psychiatrists. Also, all videotapes were with real psychiatrists who used their real names to have a maximum of genuineness and credibility. Finally, the analysis provided a consistent and unequivocal result.
The study also has some limitations. Most notably, the patients were an opportunistic sample of psychiatric patients and recruited from only one area in East London. Although it is unlikely that the findings would have been different in other patient groups in the same setting - given their consistency across all subgroups - it is unclear whether these findings can be replicated in other service types and countries or with patients who have never had been in contact with a psychiatrist before. We compared only three types of introductions and made them clearly distinct. It is possible that further variations might have yielded even more positive or different ratings. For example, one might speculate as to whether providing even more information on the expected procedure in the consultation or a different kind of personal disclosure - other than the somewhat intimate disclosure of a personal difficulty linked to an apology for being late as used in this study - would have led to more favourable ratings.
Implications
The study may be seen as an encouraging example of how experimental designs can be implemented to answer research questions of clinical relevance. In real treatments it is difficult to vary patient-clinician communication and test how patients may respond to different types of presentations and communication. The approach used in this study can obtain patient responses to a controlled variation of distinct communication styles.
The findings have direct implications for how psychiatrists should introduce themselves to their patients. Of course, the findings reflect average differences and may not apply to every single patient in every situation. Information about patients obtained before the first encounter might modify the general rule as suggested by the results in this study. Yet, in the absence of good reasons to behave differently, psychiatrists should not only state their name and explain the nature of the referral, but also outline what they intend to do in the first consultation, how long it is likely to take and what the outcome might be. At the beginning of the first consultation, patients may have various and often vague expectations about what might happen when they see a psychiatrist. Addressing this uncertainty and indicating the aims and timescale of the meeting appear to be valued by patients and raise their trust in the psychiatrist. At the same time, psychiatrists should refrain from disclosing personal information at that stage. Disclosure of personal difficulties may or may not be beneficial at a later stage of treatment, but at the beginning of the first consultation it does not help to establish a more positive relationship.
The introduction is only a minor part of a treatment, but it is the beginning of a relationship that in some cases may last a very long time. Setting off the relationship in a way that is evidence-based and that increases the chances to be seen as the right psychiatrist by the patient should be a good start.
Funding
The study was funded as own account research within East London NHS Foundation Trust.
Acknowledgements
We thank all 12 psychiatrists who volunteered to be videotaped. We are also grateful to Dr Jude Ezeonwuka and Dr Nike Ljete for their help with producing the videotapes and we also wish to thank the patients who participated in the study.
eLetters
No eLetters have been published for this article.