Volume 191 - Issue 2
An early general practice trial of antidepressants: interview with the trialist, Tim Blashki
Authors: Bruce Arroll, Timothy G Blashki and Grant A Blashki
Med J Aust 2009; 191 (2): 123-125. || doi: 10.5694/j.1326-5377.2009.tb02710.x
Published online: 20 July 2009
Published online: 20 July 2009
Lessons from the past for today’s researchers in general practice
In January 1971, a randomised controlled trial of management of depression by general practitioners was published in the British Medical Journal by Tim Blashki (T G B) and his colleagues Robert Mowbray and Brian Davies (Box 1).1 Although there had been two earlier trials of antidepressants conducted in general practice (one British and one American), this was the first in the world to have extractable data in general-practice-only patients that could be used in a meta-analysis. It is therefore the earliest study included in this year’s published Cochrane Reviews that examines antidepressants versus placebo for depression in primary care; the review’s authors were Bruce Arroll (B A), Grant Blashki (G A B) and colleagues.2
Tim Blashki, the first author of this historically important article, was a Melbourne psychiatrist who had been a GP before he commenced the study. Tim Blashki’s son, Grant (G A B), has continued the family tradition of researching the management of mental illness by GPs,3 and he recently took up Bruce Arroll’s suggestion to interview Tim to document some of his experiences from this early clinical trial. This interview took place by phone on 8 January 2009.
The interview
Could you say something about your medical training?
I graduated from medical school in Sydney in 1964, having had only about 12 lectures in psychiatry in total. My first real contact with psychiatry was as a second-year resident at Royal Melbourne Hospital, where I worked as a medical officer under Professor Brian Davies. Psychiatry interested me and came easily to me, and this sparked an interest which has continued for the rest of my life.
As a result of this experience, I went to a typical psychiatric hospital of the time. There were some very good and dedicated people working there, but, for the most part, treatments at that time were only partially effective, and there was a large custodial component in the function of the institution. My first job was unsupervised, and involved looking after a ward for young women (adolescents and those in their early 20s), many of whom had psychosis with poorly controlled symptoms. I left after 6 months and began working in general practice, as I wanted to help people experiencing mental illnesses in the community.
The general practice I joined practised in a traditional manner for that time. The focus was on patients’ somatic symptoms and on somatic treatments. Patients had great faith in the doctors, whom they had known for many years in what was a rather tight-knit community. Psychological problems were generally ignored or ascribed to some somatic problem or social difficulty. Treatment consisted of support, advice, and a variety of what were essentially placebo treatments, ranging from rose-coloured water (dill water), vitamins, Waterbury’s compound, the pharmacist’s special concoction (often a bromide-containing medication), and night sedation with drugs such as chloral hydrate and barbiturates. The use of tricyclic antidepressants, monoamine oxidase inhibitors and minor tranquilisers, often in minute doses, was just beginning. Depression was just starting to be acknowledged as an illness; most depressions were perceived as reactions to events or a failure of will. Anxiety more or less went with the depression, with the exception of phobias and “panic” states.
It became clear to me that much of what I was seeing in general practice — I thought about 70% — was psychologically based, and so I returned to psychiatry at Royal Melbourne Hospital, where I again worked under Professor Brian Davies. It was in this dual setting of psychiatry and of general practice that I began to think about psychological disorder in the community, and depression and anxiety in particular.
How did the idea of a randomised, placebo controlled trial of antidepressants in general practice come about?
In the late 1960s, I decided to do a doctorate in medicine. I chose to do this with a focus on general practice for a number of reasons.
First, as I’ve already mentioned, many of the problems that I’d seen in general practice were of a psychological nature.
Second, while most mental health problems were being seen in general practice, most of the research was done in hospitals on inpatients and the results were being extrapolated to general practice patients. The assumption was that these patients were part of the same cohort, but, having worked in both places, I believed this to be incorrect.
Third, I was interested in the notion of placebo as an effective treatment, and I had wondered whether some of the “antidepressant effect” seen in patients treated in general practice for depression was possibly the result of such a placebo effect.
Fourth, I thought that some of the response to antidepressants in this mildly to moderately affected group might be occurring because of reduced anxiety, rather than to a specific antidepressant effect.
Finally, I could find very little published research on what was essentially a large group of people in the community who had some type of mental disorder.
What was involved in getting the study off the ground?
Initially, I had two problems. The first was that GPs were said to be not particularly interested in psychiatry.4 However, I felt that given half a chance, many GPs would respond to an offer of some help and of an opportunity to be involved in a research project. This indeed proved to be the case. It involved me visiting GPs in their practices and talking about the sorts of issues that they faced, which facilitated the study and was central to its successful completion over 9 months in late 1969 and early 1970. The Research Committee of the Victorian branch of the Royal Australian College of General Practitioners were well acquainted with the project and approved it.
The second problem was that of constructing a study that had the same scientific rigour as previous studies that had been conducted in institutions, and to apply this rigour in a general practice setting. To accomplish this, it was necessary to ensure randomisation, placebo control, double blinding, and for the patients to be rated both subjectively and objectively with the use of rating scales such as the Hamilton Rating Scale for Depression5 and the Taylor Manifest Anxiety Scale.6 To confirm whether the patients did actually take the tablets that were prescribed, riboflavine was included in the formulation, and patients’ urine was tested for compliance with treatment. The coauthors of the study, Robert Mowbray and Brian Davies, made important contributions to the planning of the study.
What did the study find?
For me, the most fascinating and important finding was that clinical improvements occurred in 55% of patients who had received placebo after 7 days, and in 61% at 28 days. Indeed, there was a handful of patients who wanted to continue taking the placebo even after they were told it was a placebo, and we managed to obtain more placebo from the pharmaceutical company, so that they might continue taking their “medication”.
With respect to the specific treatments, I was not surprised that the higher dose of amitriptyline (150 mg per day) was the most effective in relieving depression and anxiety. Anecdotally, I thought that the smaller dose of 75 mg might also be effective in relieving symptoms, but this proved not to be the case. I was interested to read a recent study by Furukawa et al, published in the British Medical Journal in 2002, in which they found a good response to low-dose tricyclic antidepressants.7
With respect to side effects, there was no difference between the groups. This was very surprising, but I think it might be attributed to a problem in the study design. Participants were given a list of possible side effects before starting the medication — something I thought to be important because some might well have been troubled with the higher doses — and this may have created a bias that was reflected in the side-effect profile across all groups.
What was the social context in which the study was conducted?
I have already alluded to the apparent lack of interest in psychiatry among GPs in those days, to the stigma associated with psychiatric illness, to the notion that depression was induced by life circumstances or was a failure of will; certainly, “madness” was something to be avoided at all costs. While there was some acknowledgement of a biological tendency or disposition often reflected in a family history, when this was discerned, patients were generally sent off to the psychiatrist and the case considered exceptional.
The other social factor was that while women came to their doctors for treatment, men generally did not. Men saw it as a weakness, unmanly, and indicative of failure; their way of expressing their distress was often through aggression, excessive alcohol consumption or, at worst, carefully planned or violent suicide. So, I saw mainly depressed women who seemed more ready to say something about how they felt and express something of their vulnerability. Hence, the idea of studying only women in this project came to mind.
What is your view of mental health management in general practice today?
The management of mental health in general practice is far removed from that of 40 years ago. GPs are more educated in mental illness, and mental illness is less stigmatised. Research has provided a greater understanding of mental disorders and the treatments available, whether they be biological, social or psychological, and much therapy is now evidence-based. It is important that mental health research continue in the setting of general practice, and there are at least three important lessons for future researchers that I’d like to share (Box 2).
Fortunately, mental health is now seen as a community responsibility, and there is commensurate financial commitment. Powerful support structures now exist — community, psychiatric, psychological and social. GPs have rightly become an integral part of this process.
1 About one of the world’s first randomised controlled trials of management of depression by general practitioners1
The study was a double-blind randomised controlled trial of amitriptyline (two different doses: 75 mg and 150 mg per day), amylobarbitone (150 mg/day), and placebo, for 4 weeks.
It was conducted between 1969 and 1970, and involved 82 women with depressive illness, recruited from 21 general practices in Melbourne, who were randomly allocated into the four groups (61 women completed the study).
Improvement at 7 and 28 days was noted on several measures of depression and anxiety in all treatment groups.
Of the treatments, amitriptyline at 150 mg/day was the most consistent in relieving depression and anxiety.
Troublesome side effects were equally distributed among the four groups.
2 Lessons for researchers undertaking mental health studies in the general practice setting
The success or failure of research in general practice will usually depend on the trust established between GPs and their patients.
All modes of treatment should be assessed in the setting in which they are to be used, and this is especially true in general practice.
Research needs to be scientifically rigorous and objective, while remaining sensitive to the subjective experience of patients.
Competing interests
The original trial1 was funded by Roche Products. Grant Blashki has received many free dinners from the trialist Tim Blashki.
References
- Blashki TG, Mowbray R, Davies B. Controlled trials of amitriptyline in general practice. BMJ 1971; 1: 133-138. 0_i1091901
- Arroll B, Elley CR, Fishman T, et al. Antidepressants versus placebo for depression in primary care. Cochrane Database Syst Rev 2009. In press. 0_i1091903
- Blashki G, Judd F, Piterman L. General practice psychiatry. Sydney: McGraw-Hill Medical, 2006. 0_i1091905
- Shepherd M, Cooper B, Brown AC, Kalton GW. Psychiatric illness in general practice. London: Oxford University Press, 1966. 0_i1091907
- Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry 1960; 23: 56-62. 0_i1091909
- Taylor JA. A personality scale of manifest anxiety. J Abnorm Psychol 1953; 48: 285-290. 0_i1091912
- Furukawa TA, McGuire H, Barbui C. Meta-analysis of effects and side effects of low dosage tricyclic antidepressants: systematic review. BMJ 2002; 325: 991-1000. 0_i1091914