Randomised equivalence trial comparing three month and six month follow up of patients with hypertension by family practitioners
BMJ 2004; 328 doi: https://doi.org/10.1136/bmj.37967.374063.EE (Published 22 January 2004) Cite this as: BMJ 2004;328:204All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
Mac Donald´s ironical suggestion resumes the study´s central idea:
“follow up intervals are based on phases of the moon because nothing
better exists”. But, dangerously, conclusions glance an apparently,
simple road. EBM, magíster dixit, points out that every six months follow
up is the same as three months. But the question is what sort of
“evidence” is it ? This is unidirectional evidence; “evidence” from
doctors trying to find, to prove, their own ideas: we read in the
introduction: “doctors must decide how often to follow up ...”.
Evidence must be an intersubjective agreement, in a bidirectional
manner, a co-construction in the relationship. This means a reasonable, an
ethical, way to practise Medicine. Medical inquiry needs an
epistemological stance far away from objectivism and de-subjectivation.
Follow up must be tailored in the relationship. This is the better thing that
could exist. This study's interest is that itinvites us to think about the
ancient saying: “treating the patient rather than treating the number”.
This is what really adds, once more.
Competing interests:
None declared
Competing interests: No competing interests
It is important that properly conducted trials inform those who
produce guidelines on the management of clinical conditions. Birtwhistle
et al have contributed to this knowledge base but I am concerned about the
generalisability of their trial. By selecting those with "controlled"
blood pressure they have selected from a pool of 13% (their figures) of
hypertensive Canadians. Another problem is that hypertension is only one
of several risk factors for cardiovascular disease. Frequency of visits is
likely to be delineated by the presence of other cardiovascular risk
factors and comorbidity and the need for their management. Thus frequency
of review advice should be based on absolute cardiovascular risk rather
than the level of a single risk factor.
Competing interests:
None declared
Competing interests: No competing interests
I feel, the interval of six months is too long a period for follow up
for patients with hypertension.Frequent follow up of 4-6 weeks gives an
opportunity for checking drug compliance, reinforcing life style
modification and assessing the control of blood pressure.
Competing interests:
None declared
Competing interests: No competing interests
While my practice has been routine 3 month f/u for hypertension, this
study is reassuring that longer follow-up intervals are just as
satisfactory for control of the surrogate endpoint of blood-pressure. It
would be even better to see this study carried for a longer period of time
with measurement of patient-oriented outcomes like heart attack or stroke
or death.
The study also points out that, whatever the frequency of visits, we
need to work harder to get ALL patients under control.
Competing interests:
None declared
Competing interests: No competing interests
As an editor of a journal of record, the basis on which you have seen fit to print this paper is incomprehensible. There seems to be a worsening paucity of ideas worthy of follow-up in the profession at large in all the research journals. Now we have this one, of which this is the essence...
Eight persons, of which five were professorial, were engaged in supervising an endeavour to minimise care for essential hypertension if at all possible.
They chose 607 patients with treated and controlled blood pressure for at least three months. All the criteria for “controlled” were not given. Eighty out of 130 family doctors failed or refused to participate. The patients were divided into two groups. 302 were to be seen 3 monthly over three years. That is 302 x 4 x 3 = 3624 measurements to be taken. The figure for the 6 monthly customers evaluates to 307 x 2 x 3 = 1842 ie 5466 encounters overall among the 58 researchers, which is 95 readings on average each over the three years, or one a fortnight (5 day working week). The exclusion criteria incomprehensibly included those treated by specialists, the pregnant, and those who could not be randomised for other medical reasons. It is not known in detail how patients were identified as out of control and needing earlier reassessment or the interval of reassessment if change of drug was involved; or if other reasons did, as a matter of fact, require a more frequent follow up.
The main outcomes were three. First, whether blood pressure measurements were taken in doctors’ premises or patients’ homes. Not, one notices, standing, sitting, or lying or indeed standing and sitting and lying, nor at work, the chemists’, the gyms, at blood transfusion sessions, or other incidental times, by one dedicated person or by any body handy. The data base was kept to an absolute minimum. There is no record of the effect the research nurse might have or ought to have had based on her personality. Then, patient satisfaction was considered as if this had something to do with it. There is no criteria by which the ordinary patient is in any position to know if he ought to be properly satisfied or whether he is being patronised with the doctor‘s questionnaire. In any event we are not selling goods, insurance or computers. And finally to be considered was medication adherence which would appear to be crucially fundamental to any outcome of this “work”. We find that 20% of the time the patient may not take his tablets and presumably does not need them, yet is considered properly treated. Dichotomous variability is used in this paper but is too fanciful a term for me to understand and I think any other ordinary mortal working at the periphery in any of 140 or so countries. No doubt the blood pressure was up or down. Death from essential hypertension in a group of these six hundred might be quite rare over a three year period. If hypertension causes stroke, or coronary thrombosis, or whether some other effect is the actual cause of death I do not know. But we are told that 67 dropped out over three years though not how many died, especially once excluded, which would have been useful (for even delayed satisfaction purposes). 296 per group is not enough to base the whole care of the middle aged on, and neither is three years.
In the paper, statistical operands were applied, constructed or inferred and blood pressure readings were accepted up to 5mm Hg wrong either way. The tables are poor. In table 1 we are told that “organ damage” amounted to 39 and 42 persons in each group but that retinopathy, LVH, and tia/stroke amounted to 42 and 39 respectively. Are they meant to be inclusive or additive? The mean time for incidentally visiting the patients in the 6month group was 2.16 months, and the mean time for visiting the 3 month group was 1.89 months. The question must be did anyone measure the blood pressure at those visits? If not, why not? If they did, then clearly the purpose of this “work” seems to have been lost. Is it not negligent not to take an opportunity to measure the B.P. of anyone currently under treatment after 1.89 months? Table 2 reports the blood pressures over time. The dichotomous variable activity described above has now been dropped and blood pressures are now recorded to the second decimal point. Some sudden dedication! Table 3 records those who escaped control and those presumably who dropped out. How much out of control we don’t know. It was that which was judged ‘by doctor’. We do need to know how that was done. Who said what and why?
Blood pressure is controlled by more, and not less, assessments of the observation one is trying to control. Hourly, daily, monthly, regularly and irregularly, rested or unrested, it should give the patient his own oversight on blood pressure control and this is properly his own interest and responsibility. He may well look after it better than any physician. Mothers should teach their children how to do it. They should teach it at school. Taking my own wrist measurement blood pressure takes 40 seconds start to finish and 2 minutes to take three readings which gives a better consistency appreciation. Current electronic models cost about £40 pounds sterling.
Research at “family practitioner” level wastes time and money. We keep on taking the money. Don’t we? This “work” and its outcome are contrary to the obvious and the test is - would any physician leave himself 6 months or even three months unmonitored? Research work should not be passing interests or games or poorly founded or merely good intention. Nothing that matters can come of this quality of research (and of a great deal more in recent years - almost all of it frankly). But this “work” will gratifyingly decrease our workload if taken on. In England at least family physicians are still generally 9 to 11 Monday to Friday, so it may well help.
We should also be entitled to an indefinite article in front of our job description. ‘Doctor says this or doctor says that’ properly went out years ago. Ours should be the first of the professions to dump all whiff of the mutual petty self-appreciation societies which bedevil them - all of them I think.
Competing interests:
None declared
Competing interests: No competing interests
Mac Donald´s ironical sugestión resumes the study´s central idea:
“follow up intervals are based on phases of the moon because nothing
better exists”. But, dangerously, conclusions glances an apparently,
simple road. EBM, magíster dixit, points out that every six months follow
up is the same as three months. But the question is what sort of
“evidence” is it ? This is an unidirectional evidence; “evidence” from
doctors trying to find, to prove, their own ideas: we can read in the
introduction: “doctors must decide how often to follow up ...”.
Evidence must be an intersubjective agreement, in a bidirectional manner,
a co-construction in the relationship. This means a reasonable, an ethical,
way to practise Medicine. Medical inquiry needs an epistemological stance
far away from objectivism and de-subjectivation. Follow up must be
tailored in the relationship. This is the better which could exist. This
interesting study at all invites us to think in the ancient saying:
“treating the patient rather tran treating the number”. This is what
really adds, once more.
Competing interests:
None declared
Competing interests: No competing interests
Birtwhistle RV et al - Randomised equivalence trial comparing three month and six month follow up of patients with hypertension by family practitioners
We welcome the paper by Birtwhistle et al reporting an equivalence
trial on follow up for hypertension(1). In the preamble the authors
remark that 13% of hypertensive patients are adequately controlled and yet
their sample size calculations appear to be based on 40% control(2). Even
by the end of the first year the numbers of patients remaining in the
study demonstrate a significant shortfall per group. We are not informed
whether the attrition was equally distributed between the groups. On this
basis we are concerned about the safety of their conclusions.
The question of follow up visits raises the issue of workload in
primary care at a time when the National Service Frameworks increase the
prospect of more patients being ‘followed-up’ in UK primary care. The
patients in this study apparently visited the doctor every 2 months. In
the UK patients are reported to consult their GP on average three times
per year and more often as they get older. Opportunistic surveillance
obviates the need to embargo further appointments for ‘follow up’. We
note that patients who visited the doctor less thought that the doctor did
not ‘take their blood pressure problem seriously enough’ and yet the
doctors apparently achieved equivalent control between the groups. This
underlines the danger of creating a dependence on doctors and nurses by
organising care for chronic illness around ‘follow up’ visits.
We note also that GPs were invited to ‘use their own judgement’ about
which cases were ‘out of control’. It would have been helpful to see how
many cases judged on objective criteria were excluded from such definition
by the doctors. A local study reported reticence among colleagues to
medicate some patients even when encouraged to reduce blood pressure down
to recommended levels (3). This suggests that the management of
hypertension is not considered a matter of simply working to guidelines
and raises further doubts about the value of ‘follow up’ appointments when
patients are already visiting their general practitioner for a variety of
reasons.
On behalf of the Journal club.
Mosborough health centre.
Sheffield. UK
References:
1. Birtwhistle RV, Godwin MS, Delva MD, Casson RI, Lam M, MacDonald
SE, Seguin R, Ruhland L. Randomised equivalence trial comparing three
month and six month follow up of patients with hypertension by family
practitioners.
BMJ. 2004 Jan 24; 328 (7433): 204. Epub 2004 Jan 15
2. http://www.ucalgary.ca/~patten/blackwelder.html (accessed Feb
2004)
3.Jiwa, M. & Mathers, N.J. (2000) ‘Auditing the use of ACE
inhibitors in hypertension. Reflecting the cost of clinical governance?’
Journal of Clinical Governance, 8:(1) 27-30.
Competing interests:
None declared
Competing interests: No competing interests