Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
an apple a day?
- Stephen F Hayes (18 April 2003)
-
Brief interventions have useful long-term results
- David A Brown (18 April 2003)
-
Attending to the important but not urgent
- Derek J Marshall (25 April 2003)
-
Behavioural conselling to increase consumption of fruit and vegetables in low income adults
- Thomas R. King (25 April 2003)
-
Re: Brief interventions have useful long-term results
- David A Brown (28 April 2003)
-
Re: Brief interventions have useful long-term results
- Robert G Bunney (21 May 2003)
-
More data please
- Dorothy EM Mackerras (6 October 2003)
|
|
|||
|
Stephen F Hayes, hospital practitioner dermatology Isle of Wight
Send response to journal:
|
I am a doctor and an enthusiastic amateur/going on professional apple grower. With my life partner I planted and tend 1,000 apple and plum trees and am wondering if I should give up medicine altogether to spend more time with my apple trees. As well as apples we grow trees and sell them at car boot sales which are frequented by some of my patients. I can afford to sell trees at low prices. Growing food bearing trees hopefully helps put people back in touch with nature, perchance to improve their mental health, and also hopefully improve their diet. As I am losing my faith in the way we in the UK now choose to practice and organise medicine, and growing in my love for apples, how should I choose to spend the final decade of my active career most fruitfully? Growing antioxidants or prescribing anticoagulants? Competing interests: doctor and apple grower |
|||
|
|
|||
|
David A Brown, psychologist Airport Health Centre, Mascot Australia 2020
Send response to journal:
|
The study by Andrew Steptoe and colleagues on dietary change reminded me of a coronary risk reduction project that I was involved in during the late 70s. My role was to deliver behavioural weight reduction groups to those people from population of 6000 who wished to attend. To my surprise, a single 1-hour counselling session from one of the doctors had the same long-term benefit as my intensive 12-week behavioural weight loss course. The key factor turned out to be not the treatment given, but the degree of interest of the individuals involved. At the 3-year followup those who responded quickly to the initial invitation had lost weight, whereas those who had responded slowly to that initial invitation had gained weight. The treatment given did not predict weight loss at all. There was however a different pattern. Those people who responded quickly and attended the 1-hour session lost weight slowly over 3 years, whereas those same people who attended my sessions lost weight quickly and then regained a fair part of what they had lost. By the 3 year point the two groups had converged. Whereas those who responded slowly gained weight no matter what course they attended. My intensive courses were in an important respect a negative factor compared to one good counselling / education session. They took much more of the person's time, and they subjected at least some of them to a weight loss/weight gain cycle, which is probably unhealthy. That experience convinced me of the value of brief interventions. I now rarely see a person for psychological treatment on more than two occasions, and in most cases I see them only once - on the grounds that if they didn't get it the first time, they probably won't get it the tenth. The other important lesson that I drew from this experience was that researchers who use volunteers, get a positive result, and then say "If only everyone could have this treatment", are kidding themselves. With behaviour change research you get the best results with the pick of the crop - the first people through the door - and the results will inevitably fade the longer you leave the door open. Competing interests: None declared |
|||
|
|
|||
|
Derek J Marshall, Locum Consultant Psychiatrist, Psych ICU Ealing Hospital, Uxbridge Rd, Southall, UB1 3EU
Send response to journal:
|
It is indeed a shame that research such as this does not receive greater prominence. So many of us are caught up in doing important and urgent things - sedating the dangerous, disturbed patient, referring to hospital the man with angina , draining the subdural - that we are often thankful for the respite afforded by patients who are not particulraly unwell and don't require urgent treatment. There is a natural tendency to want to diagnose 'non-case' = no problem = do nothing. Furthermore, the sexy modern world of evidence-based medicine, in addition to the resources devoted to trials which produce evidence for a half-point change on scale x, relegates relatively low-cost pragmatic interventions - with potentially more far-reaching benefits - to 'cinderella activities'. The cost of all of this is that we neglect the non-urgent, but extremely important, aspects of healthcare, such as educating our patients about the long-term value of a healthy diet. In a simplified ideal world the mundane advice of 'an apple a day' would make exciting trials and new evidence redundant - but then who wants that ? Competing interests: None declared |
|||
|
|
|||
|
Thomas R. King, Medical Writer Merck and Company, Inc. HM-215 PO Box 4, West Point PA 19486-0004
Send response to journal:
|
In the title, abstract, and introduction (and eventually in the results and discussion), this article characterizes results of a trial carried out on a relatively mixed income sample as being specific to “low- income” bracket individuals. This would seem to be an fallacy based on a flawed study sample. 32% of the population studied fell above the low income threshold level. In the “assessment of income” section, the authors do not attempt to explain why people above the threshold were included in the study sample. They simply say: “We intended to investigate only low income adults, but some higher income volunteers also took part.” This is understated considering the proportion of the study population that was above the income threshold. This probably wouldn’t be so critical if the authors had not linked low income to low consumption of fruits and vegetables early in the introduction. Furthermore, the authors seemed to downplay the importance of the sample flaw in their conclusions, leaving it to the reader to consider whether or not income level is even a critical factor in dietary modification strategies. Another problem lies in the use of biomarkers as indicators of vegetable/fruit consumption. The authors indicate that there is a lack of published data to support the use of biomarkers in analysis of dietary interventions. Then, this very tool is used to measure the primary outcome (intake of veggies and fruits). It seems that the use of biomarkers for dietary intake requires further study before being used in an analysis of the type featured in this paper. Perhaps it would have been more relevant to focus on the health behavior models (or maybe to have tried other models) and the typical methods (surveys or interviews) used to evaluate intervention efficacy. The nurses administering the interventions and performing the evaluations were unblinded- this is problematic, and again is not adequately explained. The results of the biomarker tests were of limited utility (beta-carotene and alpha tocopherol were raised but there is limited discussion of the biologic variability that could affect these assays, and there were no changes in ascorbic acid or urinary potassium excretion levels). The necessary blood draws actually may have dissuaded individuals (perhaps some in the intended low income target population) from participating in the study. Additionally, no background is provided with regard to other potentially important social factors, for example, whether or not there are readily available sources of good fruits and vegetables for the study population. Overall, the topic is important enough that concrete scientific data is needed to support use of the proposed interventions. Models of health promotion have long been used with success to change health behaviors. The authors seemed to underestimate the use of the survey for evaluation of dietary change strategy in order to promote the use of biomarkers and technology. The limitations of the study design are understated, and the authors infused the discussion of the results with an air of significance not evident in the results, taking the limitations into consideration. Competing interests: None declared |
|||
|
|
|||
|
David A Brown, Psychologist Airport Health Centre, Sydney Australia
Send response to journal:
|
Correction - the last followup in my weight reduction study was at 22 months from initial weigh-in, not 3 years. Competing interests: None declared |
|||
|
|
|||
|
Robert G Bunney, GP Brannam Medical Centre, Barnstaple, EX32 8GP
Send response to journal:
|
David Brown points out the problem of overestimating the success of behavioural change in the wider population from the results achieved with those responding to invitation to take part in a trial. I note that Steptoe et al sent out 3858 invitations of whom only 775 ( 21%) replied. Of these 775 only 271 were randomised as 188 were excluded and 316 had replied to decline taking part. I also thought it interesting that Minerva noted only the week before that a study in the American Journal of Public Health (2003;93:635-41) calculated it would take 7.4 hours a day for primary care physicians in the US to complete all the services recomended for all their patients. So they don't. Competing interests: None declared |
|||
|
|
|||
|
Dorothy EM Mackerras, Senior Research Fellow Menzies School of Health Research, Darwin, NT, 0811, Australia
Send response to journal:
|
EDITOR - I read with interest the trial comparing methods to encourage fruit and vegetable consumption. However, I was disappointed that the clinical outcomes were presented as "There were no changes in … blood pressure or serum cholesterol" (p857). This could mean either that there was no difference or that there was a trend in one or other direction that was not significant. In an age of evidence-based practice and public policy, it is important for negative trials to publish results in a numeric form so that they can be included in future meta-analyses to allow an accurate picture of the effects. Could the data for these two parameters be presented, e.g. in the form of their Table 2? Competing interests: None declared |
|||