...nor have there been knock-out blows to the severe sepsis study, also conducted at the University of Vienna.
.
I'd certainly like to comment on the severe sepsis study.(Adjunctive homeopathic treatment in patients with severe sepsis: a randomized, double-blind, placebo-controlled trial in an intensive care unit: Homeopathy 2005 (94) 75-80.) As background, I am a specialist in intensive care and treat patients with sepsis every day. I am therefore very familiar with the literature in this field and am a reviewer for several ICU journals.
Mr Gully whilst I do not wish to be rude, suggesting that this study is good positive evidence of homeopathy's efficacy suggests that either you have not read it or more likely you do not know enough about the subject.
First, this is a small study. Seventy patient were enrolled, which for a sepsis trial in particular is very low. The reason is that sepsis is a very heterogenous condition, and small trials are unlikely to have well matched patients. As comparisons, recent sepsis trials in the literature randomised 300-500 patients. (Annane et al, Effect of treatment with low doses of hydrocortisone and fludrocortisone on mortality in patients with septic shock, JAMA 2002; Corticus (not yet in press)).
What this means is we have to look very closely at the 2 groups at baseline, and here we see the second major problem. As I mentioned sepsis is very heterogenous. This means outcomes are different for different sites of infection. Urinary sepsis tends to have a good prognosis, whilst intra abdominal is less so. However, the authors do not present this data. Instead in Table 1 they simply give "reason for admission" In this category the majority of patients are classified as just "sepsis" which is meaningless, as all patients would have had sepsis by definition, in order to be included. Thus we have no idea how well the groups were matched in terms of site of infection.
Now we come onto the statistics. Mr Gully claims this study shows efficacy of homeopathic treatment over placebo. However, 28 day mortality was NOT
significantly different between the two groups (Survival at day 30: 81.8% vs 67.7%, P 0.19 Hom vs Plac). At day 180 there is a significant difference in survival (75.8% vs 50% P=0.043). Now this is rather odd. Why should there be no difference at 30 days, but one at 180 days? Sepsis tends to have an early effect on mortality. Most large scale trials will use 28 day mortality as a primary outcome measure, as by this time the majority of patients will either have died or got better. Average length of stay in ICU's is certainly well under 180 days, and is more in the range of 5-7 days for sepsis. In this trial the medication was ceased upon discharge. So how to interpret this strange finding? Well, the sort of standard data for a sepsis trial we need is simply not presented. There is no information on length of stay, time on mechanical ventilation, cumulative inotrope dose, time on dialysis -nothing apart from the baseline information is given. Without this the likeliest reason for the findings at 180 days is just a statistical aberration, in an underpowered study which we have no way of knowing was matched properly.
Frankly if this was a trial of a conventional drug, no reputable ICU journal would publish it in its current form. If a drug rep attempted to convince me to use a conventional med based on this study I would laugh them out of the office. Attempting to trumpet it as a triumph for homeopathy shows an astonishing lack of understanding of the subject and in my opinion highlights the paucity of your arguments.