Rapid Responses to:

LETTERS: George Davey Smith and Matthias Egger Meta-analyses of observational data should be done with due care BMJ 1999; 318: 56 [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

Misleading inaccuracies do more harm than good

F P Cappuccio   (13 January 1999)

Misleading inaccuracies do more harm than good 13 January 1999
F P Cappuccio, Reader in Preventive Cardiovascular Medicine St George's Hospital Medical School Send response to journal: Re: Misleading inaccuracies do more harm than good	Dear Sir, In their continuous anxious and compelling search of faults in published research, Davey Smith and Egger's latest critical exercise (2 January, p.56) falls short of being accurate. In their reference to a published meta-analysis of observational data(1)and the re-analysis that followed(2), they commit two sins. The first is to misquote the original paper and the second, more important, is not to cross-check the facts. They quote that "Cappuccio et al argued that this could be expected ". Nowhere in the original manuscript is there such a statement. This interpretation has been carefully construed from the paragraph discussing potential sources of heterogeneity where the authors conclude that " it seems more likely that other characteristics, either not measured or not considered in the analysis, may explain the differences across studies". "Correcting the meta-analysis for this error (and several other mistakes) " they then go plotting Birkett's estimates in the figure. However, they have conveniently ignored our correspondence to Birkett's publication(3)where we clearly challenged some of Birkett's calculations. Davey Smith and Egger's letter seems to crucify human error as misleading meta-analysis without realising that whilst the inaccuracies discussed (and in part ackowledged(3)) have led to minor changes in the overall estimate of effect (the relevance of which is minimal and openly discussed in view of the results of controlled trials(4)), their letter intends to mislead by the use of misquotes and inaccuracies those readers who are unlikely to go and read the original publications. Yours sincerely Francesco P Cappuccio Reader in Preventive Cardiovascular Medicine Department of Medicine St George's Hospital Medical School, London SW17 0RE References 1)Cappuccio FP, Elliott P, Allender PS, Pryer J, Follman DA, Cutler JA. Epidemiologic association between dietary calcium intake and blood pressure: a meta-analysis of published data. Am J Epidemiol 1995; 142: 935-945 2)Birkett NJ. Comments on a meta-analysis of the relation between dietary calcium intake and blood pressure. Am J Epidemiol 1998; 148: 223- 228 3)Cappuccio FP, Elliott P, Follmann D, Cutler JA. Authors' response to "Comments on a meta-analysis of the relation between dietary calcium intake and blood pressure". Am J Epidemiol 1998; 148: 232-233 4)Cappuccio FP. The "Calcium antihypertension theory". Am J Hypertens 1999; 12: in press.