Never Mind the Content, Measure the Impact: Spin Bowling for Journal Impact Factors vs. the Importance of Patient Impact
G. Jackson
DOI: https://doi.org/10.1111/j.1742-1241.2010.02511.x
IF: 2.6
2010-10-01
International Journal of Clinical Practice
Abstract:Journals aim to improve their ‘Impact Factor’ by publishing articles of outstanding merit that as a result are widely cited. Could it be, however, that some articles achieve citations based on controversy rather than good science, and do they as a result achieve publication prominence based on the Impact Factor generating potential? The juxtaposition of a statement on conflict of interest with a very dubious article on testosterone replacement places the conflict responsibility on authors without a mention of the vested interest (real, potential or imagined) of the journals (1,2). In this study, frail elderly men of mean age 74 years who had limited mobility were randomised to receive either testosterone gel or placebo. The objective was to see if testosterone could improve mobility, which it did, but the study was stopped prematurely because major cardiovascular events were noted to be more frequent in the testosterone compared with the placebo group. The results appear alarming until the data is scrutinised – the testosterone group had more pre-existing cardiovascular risk factors such as hypertension or hyperlipidaemia (including statin therapy), and cardiovascular disease was not monitored directly. Added to this, the target testosterone was 17.4 nmol ⁄ l (in frail old men!) that required dosage regimes initiated outside the recommended starting regime and in 15% above the approved dose regime. The findings are at variance with other studies that show no cardiovascular signal – and this combined with no consistent pattern of cardiovascular events – strongly suggests that the differences were attributable to chance alone (3,4). So, was it responsible to publish even with a balanced linked editorial? (5) Citations yes, but value? Responsible publishing is further called into question by a meta-analysis of angiotensin-receptor blockers (ARBs) and cancer (6). This is an article of very little, if any, scientific merit that has had important implications for patient adherence to therapy as a result of media interpretation being, as usual, sensational. What is most concerning is the failure of the authors to follow meta-analysis guidelines and to point out that cancer screening at entry was not part of the trial protocols, yet they suggest a ‘modest but significant’ link between ARBs and cancer. Lancet Oncology, but not The Lancet, accepted the article and its Impact Factor will have certainly been enhanced, but was this responsible editing? It certainly could be considered irresponsible to have the insubstantial, uncritical linked editorial written by someone very closely linked academically to the authors (both from Cleveland Ohio) (6). Just considering the pathophysiology, how can a drug exposure with ARBs for 2–4 years (the trial’s duration) be in any way linked to cancer, unless cancer was undetected and not screened for at study entry. We have been this way before with calcium antagonists that we were virtually alone in defending (7). Calcium supplements in people with osteoporosis are now questioned as a cause of myocardial infarction (MI), but not stroke, death or the composite endpoint of MI, stroke or sudden death (8). The absence of a link between MI and mortality coupled with none of the studies in the meta-analysis having cardiovascular outcomes as primary end points with no systematic gathering of cardiovascular data does make you wonder what is the point of it all. Armchair science when you have nothing better to do – painting the shed may be more constructive. Well, I suppose I have fallen into the spin-bowler’s trap and cited the articles as the journals wanted. There is, however, a bigger picture here – these studies have important patient impact as well as journal impact, and we do a disservice to our patients if we prioritise highly doubtful analyses. Journals need to be just as responsible as authors.