How to read a paper

[GuinessLover]

How to review a paper

Here I am going to discuss how you go about reviewing a paper and come to the informed opinion on whether the results the paper has come to are valid. To do this I am going to review a paper which hopefully is interesting to the audience on this forum. I have been taught throughout my degree how to review papers, and in the near future will be involved in a research project myself, which may result in a paper being produced.

If you already know how to review a paper, I am happy for you to nitpick at my techniques and offer me some advice, but for those who do not or have never needed to review a paper, I hope this will help you understand what makes a good paper from which the conclusions are valid.

Unfortunately this basic technique is not always present in journalists, who occasionally come to writing ridiculous articles, such as the infamous MMR controversy, caused by Dr Andrew Wakefield.

The article I’m going to use to show you some techniques to help you review a paper is ‘Long-term creatine intake is beneficial to muscle performance during resistance training’ by K. VANDENBERGHE, M. GORIS, P. VAN HECKE, M. VAN LEEMPUTTE, L. VANGERVEN, AND P. HESPEL.

I comprehend that this paper may be considered to be out of date (1997), but I’m using it is free access, and will hopefully appeal to the audience on this forum (I do apologise if it doesn’t).

Unfortunately I cannot yet post URLs, but if you google the title you'll get straight to the full paper I'm going to review.

If anyone would like me to continue explaining how to review a paper and decided whether it's conclusions are valid, please state your interest below!

[ryandj2]

http://jap.physiology.org/cgi/reprint/83/6/2055.pdf

[comanight2]

im interested lets hear it dude!

[GuinessLover]

Here we have this relatively big, nine page article, with lots of long words such as ‘dynamometer consisted of a computer-controlled asynchronous electromotor’ and ‘fatiguability of the arm-flexion muscles’.
The layout looks boring, and really no-one wants to read it.

Saying that, I personally found it quite an easy read compared to most papers I’ve read, but I doubt everyone will.

So how do we go about reviewing it?


1)

First we look and see which journal the paper has been published in. Is this a journal which you or I would read? I don’t know the number of people who read journals here, but I wouldn’t say many of the general population do, and I suspect less people read The Journal of Applied Physiology.

After a quick look on the internet, I am told that the ‘impact factor’ (the effect the papers have on recent research in that field in that journal, and therefore how important the papers in that journal) of The Journal of Applied Physiology is:-

Quote

The four journals with the highest impact factors in sport and exercise science are Exercise and Immunology Reviews (impact factor 2.9), American Journal of Sports Medicine (2.3), Medicine and Science in Sports and Exercise (2.1), and Journal of Applied Physiology (2.1)

I would put the link down from where I got this data, but unfortuanately it won't let me put up URLs

So we’ve decided that this paper is going to be read, or at least cited by quite a few people in the world of sport and exercise science. Therefore we can assume that this article is going to be ok.

[GuinessLover]

2)

Next, we look at who’s funding this paper. To do this we have to go to the section at the bottom of the paper just before the references.

Quote

This study was supported by Grant G.0189.96 from the Belgian National Medical Research Council (Fonds voor GeneeskundigWetenschappelijk Ondersoek). Financial support and creatine and placebo supplements were kindly provided by Novartis Nutrition (Berne, Switzerland). The subjects’ sports outfits were supported by NIKE
(Belgium). Address for reprint requests: P. Hespel, Faculty of Physical Education and Physiotherapy, Exercise Physiology Laboratory, Tervuursevest 101, B-3001 Leuven, Belgium.

From this we can understand that the Belgian National Medical Research Council has offered some financial support [don’t the have better things to be giving money to e.g. cancer research]. Underneath that we are told that Novartis Nutrition supplies more financial support, the creatine and placebo. NIKE supply clothing.

I’m guessing Novartis Nutrition has quite an interest on what creatine does, especially if they can persuade the general market to buy it off them for a fee. Why Nike got involved don't as me?

Does this downgrade the quality of the paper? I would say no at this stage, as papers like this need the funding, and you’ll find most similar papers will have funding from a nutritional company. But keep it in mind.

[GuinessLover]

I will continue, just going to the gym. My thanks go to ryandj2 for the full paper link.

Coming up next is how we look into the Introduction and Methods of the paper. Then I'll discuss the Results. Just a quick one, do people want me to go into detail about stastical interpretation of data or just give a quick overview when it comes to results?

P.S. Admins is this in the right place?

[Weaves]

Include statistical interpretation also. I imagine quite a few people would not have a clue.

[ryandj2]

Quote

1)

First we look and see which journal the paper has been published in. Is this a journal which you or I would read? I don’t know the number of people who read journals here, but I wouldn’t say many of the general population do, and I suspect less people read The Journal of Applied Physiology.

After a quick look on the internet, I am told that the ‘impact factor’ (the effect the papers have on recent research in that field in that journal, and therefore how important the papers in that journal) of The Journal of Applied Physiology is:-


I would put the link down from where I got this data, but unfortuanately it won't let me put up URLs

So we’ve decided that this paper is going to be read, or at least cited by quite a few people in the world of sport and exercise science. Therefore we can assume that this article is going to be ok.

Not strictly true - it is a good indicator of the quality of the journal overall, but there are issues with it. It is an average of all citations to articles in the journal, so doesn't mean any one individual paper would be good quality or have a high impact.

[GuinessLover]

Well picked up on ryandj2. I was just using it as a brief qualitative assessment on how good the The Journal of Applied Physiology is considered in exercise and sports science. You do need to assess the quality of the paper as an individual paper. Even the best journals print the bad papers e.g The Lancet printing the aforementioned Dr Wakefield paper.

[GuinessLover]

3)

Now we move onto the introduction.

Quote

THE EFFECTS of oral creatine supplementation on contractile performance and metabolism of skeletal muscle recently have become an area of major interest in exercise physiology. Thus evidence has accumulated over the last several years showing that short-term high-dose creatine intake may elevate muscle creatine stores, predominantly in its unphosphorylated form (6, 11, 14), and improve one’s capacity to perform maximal intermittent exercise (2, 5, 6, 12, 24) but not a single bout of either high-intensity (7, 18, 19) or endurance type (1, 23) of exercise. The physiological mechanisms linking creatine loading to enhanced exercise performance remain, however, largely unexplained. While research to date mainly has focused attention on shortterm effects of oral creatine ingestion, long-term oral creatine supplementation has developed as a widespread ergogenic practice among various categories of athletes. Yet the physiological impact of long-term creatine intake on the human body remains virtually unexplored. On the one hand, a recent study (15) has documented that the higher concentration of muscle creatine achieved after short-term creatine loading may be conserved for a period of 4 wk by continued creatine ingestion in a dosage equaling daily body creatine loss. On the other hand, fragmentary in vivo and in vitro evidence in literature suggest that creatine probably is needed for additional cellular functions than just phosphocreatine (PCr) production. Animals in which muscle creatine depletion was induced by feeding of creatine analogs have been shown to exhibit growth retardation and general weakness and at the same time to develop muscle ultrastructural abnormalities, including disruption of thin myofilaments, dilation of mitochondria, and disruption of Z bands (17, 20, 25, 28). Interestingly, such ultrastructural abnormalities resulting from abnormalities in creatine metabolism also have been associated with the incidence of muscular dystrophy or myopathy (27). Furthermore, on one hand, prolonged oral creatine supplementation has been shown to increase type II muscle fiber diameter (17) in patients with eye muscle atrophy. On the other hand, Earnest et al. (9) have reported that short-term creatine intake increases fat-free mass in strengthtrained athletes. Accordingly, some (4, 16) but not all (10) in vitro findings indicate creatine stimulates the biosynthesis of muscle myosin. Given these earlier in vivo and in vitro findings, it is reasonable to speculate that oral creatine supplementation may exert an anabolic action in humans and thereby is likely to enhance the effects of resistance training on muscle mass and strength. The aims of the present study, therefore, were 1) to evaluate whether creatine supplementation may add to the effects of resistance training on muscle strength and on the capacity to perform high-intensity exercise and 2) to evaluate the effects of long-term creatine supplementation on body composition.

Here they give a lot of scientific waffle about why they are doing this study. I believe that there was sufficient gaps in research to warrant this study to occur. Read it if you have little knowledge about the subject matter then pick out what the aim and hypothesis are.

The aims make sense to me. The hypothesis is one of that oral creatine will lead to an anabolic effect. Unfortunately they do not state, in numbers, the effect they expect creatine to have on muscle mass and strength. A qualitative hypothesis would’ve been better.

For example ‘We predict an increase of muscle mass of 5kg with those on creatine compared to those who are not.’ This could lead to a tiny increase e.g. a gain of 100g of muscle mass over the study period, to be claimed a success, which it is quite likely that they will do.

This lack of qualitative hypothesis prevents a prospective power calculation from occurring which I consider vital.

I will discuss power calculations in another section.

Ah well, we can only plod on through and look at the method.