Wednesday, 16 October 2013

'The drugs don't work': And neither do sports suppliments

A paper published last year in the BMJ carefully considered the evidence underpinning sports performance products. It concluded:

'The current evidence is not of sufficient quality to inform the public about the benefits and harms of sports products. There is a need to improve the quality and reporting of research, a move towards using systematic review evidence to inform decisions.'

Despite this, health and fitness magazines continue to proclaim that I and others should embrace supplementation. These adverts are particularly prevalent for whey protein and creatine.* Creatine is a naturally occurring substance that helps supply energy to all cells in the body. Of course, oxygen is also a naturally occurring substance that helps supply energy to all cells in the body, but it doesn't mean I carry an oxygen tank while running.

Not yet anyway.




A typical sports supplement. The word 'PhD'  gives an air of scientific merit. There is none. 


There remain colossal problems behind research involving the use and effectiveness of creatine and other supplements. Companies such as PhD above annoy the hell out of me because they are able to sell something that almost certainly doesn't work**.

General problems across peer-reviewed literature assessing the effectiveness of these supplements include:
  • An overemphasis on male athletes instead of 'average people' who take exercise on a regular basis. 
  • Group sizes that are almost always in single figures and often unbalanced across conditions. Healthy men and women are not a special population so there is no reason I can find to have such a small sample size.  
  • There is therefore a high chance of false positives (i.e. finding a significant effect when there isn't one). 
  • A lack of randomised controlled trials. There is often no true control group and only a placebo.
  • Inappropriate use of parametric statistics given the reported sample sizes and large standard deviations.
  • Little/no control for changes in diet or the amount of exercise taken across a testing period.
  • Little/no control for participants taking other legal or illegal supplements (e.g.. Steroids) 

A growing number of studies are now failing to support the notion that creatine for example, has any beneficial effect on fitness or strength (e.g. Balsom, Harridge, Soderlund, Sjodin & Ekbolm 2008; Rahimi et al 2011). A meta-analysis from 2003 found a very small effect, but this seems a little out of date in light of new evidence.

Economics

The power of advertising makes people forget that they are probably paying for something that looks pretty, but does nothing. Take Maximusle for example. You're meant to start off with a loading dose of 33 grams four times a day. After the first week this drops down to a maintenance dose of 33 grams twice a day. This would cost £960 a year***.

Personally, I would take that money and hire a personal trainer for one hour a week and incorporate those sessions into part of regular exercise programme. Prices vary, but £15 an hour is typical. Fifty-two hours would cost £780 a year.

In saying that I don't suppose the evidence base for hiring a personal trainer is particularity great either, but something tells me that any behavioural intervention has got to be better than an over-priced and over-marketed alternative.



*I appreciate that these magazines are predominately funded by advertising. However, I am more likely to pay more for a publication that wasn't full of s**t.

**See the Sale of Goods Act 1979.

***A 1.1Kg tub (33 servings) costs £48

Weekly Servings
First week = 4*7 = 28 servings
Maintenance week = 14 servings

Monthly Servings
Total servings in first month = 28 + (14*3) = 28 + 42 = 70
Total servings in a regular month thereafter = 28 days * 2 servings a day = 56

Yearly Servings
Total servings in a year = 70 + (56*11) =70 +616= 686

Purchases Required
686/33 = 20 1.1Kg tubs required

Total cost for one year = 20 *48

£960


References

Balsom, P. D., Harridge, S. D. R., Soderlund, K., Sjodin, B. and Ekbolm, B. (2008). Creatine supplementation per se does not enhance exercise performance. Acta Physiologica Scandinavica. 149(4), 521-523.

Branch, J. D. (2003). Effect of creatine supplementation on body composition and performance: a meta-analysis. International Journal of Sport Nutrition and Exercise Metabolism. 13(2), 198-226.

Murphy, A. J., Watsford, M. L., Coutts, A. J.and Richards, D. A. B. (2005). Effects of creatine supplementation on aerobic power and cardiovascular structure and function. Journal of Science and Medicine in Sport. 8 (3), 305-313.

Rahimi, M., Kordi, M., Karimi, N., Gaeini, A., Samadi, A. and Moradi, N. A. (2011). The effects of whole body vibration training on creatine supplementation on lower extremity performance and balance in elderly males. Iranian Journal of Aging. 6(19).

Sterkowicz, S., Tyka, A. K., Chwastowski, M., Sterkowicz-PrzybycieĹ„, K., Tyka, A.and Klys, A. (2012). The effects of training and creatine malate supplementation during preparation period on physical capacity and special fitness in judo contestants. Journal of The International Society of Sports Nutrition. 9 (41). 1-8.

Sunday, 25 August 2013

Tracking CPU temperature over time

Temperature Gauge is a neat little app that logs the temperature from every sensor in any Mac. From around 2007, Apple has continued to include a bewildering number of these sensors, but I'm just going to look at a single example that can be found on any computer - The CPU.


Experiment 1: iMac (2013) vs Macbook Pro (2010)

Taking an iMac from 2013 (quad-core Intel Core i5) and a 13 inch Macbook Pro from 2010 (Intel Core 2 Duo), I compared their temperatures from switch on. After the operating system was loaded, each computer played a standard definition video lasting 5 minutes.

To keep things as controlled as possible, both computers were running the same version of Mac OS X and were tested at the same time in the same room to keep the ambient temperature constant. The fans in each machine were kept at a fixed rate using smcFanControl.

As expected, the iMac consistently runs at a reduced temperature because it contains a more efficient processor and provides more volume for air to circulate.

What you might not expect is how the two processors respond over time as the general trends are almost identical in terms of how the temperature rises and falls at specific points in time.


Experiment 2: 13 inch Macbook Pro (2010) vs 15inch Macbook Pro (2007)*

I also managed to get my hands on a 15 inch Macbook Pro from 2007 and ran the same test.

Again, the general trends were similar, but a larger laptop, despite being several years older and running a rather outdated version of the Intel Core 2 Due, ran consistently cooler.

Breathing space appears to be the most important variable**, but I wonder how this affects the lifespan of the CPU. Does a larger Macbook Pro last longer because it consistently runs at a lower temperature?

Maybe, but it's probably not worth asking Apple for the numbers:)



*Two laptops serve as a better comparison because they utilise fans of identical size.

** I should also add that the insides of all three machines were cleaned to ensure that a build up of dust inside would not influence the results. 

Sunday, 2 June 2013

How predictable is English Football? Using linear regression to forecast future league positions

Sport produces a lot of data that pundits generally ignore. Somehow, they manage to spend hours debating how a season of football was hampered or improved by the actions of a few individuals. Usually referees.

My gut feeling has always been that success for any English Premier League football team is highly dependant on financial management. This varies considerably between clubs.

Curiosity got the better of me on this one.





In statisticslinear regression is an approach to modelling the relationship between a scalar dependent variable y and one or more explanatory variables denoted XIt was the first type of regression to be studied seriously and the first to be used extensively in practical applications.

In this instance, my dependant variable is league points awarded this season, and my predictors are turnover, profit/loss before tax, net debt, interest owed on any debt and the club's wage bill. These predictors were all taken from the previous season.*

A stepwise multiple regression was conducted to evaluate whether any financial indicators were necessary to predict total points as of June 1st 2013. At step one of the analysis, annual turnover was entered into the regression equation and was significantly related to points awarded [F(1,15) = 60.35, p < .001]. The multiple regression correlation coefficient was .89, indicating approximately 78.8% of the variance in total points could be accounted for based on a club's turnover. The remaining variables: profit/loss, wage bill, net debt and interest payable were not entered into the equation [p > .29].

The regression equation can therefore be defined as:

points = .191turnover +29.48

While termed a regression equation, this is essentially the same as any equation referring to a straight line (y = mx + c).



Plot demonstrating a strong correlation and predictive power between the number of points awarded this season and annual club turnover from the previous year. League position is also noted. 

In other words, if you wanted to predict how many points a team might win next year, any prediction would be well advised to take into account a club's turnover at the end of the previous season. For example, imagine Newcastle United's turnover increases from 93 to 120 million. Entering this into the above equation, their points this time next year would be 52 (an increase from 41 at present).

Of course, this regression analysis only takes into account the data entered into the model in the first instance. Additional financial data should produce a more accurate model as almost all of my predictors (with the exception of club debt) correlated with total points awarded.

The small sample size is also of concern. Sadly, this model is almost certainly not as powerful as the numbers suggest so I wouldn't head to William Hill just yet. In particular, teams with less variation in points and turnover may be more difficult to predict as they cluster together. That said, it would be interesting to combine data from every team across each divisions over a number of years. With enough enough historical information, a superior model could forecast league position with some accuracy.

Such a result may appear to be intuitively obvious, but this reality isn't reflected in current media coverage. Despite a huge amount of time devoted to football panel shows, money alone appears to be a strong predictor of success. For example, Alex Ferguson was clearly a great manager, but part of that success hinges on the fact that he was based at a very rich club for most of his career. That alone may account for a large percentage of his success.

But no one really wants to talk about that.

I appreciate that being able to predict even 1% of a future sporting performance remains very important. In a 100m sprint, it could make the difference between gold and bronze. However, given the repetitive nature of football league tables north and south of the border, 1% would make very little difference.

This is particularly true in Scotland where no team outside the old firm have won the league in my lifetime!


*Data taken from http://www.guardian.co.uk/football/2013/apr/18/premier-league-finances-club-by-club


Sunday, 14 April 2013

Going beyond the 'nudge': Could supermarkets do more to encourage a balanced diet?

It is well acknowledged that what we eat is important and UK Governments have spent a small fortune on campaigns that encourage people to think carefully about what they eat.

These are typically known as nudge strategies. As the name suggests, they attempt to 'nudge' the public into changing their behaviour.




Whether these actually work remains open to debate. For example, most people who smoke know it is bad for their health, but reminding them of that fact will not always have the desired effect on their behaviour. On the other hand, the public smoking ban has been very effective in helping people kick the habit because it removes temptation and encourages people to adopt effective coping strategies. I guess that could be described as a bit more than a nudge!

Anyway, this got me thinking about diet as I was walking around the supermarket. Glancing at the shelves, it struck me that temptation is everywhere. Every aisle is packed with products that contain too much fat, salt or sugar. On rare occasions when you do find something that is healthy for example, a breakfast cereal that isn't packed with sugar, you are only one or two moves away from another product that is often cheaper, but comparatively unhealthy.



My view is that the government can nudge people in the right direction all they want, but when it comes to regulating the way food is pushed towards consumers at the point of sale, something that is well known to affect shopping behaviour, little is done to help encourage people towards a healthier alternative.


One Idea

Supermarket aisle's should not just be split by food type, but by health rating and then by food type. This would allow for a clear distinction between food than should be consumed regularly and occasionally.



The outcome would be two-fold. Firstly, it would be clear to customers what should and should not be consumed on a regular basis. Secondly, while not limiting individual choice it may encourage new patterns of shopping behaviour.

This might make for a really neat experiment.

Participants would be instructed to keep a diary of their normal food consumption and in a between-subjects design, control participants would then be asked to shop from an mock online supermaket orgaised in a traditional manner. An experimental group would shop from a supermarket organised in the new manner shown above.

My prediction: participants in the experimental group would place more healthy items in their basket when compared with controls. They would also show a reduction in total basket calories in comparison to their average shop.

Assuming customers spent the same amount of money or more, this would also make good business sense...especially in the long run if those same customers lived longer!



Thursday, 7 March 2013

The evolution of mobile phone design

*Updated 6/01/2014 [12th mobile]

Hard to believe that I am now on my 11th mobile phone.

Trying to convince my parents that I 'really needed' a mobile phone at 13 years of age was a challenge in 1999.

I also remember my mum being mortified as she contemplated the idea of me walking down the street 'talking to someone on the phone like a right poser'. 

Anyway, after a quick Google image search, I have compiled my own mobile phone timeline. Enjoy.


















Monday, 21 January 2013

DSM 'Field Study' Results

The Diagnostic and Statistical Manual of Mental Disorders (DSM) published by the American Psychiatric Association is tasked with providing a common language and standard criteria for the classification of mental disorders. In theory, this has many benefits for those working across different disciplines.

Version 5 has taken a decade to compile from 1,500 experts in psychiatry, psychology, social work, psychiatric nursing, paediatrics, neurology and other related fields from 39 countries.

Results from the DSM 'field trials' have now been published [1]. 

In several studies, the reliability of the new diagnostic criteria for different psychiatric disorders was measured.

Two different psychiatrists assessed each patient, and the agreement between their diagnoses was calculated using the kappa statistic, where 0 indicates no correlation at all and 1 is perfect.







Unfortunately, the reliabilities of most DSM-V disorders were not very good. This included staples such as schizophrenia, bipolar disorder, and alcoholism. Others were even worse. Depression, had a kappa of 0.28, and the new 'Mixed Anxiety-Depressive Disorder' scored -0.004. This diagnosis can now be described as meaningless.

The DSM-V field results are in fact worse than the results for the DSM-III in 1980. This version has remained mostly unchanged for the last 30 years (DSM-IV made faily modest changes). The reliabilites have got worse, despite an editorial claim of 'continued progress'. That said, the DSM-5 trials were larger and conducted in a slightly different fashion, but given years of development these results remain extremely disappointing.


The DSM-V may simply be ignored as practitioners in the US and UK are allowed to do just that. But a decade of research should not result in a book that gathers dust.

[1] Freedman R, Lewis DA, Michels R, Pine DS, Schultz SK, Tamminga CA, Gabbard GO, Gau SS, Javitt DC, Oquendo MA, Shrout PE, Vieta E, and Yager J (2013). The Initial Field Trials of DSM-5: New Blooms and Old Thorns. The American Journal of Psychiatry, 170 (1), 1-5 PMID: 23288382

Sunday, 20 January 2013

Favourite PowerPoint Quotes

'I hate the way people use side presentations instead of thinking'

'People who know what they're talking about don't need PowerPoint'





'PowerPoint presentations too often resemble a school play - very loud, very slow, and very simple.'





'UsingPowerPoint is like having a loaded AK-47 on the table: You can do very bad things with it.'





'The genius of it is that it was designed for any idiot to use. I learned it in a few hours



'Trust yourself to be a bridge between simple content and complex ideas.'