Data Dispatches @datadispatches - Tumblr Blog

Aviation & Medicine & Culture

Some excerpts from Problems and Perils of Prescription Medicines

<http://www.blacktriangle.org/reason.html>

Aviation [safety] is predicated on the assumption that people screw-up. You (healthcare professionals) on the other hand are extensively educated to get it right and so you don’t have a culture where you share readily the notion of error.

Errors can’t be eliminated but they can be managed.

Error is extremely adaptive

Error is not random, it is not out of the blue, and it is largely predictable.... It is very contextual.

Find error traps -- to do that you need a database, some kind of error collecting system. You need a reporting system; you need a near miss reporting system. You need a memory. Without that you have no idea of identifying where these errors occur.

The other much more manageable area are latent conditions I used to call them latent errors, but they are just latent conditions... they are always there.... So just like mosquitoes you swat one and another appears, so the only way of dealing with the error problem is to go to the latent condition, to go to the systemic swamps in which they breed

Then there is the obvious thing about naming blaming and shaming,

http://www.blacktriangle.org/reason.html

PowerQuery

A good thing

http://blogs.msdn.com/b/powerbi/archive/2013/07/07/getting-started-with-pq-and-pm.aspx

Everything in moderation

5 Reasons Too Much Data Can Be Risky - Five Things - WSJ

Companies are collecting growing volumes of consumer data as they seek to better target shoppers. But a raft of coming new regulations and growing unease from customers, could turn these Big Data investments into liabilities. And reversing gear on dat…

http://blogs.wsj.com/five-things/2014/04/10/5-reasons-too-much-data-can-be-risky/?mod=WSJ_hps_sections_riskcompliance

Every company a data company

One point of view

http://thenextweb.com/dd/2014/04/07/spotify-effect-every-company-now-data-company/

"When a company uses data to personalize its inventory, it creates a better UX. This UX in turn draws in millions of users and investment dollars"

Data on porn

http://jonmillward.com/blog/studies/deep-inside-a-study-of-10000-porn-stars/

Obit: David Margerison Davies

So good I'm posting it again:

His contributions to clinical pharmacology never received adequate official recognition, perhaps because he was intolerant of obstructive administrators and bureaucracy… he it was who invented the beautiful phrase, “Your manuscript has been reworded to assist our translators”, deployed when transforming some ungrammatical gobbledegook into intelligible English. He was a gifted raconteur whose clinical tales of an evening were a rare delight.

Original:

One of the first physicians in the UK to see the importance of detecting and preventing harm from drugs. Devoted himself to spreading this gospel to practising doctors through his pioneering Adverse Drug Reaction Bulletin and Textbook of Adverse Drug Reactions.

…his lifelong interest in rational use of drugs and adverse reactions to them. This interest, he claimed, was sparked by gloom induced by sulphanilamide he was given in the army; by seeing hexamethonium cause acute hypotension in two patients (one a heavy man who collapsed on top of him); and by observing a previously unrecognised reaction to cortisone.

http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(03)12699-2/fulltext

Columbia University - Columbia Data Science Class

So good I want to blog it again...

http://columbiadatascience.com/blog/page/3/

On Kaggle, GE finds data science solutions for patients and pilots

http://gigaom.com/2013/04/03/on-kaggle-ge-finds-data-science-solutions-for-patients-and-pilots/

Summary: GE, an airline and a health system have crowdsourced data science questions using Kaggle and are now paying out $600,000 to winners of two competitions.

NYT: Surgery to prevent strokes is found ineffective

Their failure highlights the peril of assuming that an apparent improvement on a lab test or X-ray, like better blood flow or a wider artery, will translate into something that actually helps patients, warned an editorial that accompanied the new findings. Only rigorous studies can tell for sure. … “Surgeons don’t want to be doing bad operations,” Dr. Langer said. “Whenever you have a paper like this, we’re all disappointed, because we like to operate. But in the end it’s a good thing.”

http://www.nytimes.com/2011/11/09/health/research/surgery-to-prevent-strokes-is-found-ineffective.html?_r=0

UPS Trucks Don't Turn Left, Saving Them 10 Million Gallons of Gas/Year

Hat tip Mike Z

http://vaviper.blogspot.com/2014/04/ups-trucks-dont-turn-left-saving-them.html

HBR: Decreased Congestion in Beijing Undermines Air-Pollution Efforts

Beijing’s recent attempt to reduce air pollution by requiring drivers to take their cars off the road one day a week has had the effect of reducing congestion and raising speeds, which has led to increases in emissions of particulates, according to a team led by Cong Sun of Tsinghua University in China. As a consequence, there has been little or even a negative impact on air quality, which may help explain why Beijing has faced increasingly serious air pollution since imposing the one-day-a-week driving restrictions, the researchers say.

SOURCE: Restricting driving for better traffic and clearer skies: Did it work in Beijing?

April 29, 2014

Ken Rudin's talk

Lots of direct advice on how to do data science right.

https://www.youtube.com/watch?v=xKrw2TKfj4w

(Hat Tip Sam Dru)

klsdjflksdfj

Machine Learning Course

Keep in mind there are ML algorithms and also data preparation and feature selection.

https://www.coursera.org/course/ml

Statistical Learning course by Hastie and Tibshirani

Big Data: intoxicating

'Big data' has this intoxicating effect. We start collecting it out of fear, but then it seduces us into thinking that it will give us power. In the end, it's just a mirror, reflecting whatever assumptions we approach it with.

http://idlewords.com/bt14.htm

Pointing out the bleedin’ obvious

From Black Triangle:

Is not such a bad thing says The Institute of Safe Medical Practice: In discussing the dilemma between evidence-based medicine and obvious patient safety strategies, Leape, Berwick, and Bates noted that, "A common theme… is that they make sense. To a lay person and to most physicians these things sound like obvious things to do. That is basically how aviation and anesthesia made progress: they did what seemed to be the obvious right thing to do."

http://www.blacktriangle.org/blog/?p=77

Trending Blogs

Recently Viewed Blogs

Data Dispatches