Hacker News Reader

Comment by jdietrich | Hacker News Reader

Before everyone piles in with their opinion, has anyone actually found the research mentioned in this article? I've searched quite thoroughly, but I can't find any sign that it has actually been published anywhere. I did find the paper from 2004 and was not impressed.

http://www.publicartonline.org.uk/resources/research/documen...

View on HN

This [1] at least has an reference section, but most of the actual research listed is based on audio not visual.

I suspect that this article is an advertisement for Chelsea and Westminster Hospital’s new book, The Healing Arts which is available from Unicorn Publishing Group. I get the feeling any data, if available, is in the book.

1) http://www.lahf.org.uk/sites/default/files/Chelsea%20and%20W...

I spent a lot of time in Chelsea and Westminster and it's not a very nice hospital.

The burns unit was clean, bright and airy. But there were limited activity available because of infection control reasons, and the layout of the hospital was hostile to physical activity. Using the stairs was hard, all the signs and designs pushed you to the elevators. And those had been programmed incorrectly so they took an age to arrive and the to get you to your floor.

As far as built environments for burns units go it was better than Birmingham Shelly Oak (Victorian building, excellent care because military connection), better than Bristol Frenchay (awful care, I think they've knocked it down now), but not as good as Salisbury (better layout, good care).

This is a severe problem for the NHS: The amount of estate that's just terrible is surprising.

Indeed not, and I am EXTREMELY skeptical of the claim that birth times are reduced TWO HOURS (!!!) by the addition of art. That's the kind of effect you see in papers that get retracted for outright fraud, let alone dubious and irreproducible studies.

I think this is the paper on the birth time.

https://www.sciencedirect.com/science/article/pii/S003039921...

Unblinded (somewhat unavoidable), informal randomisation, n=26, the sole author is an interior designer with no scientific training. Paper states an "average" difference in birth times of 2.1 hours (not clear if mean or median), no SDs or p-value stated. Published in Optics & Laser Technology for some reason; the editorial board appears to be entirely comprised of engineers.

On the plus side, at the very least the same hospital and hospital rooms were used. I initially assumed they were comparing an expensive, spacious upper-class hospital full of art with a run-down brutalist Medicaid hospital, and conflating sampling bias with causality.

"A screen was designed to hide emergency equipment with the joint aim of reducing women's anxieties and (through visual art) acting as a focal point of attention and distraction during labour, thus diminishing requirements for analgesia. Results demonstrated, in the presence of the screen, a statistically significant shortening of the duration of labour by 2.1h"

Uh, I don't have access to the full article, but was the control hiding the equipment with a blank screen or no screen at all?

Cause if the control was no screen at all, I don't think they were testing what they say they were testing:

"The significant clinical outcomes of this research provide the evidence of the value of integrating visual art into the environment of a labour and delivery room, improving the quality of the maternity service and potentially delivering real cost savings benefits to Hospitals."

Agree... and I'd love to see another control, one that is just "pretty" but is not (considered) art...

Ok, but can you please not use allcaps for emphasis on HN? It's basically yelling. This is in the site guidelines:

https://news.ycombinator.com/newsguidelines.html.

Don't discount the placebo effect. Just because you can't see it, just because you can't measure it, don't mean something doesn't exist.

The placebo effect is scientifically recognized. It is real. Let's be mindful not to dismiss it - like it or not.

It does exist but a lot of times regression to the mean is wrongly identified as placebo effect. Its rampant in observational studies.

I tried, and could not. And I’m baseline a little skeptical because the NHS has a history of pushing very poorly designed fad research precisely like this.

Reply on news.ycombinator.com