Tag Archives: evidence

The occasional note about research, journalism and public policy.

Oh, it must be Wednesday.

Posted on by

Which means work.

At present I’m trying to work out how to do a meta analysis of the correlations or odds ratios in a nested confirmatory factor analysis. Probably not needed, as there are descriptive differences and I simply cannot compare apples with gorillas. Problem is there are two papers with almost identical results. But not quite.

WIll need to look at more than OVID for this — there were only 17 hits with the first search so I’ll change words and look again (I was hitting multiple databases at the same time though, so only five papers that looked useful). May not be much else

In the meantime, CRAN let me find this:

Authors: Ken Kelley
(2445)Confidence Intervals for Standardized Effect Sizes: Theory, Application, and Implementation
Reference: Vol. 20, Issue 8, Feb 2007
Submitted 2006-10-01, Accepted 2007-07-30
Type: Article
Abstract:

The behavioral, educational, and social sciences are undergoing a paradigmatic shift in methodology, from disciplines that focus on the dichotomous outcome of null hypothesis significance tests to disciplines that report and interpret effect sizes and their corresponding confidence intervals. Due to the arbitrariness of many measurement instruments used in the behavioral, educational, and social sciences, some of the most widely reported effect sizes are standardized. Although forming confidence intervals for standardized effect sizes can be very beneficial, such confidence interval procedures are generally difficult to implement because they depend on noncentral t, F, and x2 distributions. At present, no main-stream statistical package provides exact confidence intervals for standardized effects without the use of specialized programming scripts. Methods for the Behavioral, Educational, and Social Sciences (MBESS) is an R package that has routines for calculating confidence intervals for noncentral t, F, and x2 distributions, which are then used in the calculation of exact confidence intervals for standardized effect sizes by using the confidence interval transformation and inversion principles. The present article discusses the way in which confidence intervals are formed for standardized effect sizes and illustrates how such confidence intervals can be easily formed using MBESS in R.

I am such a stats geek that this is my bedtime reading. Pathetic.

Trainwreck continues

Posted on by

The owner of theBuffalo Sabres pro football team. did the math.

And moved.

As I said yesterday, the US is a natural experiment that has proved, in the last decade, that Peronist economics does not work.

Politicians like to talk about incentives — for businesses to relocate, for example, or to get folks to buy local. After reviewing the new budget, I have identified the most compelling incentive of all: a major tax break immedi ately available to all New Yorkers. To be eligible, you need do only one thing: move out of New York state.

Last week I spent 90 minutes doing a couple of simple things — registering to vote, changing my driver’s license, filling out a domicile certificate and signing a homestead certificate — in Florida. Combined with spending 184 days a year outside New York, these simple procedures will save me over $5 million in New York taxes annually.

By moving to Florida, I can spend that $5 million on worthy causes, like better hospitals, improving education or the Clinton Global Initiative. Or maybe I’ll continue to invest it in fighting the status quo in Albany. One thing’s certain: That money won’t continue to fund Albany’s bloated bureaucracy, corrupt politicians and regular special-interest handouts.

via ADIOS, NEW YORK – New York Post.

The scientific method in action.

Posted on by

California has been run by Peronists for a while. The government works for the unions and the legislators. There is an effluvium of populism and sloganeering.

However, the budget is bloated. Taxes are soooo high that firms and people are leaving. This demonstrates,  almost perfectly, that tax and spend policies are economically disastrous.

It is a perfect experiment: 300 people who can move through 50 states with no restrictions: firms that can also move in a similar manner. The consequence is that Calif. is going to tank. Or be rescued, which will mean that the federal government will have more toxic debt that it can swallow.

I agree with Megan McArdle…

So what about California? A reader asks. Ummm, that’s a tough one. No, wait, it’s not: California is completely, totally, irreparably hosed. And not a little garden hose. More like this. Their outflow is bigger than their inflow. You can blame Republicans who won’t pass a budget, or Democrats who spend every single cent of tax money that comes in during the booms, borrow some more, and then act all surprised when revenues, in a totally unprecedented, inexplicable, and unforeseaable chain of events, fall during a recession. You can blame the initiative process, and the uneducated voters who try to vote themselves rich by picking their own pockets. Whoever is to blame, the state was bound to go broke one day, and hey, today’s that day!

There is a surprisingly sizeable blogger contingent arguing that we have to bail them out because however regrettable the events that lead here, we now have no choice. But actually, we do have a choice: we could let them go bankrupt. And we probably should.

I am not under the illusion that this will be fun. For starters, the rest of you sitting smugly out there in your snug homes, preparing to enjoy the spectacle, should prepare to enjoy the higher taxes you’re going to pay as a result. Your states and municipalities will pay higher interest on their bonds if California is allowed to default. Also, the default is going to result in a great deal of personal misery, more than a little of which is going to end up on the books of Federal unemployment insurance and other such programs.

Then there are the actual people involved. Whatever you think of, say, children who decided to be born poor, right now they are dependent on government programs, and will be put in danger if those programs are interrupted.

On the other hand, I don’t really see another way out of it. If Uncle Sugar bails out California, California will not fix its problems

via Is California Too Big to Fail? – Megan McArdle.

But there is one fly in the ointment. The Obamaborg owes Pelosi. Pelosi was infected with Peronism a long, long time ago. This could be the beginning of the USA slide towards being, like Argentina, a banana republic.

NHS trust admit catalogue of errors let schizophrenic kill pregnant woman – Telegraph

Posted on by

There but for the grace go all of us. One of the problems is that the ideolology around mental healht is about recovery. This was decided at the highest levels…

Holiday handed himself in to the care of a secure unit, but the following month a social worker let him back out into the community. This was a “missed opportunity” to have him sectioned, according to the report, which was written by the region’s strategic health authority, NHS Yorkshire and the Humber.

And the decision should not be made by one person. It should be made by a team and there should be a senior psychiatrist signing off on it.

Holiday missed his fortnightly dose of anti-psychotic drugs the day before the killing because he had not been home when a nurse called. He went on to steal a 10-inch carving knife from a store and was captured loitering in the street, as if waiting for someone to pass by.

One of the main reasons for community treatment orders is so that this does not happen. In high risk people a protocol (legal in NZ, but the UK is full of idiots making the life of both the mad and mad doctors miserable) is to admit to hospital briefly: we can and do use the police to bring people back for their injections.

If you miss your meds you can relapse. About 90% of people with schizophrenia who stop their meds will relapse. And being psychotic is very bad for your ability to cope in most people. In a few, very few, it leads to them being quite dangerous, which is why compulsory treatment exists.

He turned off into Wellsted Street, where his path crossed with Miss Stevenson, who had just left her home.

Holiday walked past her, turned and stabbed her once in the back, piercing her heart and leaving her bleeding to death. He then walked home calmly. Police tracked him down from CCTV images and Holiday admitted manslaughter on the grounds of diminished responsibility. He was ordered to be detained indefinitely at Rampton Hospital.

The report found that Holiday had been treated in at least five mental health units and was sectioned under the Mental Health Act on at least two occasions, but had a habit of escaping.

Known to be high risk. It would be interesting to see if there had been a review panel before all this.

Mr Snowdon apologised to the family of both Miss Stevenson and Holiday for the distress caused to them. He said the trust took the report “very seriously” and lessons had “already been learned”.

The NHS trust and clinicians should have acknowledged distress and that this was a tragedy from the beginning. That is not an admission before a court of law: that is being human and good care.

via NHS trust admit catalogue of errors let schizophrenic kill pregnant woman – Telegraph.

Individuals against the Borg.

Posted on by

Interesting day. Found this via No Minister. My comment is that it is moral to reward individuals who risk: it is immoral to steal. The most moral tax is atithe (which goes to the poor) or a  head tax (interestingly, the Torah has a head tax to start the covenant, but tithes to keep welfare going.)

It wrong to subsume one’s God-given moral thoughts and freedom to a collective. It is moral to oppose the Borg.

Advocates of free enterprise must learn from the growing grass-roots protests, and make the moral case for freedom and entrepreneurship. They have to declare that it is a moral issue to confiscate more income from the minority simply because the government can. It’s also a moral issue to lower the rewards for entrepreneurial success, and to spend what we don’t have without regard for our children’s future.

Enterprise defenders also have to define “fairness” as protecting merit and freedom. This is more intuitively appealing to Americans than anything involving forced redistribution. Take public attitudes toward the estate tax, which only a few (who leave estates in the millions of dollars) will ever pay, but which two-thirds of Americans believe is “not fair at all,” according to a 2009 Harris poll. Millions of ordinary citizens believe it is unfair for the government to be predatory — even if the prey are wealthy.

via The Real Culture War Is Over Capitalism – WSJ.com.

It’s getting … colder

Posted on by

But it is not getting through becaus it is not the approved view, y’know.

Garrett, by the way, was a talented musician and a political idiot. He is now just a political idiot.

Australian Antarctic Division glaciology program head Ian Allison said sea ice losses in west Antarctica over the past 30 years had been more than offset by increases in the Ross Sea region, just one sector of east Antarctica.

“Sea ice conditions have remained stable in Antarctica generally,” Dr Allison said.

The melting of sea ice — fast ice and pack ice — does not cause sea levels to rise because the ice is in the water. Sea levels may rise with losses from freshwater ice sheets on the polar caps. In Antarctica, these losses are in the form of icebergs calved from ice shelves formed by glacial movements on the mainland.

Last week, federal Environment Minister Peter Garrett said experts predicted sea level rises of up to 6m from Antarctic melting by 2100, but the worst case scenario foreshadowed by the SCAR report was a 1.25m rise.

Mr Garrett insisted global warming was causing ice losses throughout Antarctica. “I don’t think there’s any doubt it is contributing to what we’ve seen both on the Wilkins shelf and more generally in Antarctica,” he said.

Dr Allison said there was not any evidence of significant change in the mass of ice shelves in east Antarctica nor any indication that its ice cap was melting. “The only significant calvings in Antarctica have been in the west,” he said. And he cautioned that calvings of the magnitude seen recently in west Antarctica might not be unusual.

“Ice shelves in general have episodic carvings and there can be large icebergs breaking off — I’m talking 100km or 200km long — every 10 or 20 or 50 years.”

via Revealed: Antarctic ice growing, not shrinking | The Australian.

Naming and Shaming: the consequences

Posted on by

The good macdoctor has discussed the Psychwatch site. Quoting my comment:

Local patient advocate showed me a printout of the site today.

Some issues.

1. All doctors make some mistakes every year, and we all pray that no one gets hurt. But they happen. We have systems (peer review, audit, double checking of important information) to minimise them.

2. In psychiatry we are often trying to make a decision where there is little data to guide us… and where the research is changing rapidly.

3. Most doctors who have had complaints against them take them seriously. Complaints generally make docs less confident, more scared or defensive (which is generally a bad thing) and much more mistrustful of patients (which is a very bad thing).

4. A number of patients make repeated complaints to authorities. Psychiatry and GP collect these kind of patients.

5. There is a duty on all doctors to protect the public, which can include asking colleagues, ones employer, and at times the medical council for backup in getting that doctor back to competence.

6. If one still believes that these sites are worthwhile, I suggest you look at NHS Blog doctor and the hell many GPs there have gone through due to anonymous complaints from patients.

The Health and Disability Commissioner has done, in my view, managed to hold the balance between identifying issues that need correction and avoiding “naming and shaming”.

A return to that state would lead, again, to all health professionals being silent about poor practice out of fear of the inquisitors: which is what Mr Taylor will get if this type of site remains, whether he wants it or not.

The author of Psychwatch argued back that he is doing a public service.

Macdoctor:

You demonstrate a rather tiring yet quite familiar arrogance of the medical field – you believe that a medical degree and the authority to prescribe somehow make you more “right” than anyone else. I mop after you pricks all of the time – your profession is often allergic to correction, blind in your professional hubris, and have a seeming inability to understand any client intervention outside the scientific realm. I understand ethics, as my reference point is codes of ethics, and thus I can quite easily discern between ethical vs unethical behaviour; I can discern between competent and incompetent care, as my reference source is best practice protocols, protocols your own industy creates; if you truly believe that you are unqualified to determine either of these professional markers in your colleagues, then that either makes you an apologist for incompetence or a coward.

You choose.

I notice you have nothing to say about the pathology of depression? Glasser = 1; Medical Fraternity = 0.

I do not the arrogance leaking from this comment… but Psychwatch is wrong…

1.  These things (complaints) can literally drive doctors to suicide. (Dr Rant is talking about the UK General Medical Council, and one hopes that the kiwi medical council is more merciful.) However, bullying and patient intiated aggression, including vexatious complaints, exist in New Zealand. I will add that the findings around patient complaints have been found repetatively, most recently in (untrained) caregivers for NGOs: (presenting a poster on this recently at TMHS in Auckland, I was buttonholed by a consumrer who said it was all the caregivers fault).

The great Dr Crippen was campaigning earlier this year about a similar site in the UK. He notes:

My greatest worry about this website is the effect that it might have on a doctor suffering from depression. Do not get me wrong. If you think your doctor is mentally ill, or performing dangerously, not only is it your right to make a complaint, it is your duty so to do. You must take immediate action. But not by making anonymous remarks on the internet. Talk to one of the other doctors in the practice or to the chief executive of the hospital. But do not put some anonymous, wounding criticisms on an internet website. Supposing you are the person who made the above comment. Supposing this doctor is depressed and therefore not performing well. Supposing your anonymous comment tips him over the edge. You read in the paper next week that he has committed suicide. How would you feel then? Do you remember this:

TRURO An out-of-hours GP killed himself because of fears that he might be dismissed after turning up in the wrong town for an emergency call. (full story here)

Who is Dr Neil Bacon to set himself up in judgement over the medical profession? He has no legal training. He has not completed his higher medical training. Whatever he may say, it is hard to believe that his motives are altruistic. This website is a disgrace.

2.  Naming and blaming implies that the doctor is disposable. Doctors, nurses and other workers do not hang around when they are unwanted. At the moment New Zealand is benefiting from this: the NHS has made many junior doctors feel unwelcome. We can do the same thing: and if that happens people can leave, will leave and are leaving. It is disheartening, after encouraging a junior doctor into training, coaching them through the examinations, and celebrating their being qualified, to find tha the management do not have a job for them, or that Townsville or similar has given them an offer that is much better than what is available locally. All specailist qualifications in NZ are binational & I get a letter a month from Australia offering me salary packages about half as much as what I make in NZ>

3.  In the end, the science around talking therapies has been established. Firstly, for most anxiety disorders, behavioural or cogntive talking therapies are the treatment of choice. (The correct links to look at are Best Treatments, or the underlying data in Clinical Evidence or the Cochrane Database)  For mood disorders, medication, interpersonal therapy (a short form of psychodynamic therapy) or cognitve therapies are the best available treatments. For the psychoses, medication with family therapy or cognitive therapy for psychosis seem to have best evidence…

I find people who criticise doctors for “not acknowledging science” odd. All treatments need to be shown to be safe and effective — we should nof offer unsafe and useless treatment options! Much of the research we quote on talking therapies comes from trials of treatments developed by psychologists, nurses and psychotherapists. Psychwatch has set up a straw man.