Randomized trials versus observational studies

The primary criticism of observational studies is that there is no way to know the extent of unmeasured confounding.

Randomized controlled trials (RCTs) have their own limitations.  They often exclude patients with co-morbid conditions and select the most adherent patients using a pre-randomization run-in phase.

However, there is another problem with RCTs, one that is not widely recognized.  Quoting myself in a forthcoming paper (link to abstract (email me for reprint)):

In RCTs patients have uncertainty about what treatment they are receiving. A patient receiving an active drug or therapy might falsely believe that they are receiving the placebo or sham therapy. Outside of the RCT environment, a patient who is prescribed a drug by their physician will be sure that they are receiving the active drug. We would expect placebo effects to be stronger if patients were unaware that they might be given a placebo. Similarly, we might expect active treatments to be more effective if there was no uncertainty about treatment receipt. While there has been great emphasis about the importance of concealing treatment assignment, this concealment creates uncertainty within the patient about treatment assignment.

Treatment uncertainty could affect subjects’ behavior (such as adherence) and subjective well being.  Given the evidence about placebo effects, it’s not unreasonable to speculate that these uncertainty effects could be substantial.  Further, treatment uncertainty also might also affect who is willing to participate in the studies.  For example, patients’ who want the newest therapy might be unwilling to risk getting randomized to  placebo.

In the next post, I will formalize these ideas.  In the final post of this series, I will propose a solution.

Read Full Post »

Racial resentment

Political science professor Alan Abramowitz of Emory University wrote a paper which, among other things, claims that ‘racial hostility’ is a significant predictor of Tea Party support.   His conclusion was based on results from a survey.  The information on attitudes about race came from a  four-item ‘racial resentment’ scale.    Here is the first item that makes up the racial resentment scale:

Do you agree strongly, agree somewhat, neither agree nor disagree, disagree somewhat, or disagree strongly with this statement?

Irish, Italians, Jewish and many other minorities overcame prejudice and worked their way up. Blacks should do the same without any special favors.

How can one answer something like this?  What if you reject the premise of the statement?  For example, the two  sentences imply that many minority groups have overcome prejudice without any special favors (whatever that means).  It also implies that blacks have not worked their way up so far, with or without special favors.  What if you think that Italians and Jews worked their way up with special favors?  What if you think that Irish have not worked their way up?  What if you think that blacks already have worked their way up (they’ve come along way since the days of slavery)?  Plus, there is a possible false equivalence here.  Not all obstacles are of the same size.  If blacks ‘work their way up’ without ‘special favors,’ that does not mean that that is ‘the same’ as what other minority groups accomplished.

Special favors

The phrase  ‘special favors’ makes it sound like blacks would be getting some extra goodies from the government that is only available to them (it’s ‘special’).  Well, a lot of people with very little racial resentment would object to any minority group getting special favors.  That’s especially true for people who prefer small government (like Tea Party folks).  As Bob Somerby pointed out, there is no shortage of blacks who would agree that blacks should not get special favors (apparently resenting themselves).

Racial resentment=resentment of blacks

All four items making up the racial resentment scale have to do with resentment of blacks.  For example, the fourth item is:

Do you agree strongly, agree somewhat, neither agree nor disagree, disagree somewhat, or disagree strongly with this statement?

It’s really a matter of some people not trying hard enough; if blacks would only try harder they could be just as well off as whites.

I suppose it’s not surprising that the Tea Party looked like they had a lot of racial resentment when you measure it in this way (since people who identify with them tend to be white conservatives).

What if some of the questions tried measure resentment of whites?  For example:

Do you agree strongly, agree somewhat, neither agree nor disagree, disagree somewhat, or disagree strongly with this statement?

Part of the reason that income inequality in this country is so extreme is because there is still an exclusive club among rich white people.

Based on responses to that question, perhaps Democrats would look like they have a lot of ‘racial resentment.’

I’m not saying that racism is not more prevalent among people who identify with the Tea Party, but let’s do a better job of measuring these things.

Read Full Post »

sunk costs — when to count them

Whether or not you should weigh past expenditures in your judgment depends on the question you want to answer.

Sometimes you should ignore sunk costs, like when deciding whether to continue playing Farmville (link):

The urge to stay the course and keep your farm flourishing gets more powerful the more you invest in it, the more you ask others for help, the more time you spend thinking about it. People set alarms to wake up in the middle of the night to keep their farm alive. You continue to play Farmville not to have fun, but to avoid negative emotions. It isn’t the crop you are harvesting, but your fallacies. You return and click to patch cracks in a dam holding back something icky in your mind – the sense you wasted something you can never get back.

…

Farmville players are mired in a pit of sunk costs. They can never get back the time or the money they’ve spent, but they keep playing to avoid feeling the pain of loss and the ugly sensation waste creates.

However, when assessing whether effort was ‘worth it’, you should take into account the pain and suffering (or benefit — sunk benefit is a concept too) that you have already experienced.  For example, suppose you slaved away at a job you hated for decades and were miserable the whole time, but now you are enjoying a nice retirement.  It would be easy to feel like it was worth it now, because you are no longer suffering.   But if you are advising a young person about whether they should follow your path, you need to take that cost into account.

Similarly, if you are an adult and thinking about whether you are glad you were brought into the world (and whether we should liberally add people to the world), it’s easy to discount the suffering you might have experienced as a teenager (since it’s in the past).

My impression is that people weigh past cost more heavily in the former scenario (farmville) than in the latter scenarios.

Read Full Post »

Song taste persists, sort of

My musical tastes have changed a lot since I was less than 2.5 decades old.  I sometimes hear songs that I know I probably would have liked in high school, but do not like now. Many of the songs that I like today I know I would not have liked as a teenager.

However, the songs that I listened to a lot in the past, I still enjoy now, even when they are the type of songs that I dislike now.    I don’t think this is nostalgia — something else is going on.

It’s not that way with movies.  I currently hate a lot of the movies that I loved as a teenager.

It’s not that way with food.  I’ve been very disappointed whenever I’ve eaten a lot of the food that I loved as a kid.

It’s also not that way with books.

Read Full Post »