Do US Veterans Have Increased Risk for Alcohol Abuse?

In 2007, I took a Biostatistics course (taught by a wonderful lecturer, Dr. Melody Goodman) to brush up on my statistics. The final project was to do secondary data analysis of some national, large dataset. 

I started with the "common knowledge" that US veterans are more likely the nonveterans to have drinking problems. I tested this hypothesis by analyzing responses to the 2005 Behavioral Risk Factor Surveillance System (BRFSS) survey. Initially, I was seduced into thinking that I would get definitive answers because the study interviewed >350,000 people (>900 Mb of data). 

I ended up being dismayed because the various interpretations that could be applied to the results precluded a definitive answer. 

But, I was also delighted to uncover a relationship between age and veteran drinking that had been overlooked. The full paper can be downloaded here.

The BRFSS survey distinguishes two types of problem drinking:


- Binge drinking = having five or more drinks on a particular occasion in the past 30 days. 

- Heavy drinking = engaging in Binge drinking five or more times in the past 30 days.  

Veterans were identified by the answer to this question:

- Have you ever served on active duty in the United States Armed Forces, either in the regular military or in a National Guard or military reserve unit?

Note that "active duty" does not imply that the person ever saw battle!

One more important thing about the BRFSS. They conducted telephone interviews with the >350,000 participants. But, only with people who have landlines. If you used a cell phone, you were not included. If you did not have a phone or even a home, you were not included.

A published result from an earlier study (ref 2 in paper) suggested that veterans were 1.06 times more likely than non-veterans to engage in binge drinking and 1.17 times more likely for heavy drinking. The binge drinking number was considered to be "not significant" because the 95% confidence interval included 1.0 (an "odds ratio" of 1.0 means that there is no difference between groups). The heavy drinking number was significant.

My analysis found similar results. Binge drinking was 1.11 times more prevalent in vets than non-vets. Heavy drinking was 1.32 times more prevalent. Both numbers were significant using the 95% confidence interval standard.

However, my analysis also considered the age of individuals. At right are the results for binge drinking. 

The graph shows that young vets (20-40 yrs) are more likely than young non-vets to binge drink. 

But, older vets (60-80 yrs) are less likely than older non-vets to binge drink.

I found similar results for heavy drinking (see paper).


Now, the fun part! What does it mean? Why do young vets binge drink but old vets do not (compared to non-vets)?

- Maybe alcohol-abusing vets are more likely to die prematurely (or become homeless - remember the survey technique) than non-vets. 

- Maybe vets, as they get older, are more likely (than non-vets) to seek and receive help to deal with their alcoholism. 

- Maybe there is a cohort effect. Recent vets are more likely to survive after suffering injuries that would have killed soldiers from previous wars. These survivors have severe handicaps (e.g. loss of limbs) that could lead to depression and to alcohol abuse. 

- Maybe young vets exaggerate their alcohol use to appear "macho" to someone in the room during the telephone interview. Young non-vets may do this too.

- Maybe old vets downplay their alcohol use to appear "in compliance" with someone in the room during the telephone interview. Old non-vets may be guilty as well.

- Maybe "front line" vets should be distinguished from other active duty vets.

This is the problem with many surveys. You get a ton of data to analyze, you find an effect, but you cannot really identify the underlying cause of the effect. Add to that the tendency of people to stretch the truth. But that doesn't stop people from publishing or publicizing their findings. And, maybe setting policies according to the findings.

I know I wouldn't be able to do this for a living! I'll stick to analyzing molecules.

Oddity: French speed limits

Kate and I drove for 1 week in the Normandy region of France. I noticed a curious thing about the speed limits in France. Of course, they are in km/hr.

30
50
70
90
110
130





So, speed limit = 10(1+2n), n=1, 2, 3, 4, 5, 6

I don't remember such quantization when driving in Germany, about 15 years ago.

After France, we were in Finland for a week. We didn't drive there, but I watched the signs while being driven and speeds like 40 and 60 were often seen.

iSmile

02-01-2011

My iPod touch is just over a year old. I’ve found all sorts of uses for it. It serves as my photo album (to show off pictures of my family, our trips, and even my research). It’s my alarm clock (beats trying to figure out how to set the one at a hotel when traveling), my shopping list, my calendar (when my laptop isn’t nearby), my collection of NYT favorite articles, my recipe book (Evernote works well for this) and the list goes on. Oh, I also listen to music on it occasionally! But, that’s not what I want to tell you about. The most-used feature of my iPod is the Podcast. First, a confession: I am an NPR junkie! The only problem is that I’m not always near a radio when “my” shows are on. Ergo, the podcast has become my sure-fire NPR administration route.

If you see me with earbuds attached, walking to or from SBU (or biking during more moderate days; when was the last one of those?), most likely I’m listing to an NPR podcast. The Story of the Day (even if it’s a week old). David Bouchier’s essay, Car Talk, Wait Wait Don’t Tell Me, Will Shortz’ Sunday puzzle segment. I do stray outside of npr.org by including the Nature podcast.  An eclectic mix but there is a common theme. Most of the time, these podcasts make me smile. Of course, different smiles are involved. The gentle smile evoked by David Bouchier’s wry humor. Solving a Will Shortz word puzzle may well put a smirk on my face. The grin broadens when I hear a stupid crook story on Wait Wait (did you hear about the house burglar who left his charging cell phone behind?) And, I nearly guffaw at the antics of Click and Clack on Car Talk.

At first, I didn’t notice. After a while, though, I saw people smiling at me as I walked along. On Long Island?! But, I thought about it. Hey, I’m smiling. Maybe they notice this. Now, I pay more attention. It doesn’t seem to matter that my smile is triggered by the sound waves exclusively in my aural passages, people see me smile and respond in kind.

I hope you aren’t expecting some sort of save-the-world ending here. I’ve become too cynical for that! Nevertheless, I try to smile even when I’m detached from earbuds. If it lightens a small sliver of someone’s space-time continuum, that’s good enough for me. Besides, I’m too self-conscious to do what my friend Steve advocates: randomly high-fiving strangers! For now, I’ll stick with iSmile. uSmile?

Jellyfish, Biophysics and Surgery

03-03-2010

I just returned from the annual meeting of the Biophysical Society in rainy San Francisco. The Society is the first scientific society I joined - 34 years ago! Oh, you didn’t know that they accepted high school students as members? - heh heh heh.

The meeting was filled with the usual things that biophysicists think about: protein structure and function, lipid membranes, molecular modeling … all that neat stuff. I presented a poster on a strange phenomenon Ewa Nurowska and I observed about the chloride conductance in glycine receptor channels. Not to worry, I’m not going to describe that here.

What I really want to talk about is surgery! Yes, surgery had a prominent place at a meeting of biophysicists. The occasion was the National Lecture, presented by Roger Tsien - one of a trio who shared the Nobel Prize for Chemistry in 2008. You may not recognize Roger’s name, but you probably have seen his work in one form or another. If you have ever seen an image of a cell glowing green - that’s due to Roger’s discovery and the development of green fluorescent protein. GFP was isolated from a type of jellyfish. Roger’s group modified the GFP gene to make it more efficient. He subsequently developed variants of GFP that absorb ultraviolet light and then fluoresce at visible wavelengths that cover the whole rainbow. The colorful pictures allow researchers to determine, for example, what cells are expressing what proteins. It has become an indispensable tool in many fields of biological research.

Roger went on to describe how he is now developing probes for use in MRI and electron microscopy, but also that he wanted to determine whether optical dyes could have medical application. The way to get GFP into cells, however, is to incorporate it into the genome - and that’s neither practical nor ethical with humans. Also, GFP is fine for single cells but the emitted light would be absorbed by the tissue and fat of an intact person.

To overcome these problems, he started with one of his red dyes, Cy5, (long wavelengths penetrate tissue better) and linked it to a carrier polypeptide. The carrier molecule is positively charged and will stick to cell membranes. But, it becomes sticky only after being partially degraded by a protease that is excreted from many tumor cells. So, when the dye-polypeptide complex is externally applied to a surgical field containing normal and cancerous tissue, only the tumor cells are labeled. Now, you may ask, what good is a red dye in a bloody surgical field? Well, the dye emits a specific red wavelength, so a digitized image can highlight what the human eye cannot. The surgeon can view a false-color image of the tumor in the surgical field. Roger’s videos showed several examples with mice. The most impressive video was one in which two dyes were used. The surgeon observed a lime green tumor, cyan blue nerves (both false colors) and the usual surgical field. She excised the cancer, carefully avoiding the branch of the nerve that was surrounded by the tumor. Way cool!

You might have thought that advances in cancer treatment would be based on chemo or radiation therapy. But don’t discount the possibility of a more colorful operating room.

You can read more about Roger’s work in the March 2, 2010 issue of PNAS (Nguyen et al, "Surgery with molecular fluorescence imaging using activatable cell-penetrating peptides decreases residual cancer and improves survival", PNAS 107:4317).