Keith Warren
A researcher at Ohio State University used techniques of chaos theory to do a case study of the drinking patterns of an alcohol abuser over several years.
The results, while very preliminary, may offer new insights into patterns of alcohol abuse and how doctors and therapists can best provide help.
For example, the results suggest that alcohol abusers may show signs of improvement that are really just part of the natural cycle of their disease and not indicative of true recovery.
Chaos theory is applicable to alcohol abuse because it can help us understand sudden changes in the behavior of a system, unpredictability, and irregular behavior, all of which occur in alcoholism,” said Keith Warren, co-author of the research and assistant professor of social work at Ohio State.
--------------------------------------------------------------------------------
The results also suggest that most treatments for alcoholism may not last long enough to make a real positive impact, according to Warren.
--------------------------------------------------------------------------------
Chaos theory looks for simple models of complex behavior. The tools for building such models are nonlinear equations, which can show complex and surprising behavior such as rapid shifts from stability to an oscillating state.
Warren said alcoholism can be seen as such a complex system.
“There’s been an assumption that alcoholism is a relatively stable disease, but that may not be correct. The disease is very dynamic, it changes a lot over time, and chaos theory may be able to help us understand it better.”
Warren’s most recent study appears in the current issue of the journal Addictive Behaviors. Another related study appeared last year in the journal Psychological Reports.
In both studies, Warren and his co-authors studied a record of daily alcohol intake kept by a 40-year-old male alcohol abuser and binge drinker over a period of about five years.
They analyzed the data and created various time series – including daily, weekly and monthly – to determine patterns in his drinking.
The results showed that any single day’s intake of alcohol influenced how much he drank the next day, which is suggestive of a dynamic system. It also reflects the truism that, for alcoholics, one drink is likely to lead to another.
Moreover, there was an oscillating pattern to the subject’s abuse, Warren said. When he felt he drank too much one day, he would often respond by trying to drink less the next day. But the key is that he wouldn’t reduce his consumption as much as he increased it the previous day.
“When he drinks too much he realizes it and tries to control it by drinking less the next day – but he never quite pulls his drinking back enough,” Warren said. “You end up with a very bumpy relapse into drinking, one that shows one step forward and two steps back.”
At least for this person, a relapse didn’t usually start with a drinking binge, Warren said. It would be a very oscillating cycle of increases in drinking followed by smaller decreases followed by larger increases.
These cycles of alcohol consumption could last varying lengths of time, he said.
“This could be very deceptive to someone treating a person for alcoholism,” Warren said. “An alcohol abuser may be showing improvement for a while, but it is really just part of their cycle of relapse. By definition, if alcohol intake tends to oscillate, than an abuser is likely to be ‘improving’ half the time. But obviously, it is not real progress.”
The results also suggest that most treatments for alcoholism may not last long enough to make a real positive impact, according to Warren. For alcoholics like the one they studied, there may be substantial periods where they don’t drink at all, but that may be part of their natural cycle.
“If a person thrives in a month-long alcoholism treatment program, it may simply mean that his intake of alcohol had not yet had a chance to rebound following a decrease,” he said. “This may be one important reason why alcoholics so often relapse after treatment.”
The results of these studies, if replicated and attempted with larger groups of patients, may also have implications for a controversy in the alcohol treatment community. Some doctors have argued that teaching alcoholics to drink in moderation may be more successful than promoting abstinence from alcohol.
But in the Addictive Behaviors paper, Warren and his colleagues argue that moderate drinking will be difficult to maintain.
“If an alcohol abuser could keep his drinking absolutely stable, he might be successful,” Warren said. “But abusers face a lot of pressures that are likely to encourage them to drink a little more. Our results suggest that it’s likely that any increase will be followed by a decrease that doesn’t quite compensate for the initial increase. After a few of these cycles -- two steps into relapse and one step back -- drinking could easily be out of control.”
Warren said these results are obviously very preliminary because they are a case study of a single person. One problem for researchers in this field is that when you do studies involving nonlinear dynamics – such as this case study – you can only study one single person at a time, since combining different nonlinear time series will result in a data set that will be indistinguishable from random noise. Still, he said it is important to do more of these single-subject time series to see if the results are generalizable.
“However, even an exploratory study such as ours may yield some insight into the dynamics of substance abuse,” Warren said.
Warren conducted the research with Raymond Hawkins, a psychology lecturer at the University of Texas, Austin, and Julien Sprott, a professor of physics at the University of Wisconsin, Madison.
Written by Jeff Grabmeier, (614) 292-8457; Grabmeier.1@osu.edu
Journal
Addictive Behaviors