Pre‐resolution drinking problem severity profiles associated with stable moderation outcomes of national recovery attempts

Latent profiles of pre-resolution drinking problem severity.


Background. Higher problem severity contraindicates moderation drinking in treatment samples, but has not been well investigated in natural recovery samples with more prevalent moderation outcomes, nor have single studies assessed multiple severity indicators. Therefore, we integrated five prospective studies of recent natural recovery attempts to identify multi‐indicator profiles that distinguished moderation from abstinence or unstable resolution involving relapse. The study evaluated whether moderation was distinguished by a generalized lower severity profile, or whether more complex profiles better differentiated outcomes. Methods. Community‐dwelling problem drinkers in the Southeastern U.S. (N = 616, 67% male, 65% white, mean age = 46.5 years) enrolled soon after stopping alcohol misuse without treatment were followed prospectively for a year. Outcome predictors assessed at enrollment included pre‐resolution drinking practices, alcohol‐related problems, alcohol dependence, and a behavioral economic measure of the reward value of drinking based on pre‐resolution spending on alcohol vs. saving for the future. Results. Latent profile analysis of severity indicators supported a four‐profile solution: (1) global low risk on all indicators, (2) global high risk on all indicators, (3) high risk limited to drinking practices only, and (4) high risk limited to alcohol dependence and alcohol‐related problems only. Outcomes differed by profile membership (p < .01). Multinomial logistic regression analyses showed that the global low risk and heavy drinking risk only profiles were associated with stable moderation during the one‐year follow‐up. The high dependence‐alcohol problems risk profile was associated with both abstinence and relapse during the follow‐up (ps < .05). Conclusions. Consistent with prior research, moderation was associated with lower alcohol dependence, problems, and reward value. Participants who simply drank heavily and did not have elevated risk on other indicators also had a higher probability of moderation. Results support using multi‐dimensional severity indicators that encompass functional variables in addition to drinking practices to predict outcomes.

Alcoholism: Clinical and Experimental Research

What is this paper about?

This paper uses data from five studies of people who have alcohol problems who are attempting to recover from alcohol problems without formal treatment. We use an advanced statistical technique to identify sub-groups of these people who are similar to each other before they resolved their goal.

How does this paper improve public health?

Researchers studying alcohol sometimes use drinking practices (such as the number of heavy drinking days) to identify the extent of a person’s drinking problems. Our study shows the importance of using multiple measures (including drinking practices, alcohol-related problems, dependence) when studying natural recovery outcomes.


  • Is this paper peer reviewed? This paper is published in the journal Alcoholism: Clinical and Expermental Research. This journal uses double-blind peer review, where the authors and the reviewers do not know each other’s identities.

  • Who paid for this project? This work was supported in part by the National Institutes of Health/National Institute on Alcohol Abuse and Alcoholism grants R01AA08972, R01AA017880, and R01AA023657. The National Institute of Alcohol Abuse and Alcoholism are government funded using tax payer dollars. The content of this paper is solely the responsibility of the authors and does not necessarily represent the official views of the funders.

  • Are there any conflicts of interest? There are no conflicts of interest.