Drinking to ease the burden: a cross-sectional study on trauma, alcohol abuse and psychopathology in a post-conflict context

Abstract

Background: It is likely that alcohol use and abuse increase during and after violent conflicts. The most prominent explanation of this phenomenon has been referred to as self-medication hypothesis. It predicts that psychotropic substances are consumed to deal with conflict-related psychic strains and trauma. In northern Uganda, a region that has been affected by a devastating civil war and is characterized by high levels of alcohol abuse we examined the associations between war-trauma, childhood maltreatment and problems related to alcohol use. Deducing from the self-medication hypothesis we assumed alcohol consumption moderates the relationship between trauma-exposure and psychopathology.

Methods: A cross-sectional epidemiological survey targeting war-affected families in post-conflict northern Uganda included data of male (n = 304) and female (n = 365) guardians. We used standardized questionnaires in an interview format to collect data on the guardians' socio-demography, trauma-exposure, alcohol consumption and symptoms of alcohol abuse, PTSD and depression.

Results: Symptoms of current alcohol use disorders were present in 46 % of the male and 1 % of the female respondents. A multiple regression model revealed the unique contributions of emotional abuse in the families of origin and trauma experienced outside the family-context in the prediction of men's alcohol-related symptoms. We found that alcohol consumption moderated the dose-effect relationship between trauma-exposure and symptoms of depression and PTSD. Significant interactions indicated that men who reported more alcohol-related problems experienced less increase in symptoms of PTSD and depression with increasing trauma-exposure.

Conclusions: The gradual attenuation of the dose-effect the more alcohol-related problems were reported is consistent with the self-medication hypothesis. Hence, the functionality of alcohol consumption has to be considered when designing and implementing addiction treatment in post-conflict contexts.

Keywords: Addiction; Alcohol; Conflict; Depression; Mental health; PTSD; Self-medication; Substance abuse; Trauma; War.

Fig. 1

Percentages of men and women categorized according to risk levels as proposed by the AUDIT manual and drinking typology. Drinking Typologies [103] (adapted): abstainer: never had a drink or had none in the past year; infrequent light drinker: drinking up to four times a month, always less than 5 standard drinks per occasion; frequent light drinker: drinking two or more times weekly and less than 5 standard drinks per occasion. Infrequent heavy drinker: drinking up to four times a month, sometimes 6 or more standard drinks per occasion. Frequent heavy drinker: drinking two or more times weekly and 5 or more standard drinks per occasion. We were not able to categorize 2 % (n = 7) of the males since they did not fit in any category. Six men were frequent light drinkers, but had binges of 6 or more standard drinks less than monthly or monthly. One male drank infrequently, but always 5 standard drinks per occasion, he never had binges of 6 or more standard drinks

Fig. 2

Symptoms of depression (a) and PTSD (b) as a function of low (Risk Level I), medium (Risk Level II) and high-risk (Risk Levels III and IV) drinking and trauma exposure. For instance, a married, non-abducted man from the community Agweno, who has not completed primary education, is currently abstinent and with his age, trauma-exposure and childhood maltreatment in the family of origin set to the respective sample means has an estimated depression-score of 1.53, which is still below the most frequently used cutoff score for clinically relevant depression of 1.75 [104]. A 10 % increase of reported traumatic event-types for this man is associated with a 5.4 % increase in the estimated depression-score. The same man currently presenting at the threshold of dependent drinking (AUDIT-sumscore = 20) would have a significantly lower estimated depression-symptom-score of 1.37. A 10 % increase of reported traumatic event-types for him would be associated with significantly less increase in the estimated depression-score of 2.8 %