Alcohol use disorder (AUD) is a disease characterized by the problematic use of alcohol. According to the Diagnostic and Statistical Manual of Mental Disorders (DSM-5), the disorder is diagnosed when two or more criteria are presented out of eleven that are evaluated, including the amount of alcohol ingested and the appearance of withdrawal symptoms when discontinuing consumption. AUD is considered severe if six or more diagnostic criteria are detected (APA, 2013).

The prevalence of AUD in the Western world is high, affecting 12% of the European adult population. In 2016, alcohol consumption was responsible for 3% and 8% of mortality in women and men, respectively (WHO, 2018).

The prevalence of alcohol consumption and amount has historically been higher in men than in women (Slade et al., 2016). However, these differences have been narrowing in recent years (Slade et al., 2016; White, 2020). The frequency and amount of consumption is increasing among women, while in men, these patterns remain stable (Grucza et al., 2018). In adolescents, the pattern of binge drinking has declined more rapidly in men than in women, causing sex differences to narrow and even reverse (Johnston et al., 2016).

In the older population, an increase in the prevalence of alcohol consumption and in the binge drinking pattern is described, especially among women (Han et al., 2017). However, men over 50 years of age show a higher prevalence of consumption and standard drink units per day, although with geographic variations among the 21 countries analyzed (Calvo et al., 2020).

There are sex-specific differences in the risk factors for developing AUD, as well as in the medical and psychiatric comorbidity of individuals who are dependent or abuse alcohol (Erol & Karpyak, 2015). For example, in women, anxiety disorder, childhood trauma and disturbances in mood or emotions (internalizing symptoms) have been described as risk factors (Guinle & Sinha, 2020; McHugh & Weiss, 2019). In men, developing AUD has been linked to novelty seeking, conduct disorder, parental loss and low self-esteem (Kendler et al., 2015).

In relation to the medical comorbidities associated with AUD, a greater vulnerability to the neurotoxic effects of ethanol, liver disease, cardiovascular disease, breast cancer, infertility and early menopause has been reported (Nolen-Hoeksema, 2004). Psychiatric comorbidities such as major depression, dysthymia and anxiety disorders occur more frequently in women, while in men, antisocial personality disorder is more frequent (Boykoff et al., 2010; Kendler et al., 2015; McHugh & Weiss, 2019; Nolen-Hoeksema, 2004). In addition, depressive symptoms seem to be associated with relapses in alcohol consumption, mainly in women (Schutte et al., 1997).

The increase in the frequency and quantity of alcohol consumption among women could be reflected in the increasing morbidity and mortality of this population. In this sense, an increase in the number of visits to the emergency room and in alcohol-related mortality has been described (White et al., 2018, 2020).

It has been suggested that sex differences should be considered in the design of specific therapeutic interventions for AUD (Greenfield et al., 2007). However, a sufficient number of patients is needed to establish differential aspects of the disorder and the severity of medical and psychiatric complications (Agabio et al., 2017).

The CohRTA multicenter study has as its main objective to characterize AUD in men and women in Spain as they begin their first treatment for the disorder (Sanvisens et al., 2021). We aimed to analyze sex differences in the context of sociodemographic, clinical and disease burden aspects to better understand different treatment interventions in this population.

Material and Methods

Participants and Methods

From January 2014 to March 2021, 896 patients who began their first treatment for AUD in six public centers of the Spanish health system affiliated with the CohRTA study were enrolled. The centers were Hospital Son Espases (Palma de Mallorca), Hospital Clinic (Barcelona), Hospital del Mar (Barcelona), Hospital 12 de Octubre (Madrid), Hospital Universitari Germans Trias i Pujol (Badalona) and Hospital Universitari de Bellvitge (Hospitalet de Llobregat).

The study was approved by the Clinical Research Ethics Committee (CEIC) of the coordinating center (Hospital Universitari Germans Trias i Pujol) and by the CEIC of each participating center. The patients signed an informed consent form that included the transfer of clinical data and biological samples. The methods used in this study have complied with the ethical standards for medical research, the principles of good clinical practice established by the Declaration of Helsinki, and Organic Law 3/2018 on Data Protection and Guarantee of Digital Rights (LOPD-GDR) and Law 14/2007 on Biomedical Research.

The diagnosis of AUD was made according to DSM-5 criteria (APA, 2013), and the patients followed the same clinical protocol in all participating centers. The CohRTA multicenter study has a database with 1) sociodemographic variables (i.e., age, sex, employment status, educational level, marital status); 2) alcohol consumption questionnaire: age at first alcohol exposure, age of onset of regular alcohol consumption (i.e. weekly drinking), duration of regular alcohol consumption (i.e. elapsed time between age of onset of regular alcohol consumption, and age at admission to treatment), amount of alcohol consumption (last 30 days) in Standard Drinks per day (SD/day), and number of alcohol intoxications (lifetime); 3) use of other substances (i.e., cocaine); 4) Cumulative Illness Rating Scale-Substance Abuse (CIRS-SA), which assesses the involvement of organs and systems and the severity of comorbidity; 5) psychiatric comorbidity; and 6) laboratory parameters with hemogram, biochemistry, viral serologies and screening for drugs of abuse in urine when entering the study.

The CIRS-SA scale is validated in the Spanish population (Castillo et al., 2004; Rivas et al., 2013) and analyzes comorbidities in the following organs/systems: (i) heart; (ii) vascular and hematopoietic; (iii) respiratory; (iv) eyes, ears, nose; (v) throat and larynx; (vi) gastrointestinal tract; (vii) liver, including hepatitis B and C viruses; (viii) renal; (ix) genitourinary; (x) musculoskeletal; (xi) neurological; (xii) endocrinology, breast and infectious diseases; and (xiii) HIV infection. Each category receives a score between 0 and 4 according to the severity of the illness, with 0 being the absence of comorbidity and 4 being a very severe manifestation. In addition, CIR-SA generates three indicators: (a) total score (range from 0 to 54), (b) number of organs or systems affected (range from 0 to 13) and (c) severity index (total score divided by the number of organs or systems affected).

Antecedent of psychiatric comorbidity might be established following a review of the medical records to obtain diagnoses from specialists in Psychiatry: major depression, psychosis, attention and hyperactivity disorder, posttraumatic stress disorder, generalized anxiety disorder, panic disorder, social phobia and mania.

Upon entry into the study, blood samples were obtained for hematological parameters (i.e., hemogram), liver function (i.e., aspartate aminotransferase (AST), alanine aminotransferase (ALT), gamma-glutamyl transferase (GGT) and acute phase reactants, such as erythrocyte sedimentation rate (ESR). Serologies were performed to detect antibodies against HIV, HCV and HBV. In addition, a urine drug screen was carried out.

Statistical analysis

Categorical variables are expressed as relative frequencies, and continuous variables are expressed as the median and interquartile range (IQR). Sex differences in categorical variables were determined using Fisher's exact test or the chi-square test. Mann‒Whitney U test was applied as a nonparametric test for continuous variables.

A logistic regression model was used to analyze sex differences, with men being the reference category. For the regression model, some variables with various categories were dichotomized (i.e., oropharynx and larynx). The multivariate analysis included those variables that were statistically significant in the univariate analysis; specifically, laboratory parameters were included in the model when statistically significant globally and pathologically. Results with a p value < 0.05 were considered statistically significant. Statistical analyses were performed with IBM SPSS Statistics version 22 (IBM, Armonk, NY, USA).

Results

A total of 896 patients (634 men) were included in the study. The age at the first treatment (median [IQR]: 48 years [41–56]) was similar in men and women. However, statistically significant differences (p < 0.001) were observed in work and family situations: women were more dedicated to household chores (5.9% vs. 0.2%), they had a lower prevalence of disability (13% vs. 19.7%) and they were divorced more frequently (36.8% vs. 30.6%), while men tended to be single (36.8% vs. 24%).

Regarding educational level, 21% of the men and 13% of the women had a primary school education; 56% of both genders had secondary education, and women had a higher frequency of university studies than men (25.3% vs. 16.3%, respectively) (p = 0.002).

Alcohol consumption

Table 1 shows the characteristics of alcohol consumption. Men started drinking, on average, one year earlier than women (median [IQR]: 15 [13–17] years vs. 16 [14–18] years; p = 0.002) and tended to consume alcohol regularly from an earlier age (median [IQR]: 20 [17–26] years vs. 26 [20–37] years; p < 0.001). In addition, men consumed a greater amount of alcohol than women (median [IQR]: 15 [10–24] SU/day vs. 12 [8–20] SU/day; p < 0.001).

Table 1 Characteristics of 634 men and 262 women requesting a first treatment for AUD in the CohRTA Study
Full size table

In addition, a family history of AUD was more frequent among women (p = 0.004), and in relation to the consumption of other substances, men showed a higher prevalence of cocaine use than women (p = 0.003).

Medical comorbidity

Table 2 shows the sex differences in baseline medical comorbidities. Overall, the most frequent organ or system involvement was the liver (73%), followed by respiratory (55%) and vascular (52%) systems. Compared with women, men showed a higher prevalence of severe or very severe medical comorbidities in most organs or systems, although statistically significant differences were observed only in oropharynx and larynx (p < 0.05). In addition, the severity index score of the CIRS-SA scale was higher for men (p < 0.05).

Table 2 Characteristics of comorbidity in 483 men and 213 women requesting a first treatment of AUD in CohRTA Study
Full size table

Psychiatric comorbidity

Table 3 shows the prevalence of psychiatric comorbidities. The most frequent disorders were major depression (MDD) (24.0%), generalized anxiety (6.1%) and psychotic disorder (2.5%). The prevalence of MDD was significantly higher in women than in men (46.7% vs. 22.0%, p < 0.001).

Table 3 Psychiatric comorbidity in 634 men and 262 women seeking a first AUD treatment in the CohRTA Study
Full size table

Anthropometric and blood parameters

The BMI and general laboratory parameters are described in Table 4. Men had a higher BMI than women (25.6 vs. 24.0 kg/m2, p < 0.001). Men also showed a higher median leukocyte count and hemoglobin level than women (p = 0.018 and p < 0.001, respectively), while women showed a higher frequency of mean corpuscular volume (MCV) > 95 fl (p < 0.001), and ESR > 20 mm (p = 0.001) than men.

Table 4 Body mass index (BMI) and blood parameters in 896 patients at first AUD treatment in the CohRTA Study
Full size table

In relation to biochemistry markers, women showed higher total cholesterol levels (p < 0.001). However, men had higher concentrations of triglycerides (p = 0.001), basal glucose (p < 0.001), creatinine (p < 0.001), AST (p = 0.027), and ALT (p = 0.004), as well as higher prevalence of albumin < 35 g/L (p = 0.007), total bilirubin > 1.2 mg/dL (p = 0.003), and GGT > 55 U/L (p < 0.001).

Sex-specific associations

Multivariate analysis (described in Table 5) showed that the diagnosis of MDD was four times more likely in women than men (OR = 4,29, 95% CI: 2,57–7,16), as was elevated ESR levels (> 20 mm) (OR = 2,55, 95% CI: 1,54–4,22) and total cholesterol (> 200 mg/dL) (OR = 1,61, 95% CI: 0.99–2.58).

Table 5 Logistic regression model in the AUD sample
Full size table

Age of onset of alcohol consumption (OR = 0.94, 95% CI = 0.90–0.99), BMI > 25 kg/m2 (OR = 0.56, 95% CI: 0.35–0.89), cocaine use (OR = 0.11, 95% CI = 0.03–0.46), elevated levels of total bilirubin (> 1.2 mg/dL) (OR = 0.47, 95% CI = 0.24–0.94) and GGT > 55 U/L (OR = 0.54, 95% CI = 0.33–0.88) were more likely in men than in women.

Discussion

The results of this study show sex differences when starting the first treatment for AUD. Men and women begin their first treatment at a similar age, although women report fewer years of regular alcohol consumption. Although women would be less likely to seek treatment for AUD (Agabio et al., 2017), these results indicate that when they do it for the first time, they have been exposed to excessive alcohol consumption for fewer years.

If women show less time of excessive alcohol use, the expected comorbidity could be less than that of men. However, in our results the expected comorbidity is similar for both sexes in most organs or systems; previous studies show a greater vulnerability of women to the harmful effects of alcohol after fewer years of excessive alcohol use in comparison with men (Mumenthaler et al., 1999; Randall et al., 1999).

The global medical comorbidity according to CIRS-SA is mild or moderate for both men and women; however, a high prevalence of hepatic, respiratory, vascular and neurological affectation is shown at the beginning of treatment, which suggests the need for treatment as secondary prevention to the progression of the comorbidity. The literature includes liver diseases as the most frequent complication and the first cause of death in these patients (Timko et al., 2006). Further, preclinical and clinical studies show that women are more susceptible to liver damage caused by alcohol (Alharshawi et al., 2021; Fulham & Mandrekar, 2016).

Our results also show sex differences in laboratory parameters. Specifically, elevated levels of ESR and total cholesterol are more prevalent in women with AUD. The ESR rate is a marker of systemic inflammation that tends to be more frequently altered in women and to increase with age (Alende-Castro et al., 2019; Ernst et al., 2010). Although we have not found any studies that show sex differences in total cholesterol levels associated with AUD, there is evidence that chronic alcohol consumption alters the synthesis of hepatic cholesterol (Charlet & Heinz, 2017; Rosoff et al., 2020).

On the other hand, this study shows that men are more likely to have elevated levels of GGT and total bilirubin. GGT, in addition to being an indicator of alcohol consumption, is a marker associated with diabetes mellitus, metabolic syndrome, arteriosclerosis and cardiovascular diseases (Hernández-Rubio et al., 2022; Lu et al., 2020; Ndrepepa et al., 2018). According to our results, the literature shows differences between the sexes in the study of GGT as a biomarker of hepatic steatosis with high GGT activity as a predictor of an increase in mortality among men (Haring et al., 2009).

It is noteworthy that when starting the first treatment for AUD, men are more likely to be overweight/obese than women. In contrast, a study published in 2011 showed that women were almost five times more likely to be obese grade II (BMI > 30 kg/m2) than men (Sanvisens et al., 2011). It is possible that differences in study design and patient selection may have influenced the findings. In any case, BMI has been poorly described in patients requesting treatment for AUD and suggests the need to analyze the impact of alcohol consumption on overweight and obesity.

Another interesting piece of information from the results of this study is that almost 70% of the women had a family history of AUD, which was more frequent than in men. There is literature that relates a family history of alcohol dependence with an increased risk of developing an AUD, as well as developing depressive and anxiety disorders (Hill et al., 2001; Sjoerds et al., 2013).

Men in our study showed a higher prevalence of cocaine use than women. Epidemiological studies have described that substance use disorders are between 3 and 5 times more common in men than in women, although the sex difference in the consumption of alcohol, cannabis and cocaine has been narrowing in recent years (Hecksher & Hesse, 2009).

With regard to psychiatric comorbidity, MDD was the most prevalent diagnosis in patients with AUD, being more frequent in women. This result is consistent with previous studies suggesting that emotional states have a gender-specific impact, based on the evidence of high rates of depressive comorbidity and anxiety disorders among patients in treatment for AUD, especially women (Hasin et al., 2007). There are studies that indicate that women experience complex negative emotional states and psychological stress (Brady & Sinha, 2005; Fox & Sinha, 2009), while men show more substance-induced depression (Karpyak et al., 2016).

Finally, this study has several limitations that should be mentioned. First, it is a cross-sectional design that prevents conclusions about the causality of the findings. The lack of information in relation to specific hormonal situations of women (e.g., menopause, treatment with oral contraceptives) could have caused an overestimation of the frequency of certain alterations. However, the main strength of this study is that, to our knowledge, it is the largest patient cohort on the topic in which sex differences were analyzed comprehensively.

In conclusion, the present findings support the differences between the sexes as a result of the study of the clinical variables collected for the treatment of AUD. The main differences between men and women with AUD are that men present high dysregulation in liver markers (GGT and total bilirubin) and higher risk consumption patterns, while women present high prevalence of psychiatric comorbidity, inflammatory markers (ESR) and elevated cholesterol. These results provide valuable clinical information to establish improvements in therapeutic and preventive measures based on sex in a population especially vulnerable to chronicity, such as patients with AUD.