Paper 1
2025-04-06
Authors: Lina-Marie Schwietale & Olga Kricsfalusi
Alcohol Consumption and Smoking Behavior as Influencing Factors of Depressive Symptoms and Gender Differences: A Study on the German Population
1. Introduction and Literature Review
Depression, also known as depressive disorder, is a significant mental health condition that profoundly affects individuals’ wellbeing and daily lives worldwide (WHO, 2023a). It arises from a combination of biological and social factors. According to the WHO (2023a), depressive disorders are approximately 50% more prevalent in women than in men. Simultaneously, lifestyle factors – particularly alcohol and tobacco consumption - pose major public health challenges due to their association with various physical and psychological illnesses. In Germany, these concerns are especially pronounced, with over 2 million people experiencing depressive symptoms in 2024, placing the country 16th in global depression prevalence among 180 nations (World Population Review, 2023).
Alcohol consumption plays a particularly significant role in this context. Globally, approximately one-third (32.5%) of the population has a history of alcohol use (Qi et al., 2024). Germany ranks fifth worldwide in alcohol consumption (Statista, 2025). Numerous studies have demonstrated a direct link between alcohol consumption and depressive symptoms (Qi et al., 2024; Burlaka et al., 2024). One study highlights the importance of not only the quantity but also the frequency of alcohol consumption in determining its impact on depression. The findings suggest that consumption frequency is a key factor influencing the relationship between alcohol use and depressive symptoms (Qi et al., 2024). Interestingly, Qi et al. (2024) also found that individuals who consume alcohol in small quantities have a lower likelihood of developing depression compared to non-drinkers. However, a recent German study contradicts this, emphasizes that even small amounts of alcohol can result in negative health outcomes (Maisch, 2024).
Similarly, tobacco consumption remains a pressing public health challenge, contributing significantly to the global disease burden. The world is currently facing a tobacco epidemic, with over 8 million deaths annually attributed to smoking (WHO, 2023b). Alarmingly, approximately 1.3 million of these deaths result from passive smoking, underscoring the urgency of addressing this issue (WHO, 2023b). Cigarette smoking is the primary form of tobacco consumption. A systematic review by Fluharty et al. (2016) examined the association between cigarette smoking, depression, and anxiety, revealing mixed results. While some studies found evidence linking smoking status to later depression, others did not. A cross-sectional study by Wu et al. (2023) further explored this relationship among US adults. By adjusting for potential confounders and examining smoking status, smoking volume, and smoking cessation, the researchers found that smoking behavior overall increases the risk of depression. Moreover, higher smoking frequency and volume correlated with an increased risk of depression, while smoking cessation was associated with a reduced risk. While strong evidence supports this association, some research remains inconclusive. For instance, Burlaka et al. (2024) investigated the connection between substance use - including tobacco, alcohol and cannabis - and depressive symptoms but did not find a specific link between tobacco use and depression.
Gender differences further influence the relationship between substance use and mental health outcomes. Burlaka et al. (2024) reported that female college students exhibited significantly higher depressive symptoms compared to their male counterparts. Heavy alcohol consumption appears to have a stronger impact on inducing depressive symptoms in women (Kim et al., 2024). Notably, gender disparities extend beyond alcohol, as smoking-related depression and anxiety also differ between males and females (Fluharty et al., 2016).
Despite the extensive research on alcohol consumption and its effects in Germany, there remains a significant gap in studies specifically investigating the association between alcohol use and depression. Given Germany’s high alcohol consumption rates, this study aims to examine the link between alcohol use and depressive symptoms, thereby contributing to a deeper understanding of alcohol’s impact on mental health within the German population. Additionally, while many studies have explored smoking and its effects, there is still limited research on the relationship between tobacco use and depressive symptoms. This paper will address this gap by analyzing the connection between tobacco consumption and depression, with the goal of informing preventive measures.
Understanding gender-based differences in substance use and depressive symptoms is crucial, as these disparities may influence how alcohol and tobacco consumption impact mental health. Preventive strategies targeting alcohol-related issues should account for these differences (Zhan et al., 2012). Ultimately, this study aims to contribute to global depression prevention efforts by identifying some of its underlying causes, reinforcing the idea that tackling depression is not merely an individual issue but a collective societal responsibility (Kim et al., 2024). Based on prior research, we formally hypothesize the following:
H0-1: The level of alcohol consumption is not negatively associated with depressive symptoms.
HA-1: The less alcohol people consume, the fewer depressive symptoms they experience.
H0-2: The level of tobacco consumption is not negatively associated with depressive symptoms.
HA-2: The less tobacco people consume, the fewer depressive symptoms they experience.
H0-3: The strength of the association between alcohol/tobacco consumption and depressive symptoms is not stronger for females compared to males.
HA-3: The strength of the association between alcohol/tobacco consumption and depressive symptoms is stronger for females than males.
2. Methods
In this research, we conducted the analysis using secondary data from the European Social Survey (ESS) Round 11, released in November 2024. The ESS provides high-quality, cross- national data on social and health-related behaviors, including alcohol consumption, smoking behavior, and depressive symptoms, which are central to this study. The data collection for ESS Round 11 employed a probability sampling method, representing individuals aged 15 and older (European Social Survey, 2023). Probability sampling ensures a non-zero probability of selection and involves random selection, producing unbiased estimates of population parameters, such as the sample mean and standard deviation (Bhattacherjee, 2012). Furthermore, the data underwent comprehensive quality assessment procedures and has received ethical approval, ensuring compliance with data protection and confidentiality standards (European Social Survey, 2023). Therefore, we consider the use of this data both meaningful and appropriate, as it directly aligns with our research objectives of analyzing the influence of alcohol and tobacco consumption on depressive symptoms in Germany, with a focus on gender differences.
2.1. Variables
To investigate the hypotheses in this study, the following variables were examined: depression, alcohol consumption frequency and cigarette smoking behavior. Additionally, covariates such as age, education level, and subjective general health were included in the analysis model. Further details on each variable are provided below.
2.1.1. Depressive symptoms
The CES‐D8 Depression Scale was employed to operationalize depression as the dependent variable. The CES-D8 is an 8-item questionnaire designed to measure the frequency of depressive symptoms over the past week. Respondents rated how often they experienced each symptom during the last week using a Likert-type scale, ranging from 0 (None or almost none of the time) to 3 (All or almost all of the time). See plot 1 & table 1:
| Item | None or almost none of the time | Some of the time | Most of the time | All or almost all of the time | Mean | Count |
|---|---|---|---|---|---|---|
| How much of the time during the past week did you feel depressed? | 60.6 | 32.4 | 5.2 | 1.8 | 1.5 | 2414 |
| How much of the time during the past week did you feel that everything you did was an effort? | 37.8 | 47.4 | 11.6 | 3.2 | 1.8 | 2418 |
| How much of the time during the past week did you feel your sleep was restless? | 38.1 | 40.1 | 14.4 | 7.4 | 1.9 | 2418 |
| How much of the time during the past week did you feel happy? | 3.6 | 22.1 | 52.4 | 21.9 | 2.9 | 2415 |
| How much of the time during the past week did you lonely? | 78.1 | 16.9 | 3.2 | 1.8 | 1.3 | 2419 |
| How much of the time during the past week did you feel you enjoyed life? | 4.9 | 25.9 | 46.6 | 22.6 | 2.9 | 2416 |
| How much of the time during the past week did you feel sad? | 59.4 | 35.4 | 3.6 | 1.6 | 1.5 | 2417 |
| How much of the time during the past week did you feel you could not get going? | 62.7 | 32.3 | 3.9 | 1.1 | 1.4 | 2410 |
2.1.2. Alcohol consumption
Frequency of alcohol consumption was examined as one of the independent variables in this study. Participants from the ESS survey were asked how often they had a drink containing alcohol in the past 12 months. Responses were measured on a 7-point scale ranging from 1 (Every day) to 7 (Never).
2.1.3. Smoking behavior
The independent variable cigarette smoking behavior was examined depending on the frequency. Participants responded using a 6-point scale, ranging from 1 (I smoke daily, 10 or more cigarettes) to 6 (I have never smoked).
2.1.4 Confounding variables
The confounder age was measured as a continuous variable, representing the participant’s self-reported age in years. Education level was assessed using the International Standard Classification of Education (ISCED), ranging from ISCED 1 (completed primary education) to ISCED 6 (doctoral degree). Subjective general health was measured using a Likert-type scale from 1 (Very Good) to 5 (Very Bad).
2.2. Statistical analysis
table(df_de$cldgng_n)
All statistical analyses were performed using the statistical software R, specifically R Studio. Information about the sample and key variables was summarized using descriptive statistics. A weighted model was used to ensure that all respondents are appropriately represented, providing more accurate results. Two of the eight questions from the depression scale (D23 and D25) were negatively worded and were therefore reverse coded for data analysis. Subsequently, the eight items of the questionnaire were computed into a single scale, and Cronbach’s alpha was calculated to measure internal consistency. The Kruskal-Wallis H test was used to test both H0-1 and H0-2. This test was chosen because the dependent variable, depression, is ordinal (Likert-type scale), while the independent variables, alcohol consumption and smoking behavior, are categorical (Laerd Statistics, 2014). Additionally, it allows for testing statistical differences between groups. To test H0-3, regression models with interaction terms were conducted to examine whether the association between alcohol consumption/smoking behavior and depressive symptoms is stronger for females compared to males. Finally, a multivariate linear regression model was performed to assess the influence of confounding variables on the tested associations.
3. Results
3.1. Sample Description
Table 1 presents the total number of participants included in the study with both genders almost equally represented. The mean age of participants is 36.3838509 and the standard deviation is 18.9972654.
| Gender | Count |
|---|---|
| Male | 1214 |
| Female | 1206 |
3.2. Data Analysis
The research followed a two-stage analysis, first assessing the reliability of the dependent variable and then evaluating the hypotheses. Reliability was tested using Cronbach’s alpha, indicating acceptable consistency and a reliable scale 0.7941604.
3.3. Hypotheses testing results
The Kruskal-Wallis test was conducted to examine the association between alcohol consumption frequency/smoking behavior and depressive symptoms. This is a reliable method for determining whether there are statistical differences between two groups. The results show very small p-values, 9.1818034^{-10} and 2.2494598^{-8} for both alcohol consumption and smoking behavior with respect to depressive symptoms. This indicates a significant association, allowing us to confidently reject the null hypotheses H0-1 and H0-2. Therefore, it can be concluded that both alcohol consumption frequency and smoking behavior are related to different levels of depression, thus supporting HA-1 and HA-2. Additionally, the large H values - 53.5285538 and 44.0625704 further confirm that these variables are significantly associated with depression levels.
The interaction results of the regression model indicate no statistically significant differences between genders in the association between alcohol consumption and depressive symptoms. The model fit statistics further support this finding, with R2 0.0386339 and adjusted R2 0.0333849. Similarly, there is no evidence for gender differences in the relationship between smoking behavior and depressive symptoms. The model explains only very low number of the variance in depressive symptoms, with an adjusted R2 0.0295665, suggesting a weak explanatory power.
While testing the regression models, an additional observation emerged: the model showed a statistically significant p-value for gender differences in depressive symptoms, indicating that females are more prone to depression than males.
The multivariate linear regression model for confounders showed that age and subjective general health are significantly associated with depressive symptoms, whereas education level does not have a significant impact on depression. However, the effect size for age is very small, indicating a minimal practical impact. Subjective general health, on the other hand, is strongly associated with depressive symptoms, where worse health correlates with higher depression scores. Specifically, individuals with Good/Fair Health have a depression score, while those with Bad/Very Bad Health have a significantly higher score.
##
## Call:
## lm(formula = depression ~ alcfreq + gndr + cgtsmok + gndr + agea_n +
## eisced + health, data = df_de)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.04472 -0.27157 -0.05028 0.20790 2.16524
##
## Coefficients:
## Estimate Std. Error
## (Intercept) 1.5204819 0.0786949
## alcfreqSeveral times a week -0.0221115 0.0399626
## alcfreqOnce a week -0.0123071 0.0406397
## alcfreq2-3 times a month 0.0137760 0.0435646
## alcfreqOnce a month 0.0075431 0.0464458
## alcfreqLess than once a month 0.0626197 0.0427665
## alcfreqNever 0.1115571 0.0431254
## gndrFemale 0.0806295 0.0177140
## cgtsmokI smoke daily, 9 or fewer cigarettes -0.0257815 0.0431473
## cgtsmokI smoke but not every day 0.0010662 0.0474679
## cgtsmokI don’t smoke now but I used to -0.0882779 0.0299806
## cgtsmokI have only smoked a few times -0.0688746 0.0352542
## cgtsmokI have never smoked -0.1048450 0.0281665
## agea_n -0.0023650 0.0005396
## eiscedES-ISCED II, lower secondary 0.1305194 0.0710685
## eiscedES-ISCED IIIb, lower tier upper secondary 0.0320965 0.0671653
## eiscedES-ISCED IIIa, upper tier upper secondary 0.0605257 0.0752024
## eiscedES-ISCED IV, advanced vocational, sub-degree 0.0121947 0.0678068
## eiscedES-ISCED V1, lower tertiary education, BA level 0.0108585 0.0706733
## eiscedES-ISCED V2, higher tertiary education, >= MA level 0.0051974 0.0690300
## healthGood 0.1644785 0.0242680
## healthFair 0.3517344 0.0270187
## healthBad 0.6969733 0.0366309
## healthVery bad 0.8613992 0.0850057
## t value Pr(>|t|)
## (Intercept) 19.321 < 2e-16 ***
## alcfreqSeveral times a week -0.553 0.580107
## alcfreqOnce a week -0.303 0.762043
## alcfreq2-3 times a month 0.316 0.751862
## alcfreqOnce a month 0.162 0.871000
## alcfreqLess than once a month 1.464 0.143267
## alcfreqNever 2.587 0.009746 **
## gndrFemale 4.552 5.59e-06 ***
## cgtsmokI smoke daily, 9 or fewer cigarettes -0.598 0.550216
## cgtsmokI smoke but not every day 0.022 0.982082
## cgtsmokI don’t smoke now but I used to -2.944 0.003266 **
## cgtsmokI have only smoked a few times -1.954 0.050861 .
## cgtsmokI have never smoked -3.722 0.000202 ***
## agea_n -4.382 1.23e-05 ***
## eiscedES-ISCED II, lower secondary 1.837 0.066406 .
## eiscedES-ISCED IIIb, lower tier upper secondary 0.478 0.632785
## eiscedES-ISCED IIIa, upper tier upper secondary 0.805 0.420995
## eiscedES-ISCED IV, advanced vocational, sub-degree 0.180 0.857290
## eiscedES-ISCED V1, lower tertiary education, BA level 0.154 0.877904
## eiscedES-ISCED V2, higher tertiary education, >= MA level 0.075 0.939989
## healthGood 6.778 1.54e-11 ***
## healthFair 13.018 < 2e-16 ***
## healthBad 19.027 < 2e-16 ***
## healthVery bad 10.133 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.4132 on 2354 degrees of freedom
## (42 observations deleted due to missingness)
## Multiple R-squared: 0.2254, Adjusted R-squared: 0.2178
## F-statistic: 29.78 on 23 and 2354 DF, p-value: < 2.2e-16Below the results of the weighted model are represented:
##
## Call:
## lm(formula = depression ~ alcfreq + gndr + cgtsmok + gndr + eisced +
## health, data = df_de, weights = pspwght)
##
## Weighted Residuals:
## Min 1Q Median 3Q Max
## -1.6290 -0.2069 -0.0456 0.1482 3.5826
##
## Coefficients:
## Estimate Std. Error
## (Intercept) 1.427706 0.062771
## alcfreqSeveral times a week -0.002493 0.041319
## alcfreqOnce a week 0.022997 0.041279
## alcfreq2-3 times a month -0.001568 0.043724
## alcfreqOnce a month 0.038236 0.046749
## alcfreqLess than once a month 0.065923 0.042301
## alcfreqNever 0.131503 0.042488
## gndrFemale 0.082409 0.018036
## cgtsmokI smoke daily, 9 or fewer cigarettes 0.014266 0.040834
## cgtsmokI smoke but not every day 0.011359 0.047368
## cgtsmokI don’t smoke now but I used to -0.090489 0.028890
## cgtsmokI have only smoked a few times -0.086395 0.034457
## cgtsmokI have never smoked -0.101043 0.026737
## eiscedES-ISCED II, lower secondary 0.127235 0.049089
## eiscedES-ISCED IIIb, lower tier upper secondary 0.007312 0.047585
## eiscedES-ISCED IIIa, upper tier upper secondary 0.205852 0.060683
## eiscedES-ISCED IV, advanced vocational, sub-degree -0.017402 0.049673
## eiscedES-ISCED V1, lower tertiary education, BA level 0.097623 0.056692
## eiscedES-ISCED V2, higher tertiary education, >= MA level -0.029776 0.052404
## healthGood 0.164478 0.025879
## healthFair 0.375820 0.028440
## healthBad 0.719765 0.036100
## healthVery bad 0.863829 0.083081
## t value Pr(>|t|)
## (Intercept) 22.745 < 2e-16 ***
## alcfreqSeveral times a week -0.060 0.951888
## alcfreqOnce a week 0.557 0.577497
## alcfreq2-3 times a month -0.036 0.971388
## alcfreqOnce a month 0.818 0.413492
## alcfreqLess than once a month 1.558 0.119265
## alcfreqNever 3.095 0.001991 **
## gndrFemale 4.569 5.15e-06 ***
## cgtsmokI smoke daily, 9 or fewer cigarettes 0.349 0.726855
## cgtsmokI smoke but not every day 0.240 0.810511
## cgtsmokI don’t smoke now but I used to -3.132 0.001757 **
## cgtsmokI have only smoked a few times -2.507 0.012233 *
## cgtsmokI have never smoked -3.779 0.000161 ***
## eiscedES-ISCED II, lower secondary 2.592 0.009603 **
## eiscedES-ISCED IIIb, lower tier upper secondary 0.154 0.877882
## eiscedES-ISCED IIIa, upper tier upper secondary 3.392 0.000705 ***
## eiscedES-ISCED IV, advanced vocational, sub-degree -0.350 0.726116
## eiscedES-ISCED V1, lower tertiary education, BA level 1.722 0.085203 .
## eiscedES-ISCED V2, higher tertiary education, >= MA level -0.568 0.569955
## healthGood 6.356 2.48e-10 ***
## healthFair 13.214 < 2e-16 ***
## healthBad 19.938 < 2e-16 ***
## healthVery bad 10.397 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.418 on 2359 degrees of freedom
## (38 observations deleted due to missingness)
## Multiple R-squared: 0.2551, Adjusted R-squared: 0.2482
## F-statistic: 36.73 on 22 and 2359 DF, p-value: < 2.2e-16Based on these statistical analyses, HA-3 is not supported by the evidence, since H0-3 cannot be rejected. Our study found no statistically significant gender differences in the relationship between alcohol consumption, smoking behavior, and depressive symptoms. Additionally, while subjective general health significantly influences depressive symptoms, age has a minimal effect, and education level does not show a meaningful impact in this model. Thus,subjective general health emerges as the strongest predictor, indicating that it is a major confounder in the relationship between alcohol and smoking behaviors, gender, and depression.
4. Predictors of Clinically Significant Depression
A cut-off score of 9/24 from the CES-D-8 was used for prediction of clinically significant depression. This predictor was considered appropriate, because according to Knaul et al. (2018) a cut-off score of 9/24, the sensitivity and specificity of the 8-item CES-D were 98 and 83%, respectively. The Cohen’s j for a cut-off score of 9 was 0.7855, suggestive of strong agreement and the ROC area was adequate at 0.88.
##
## 0 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
## 1 68 148 230 341 340 274 230 217 144 119 77 43 51 32 33 20 13 10 5
## 27 28 29 30 31 32
## 6 7 5 2 2 2##
## -8 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
## 1 68 148 230 341 340 274 230 217 144 119 77 43 51 32 33 20 13 10 5
## 19 20 21 22 23 24
## 6 7 5 2 2 2##
## 0 1
## 1993 427##
## 0 1
## 0.8235537 0.1764463## [1] 0.006273404## [1] 0.01771949## [1] 0.1115418## [1] 0.3147577## [1] 0.01519376## [1] 0.04230727Interpretation
From n2420 - 427 (18%) show clinically significant depressive symptoms and 1993 (82%) not.
aModel: For every one unit change in gender from male to female, the log odds of binary outcome (having vs not having clinic level depression) increases by 1.5. This is highly significant.
bModel: Compared to individuals reporting “Very good” health, those reporting “Bad” health have between 9.8 and 25.7 times higher odds of the outcome (e.g., reporting low well-being), with 95% confidence. Since the entire interval is above 1, the association is statistically significant.
cModel: Compared to individuals with ISCED I (primary education), those with higher education levels show lower but not statistically significant odds of the outcome.
Pseudo R showed acceptable model fit.
The multivariate linear regression performed earlier confirmed the above mentioned results with subjective general health significantly associated with depressive symptoms, especially those with Bad/Very Bad Health have a significantly higher score, whereas education level does not have a significant impact on depression.
5. Discussion
This study aimed to examine the association between alcohol and tobacco consumption and depressive symptoms among the German population, as well as potential gender differences in this relationship. Additionally, the study included confounding variables, namely age, education level, and subjective general health. The findings aim to contribute to the existing body of knowledge and inform preventive measures to address the rising mental health challenges.
As reported in the results section, evidence was found supporting both HA-1 and HA-2, aligning with the studies discussed in the literature section. Qi et al. (2024) identified a relationship between alcohol consumption frequency and depressive symptoms, while Li et al. (2022) highlighted excessive alcohol consumption as a risk factor for mental health. However, studies by Liang et al. (2021) and Qi et al. (2024) provided evidence that low to moderate alcohol consumption was significantly associated with a lower risk of depressive symptoms compared to never drinkers in both men and women. Although our results support HA-1, confirming that lower alcohol consumption is linked to fewer depressive symptoms, further research is needed to examine the effects of alcohol consumption at different frequency levels.
The literature on the relationship between tobacco consumption and depressive symptoms presents mixed findings. However, Wu et al. (2023) provided evidence supporting an association between smoking behavior and an overall increased risk of depression. Additionally, a longitudinal analysis by Park et al. (2024) clearly identified smoking as a significant factor in the development of depressive symptoms. This research also provided statistical support for HA-2, confirming that lower tobacco consumption is associated with fewer depressive symptoms. Nevertheless, some studies do not provide clear evidence of a straightforward relationship between smoking and depressive symptoms (Burlaka et al., 2024; Du et al. 2022).
As shown in the results section, H0-3 could not be rejected, meaning that this research does not support the hypothesis that the strength of the association between alcohol/tobacco consumption and depressive symptoms is stronger in females than in males. This finding contradicts some studies mentioned in the literature review. For instance, Burlaka et al. (2024) and Kim et al. (2024) both found a gender difference in depressive symptoms, with females generally reporting significantly higher levels of depressive symptoms compared to males. Additionally, a cross-sectional study by Yue et al. (2015) found that the association between alcohol consumption and smoking was stronger in females than in males. However, since this study was conducted in a developing country with a different target group, the differences in the results may explain the variation when compared to Germany. Nevertheless, given the inconsistencies in the literature, further research is needed to explore the factors that may influence the relationship between alcohol consumption and gender differences in depressive symptoms.
Interestingly, our analysis revealed an additional observation regarding gender differences in depressive symptoms, indicating that females are more susceptible to depression than males. Previous research has provided evidence of gender differences in depressive symptoms, with studies showing that women tend to have higher mean scores on depression scales (Guo et al., 2024; Su et al., 2024).
In our research, we found a negative association between age and both smoking and alcohol consumption patterns, as well as depression, potentially influencing the relationships among these variables. This aligns with Gao et al. (2023), who found that chronological age is not significantly associated with depressive symptoms. However, education level in our sample did not have a strong influence on depressive symptoms, contradicting the findings of Zhao et al. (2024). Subjective general health showed a strong association with depressive symptoms, with poorer health linked to higher depression scores. While our findings suggest that poor subjective general health correlates with increased depression, it is also possible that depressive symptoms contribute to a more negative perception of health, even in the absence of physical health conditions (Ito et al., 2024).
5.1. Limitations
Our study has several limitations. First and foremost, while the bivariate analysis revealed a statistically significant relationship between alcohol consumption/smoking behavior and depressive symptoms, further research is needed to identify which specific group this association occurs in. Additionally, we rejected hypotheses H0-1 and H0-2 based on contrary evidence however, we cannot definitively accept them, as future studies may yield different results (Bhattacherjee, 2022). Moreover, there is a potential risk of confounding variables that were not included in the analysis, which could have influenced the results. Finally, while we found secondary data to be highly suitable for this research, data collected in a more systematic or scientific manner would have been even more purposeful for this study.
5.2. Quality criteria
It was crucial for this statistical study to ensure high-quality research on the associations between alcohol and tobacco consumption and depressive symptoms. This study took into account quality standards including reliability, validity, and objectivity, while also acknowledging its limitations. Reliability refers to how replicable the study would be to other researchers (Krefting et al., 2013). Statistical analyses were conducted with the use of R, a widely used software for statistical analysis, ensuring reproducibility of the results. Furthermore, the study made use of secondary data, making it easier to replicate. However, a potential reliability issue exists in the form of measurement bias, since the data were self-reported through a survey. When using the same dataset, reliability would be very high, while using other data collection methods could lead to discrepancies in the results.
The validity of the study consists of two forms: internal and external validity (Krefting et al., 2013). The internal validity of a study determines whether it actually measures what it intends to measure (Krefting et al., 2013). In this study, confounding variables were accounted for such as age, education level, and subjective general health. However, some other confounding variables that might have a played a role were not accounted for, which could have influenced the results. The use of appropriate statistical analysis methods further strengthened the internal validity of this study. External validity concerns the generalizability of the results, meaning how the studies’ results can be generalized to other contexts (Krefting et al., 2013). The study sample was merely focused on the German population, enhancing its application in this demographic setting. However, the study findings are not fully generalizable to different populations, for instance under other socioeconomic or cultural circumstances.
Objectivity refers mainly to how personal biases were removed and how information was gathered in the study (Krefting et al., 2013). The study upheld objectivity through the systematic dataset that was used as well as the statistical analyses that were conducted. Findings were presented in a transparent manner, by discussing supporting as well as contradicting literature. Furthermore, by mentioning the limitations, the study remains a neutral stance to further research. Overall, this study meets essential quality criteria including reliability, validity, and objectivity. However, future research should address other potential confounders to examine gender differences in different contexts and use primary data collection methods to strengthen the quality of the study.
References
Bhattacherjee, A. (2022). Social Science Research: Principles, Methods, and Practices. In Digital Commons. University of South Florida. https://digitalcommons.usf.edu/cgi/viewcontent.cgi?article=1002&context=oa_textbooks
Burlaka, J., Johnson, R. M., Marsack-Topolewski, C. N., Hughesdon, K., Owczarzak, J., Serdiuk, O., Bogdanov, R., & Burlaka, V. (2024). Association between Current Substance Use, Healthy Behaviors, and Depression among Ukrainian College Students. International Journal of Environmental Research and Public Health, 21(5), 586. https://doi.org/10.3390/ijerph21050586
Du, X., Wu, R., Kang, L., Zhao, L., & Li, C. (2022). Tobacco smoking and depressive symptoms in Chinese middle-aged and older adults: Handling missing values in panel data with multiple imputation. Frontiers in Public Health, 10. https://doi.org/10.3389/fpubh.2022.913636
European Social Survey. (2023). European Social Survey Round 11 Sampling Guidelines: Principles and Implementation. https://www.europeansocialsurvey.org/sites/default/files/2023-06/Round-11-ESS- sampling-guidelines.pdf
Fluharty, M., Taylor, A. E., Grabski, M., & Munafò, M. R. (2016). The Association of Cigarette Smoking With Depression and Anxiety: A Systematic Review. Nicotine & Tobacco Research, 19(1), 3–13. https://doi.org/10.1093/ntr/ntw140
Gao, S., Deng, H., Wen, S., & Wang, Y. (2023). Effects of accelerated biological age on depressive symptoms in a causal reasoning framework. Journal of Affective Disorders, 339, 732–741. https://doi-org.mci.idm.oclc.org/10.1016/j.jad.2023.07.019
Guo, S., Chu, C.-B., & Zheng, X.-Y. (2024). Changes in gender disparities of depressive symptoms among middle-aged and older adults in China: an age-period-cohort analysis. Social Psychiatry and Psychiatric Epidemiology: The International Journal for Research in Social and Genetic Epidemiology and Mental Health Services, 1–14. https://doi- org.mci.idm.oclc.org/10.1007/s00127-024-02747-6
Ito, N. T., Oliveira, D., Rodrigues, F. M. S., Castro-Costa, E., Lima-Costa, M. F., & Ferri, C. P. (2024). Depressive symptoms and self-rated health among Brazilian older adults: Baseline data from the ELSI-Brazil study. Brazilian Journal of Psychiatry, 46. https://doi- org.mci.idm.oclc.org/10.47626/1516-4446-2023-3331
Kim, Y., Kim, J., Oh, J. W., & Lee, S. (2024). Association between drinking behaviors, sleep duration, and depressive symptoms. Scientific Reports, 14(1), 5992. https://doi.org/10.1038/s41598-024-56625-x
Knaul, F. M., Farmer, P. E., Krakauer, E. L., De Lima, L., Bhadelia, A., Kwete, X. J., Arreola-Ornelas, H., Gómez-Dantés, O., Rodriguez, N. M., Alleyne, G. A. O., Connor, S. R., Hunter, D. J., & Rajagopal, M. R. (2018). Alleviating the access abyss in palliative care and pain relief—An imperative of universal health coverage: The Lancet Commission report. Journal of Global Health, 7(1), 010302. https://doi.org/10.1007/s41999-017-0016-0
Krefting, L., Kuper, A., Lingard, L., & Levinson, W. (2013). Nursing Research: Generating and Assessing Evidence for Nursing Practice. Academic Medicine, 88(4), 687–689. https://www.hopkinsmedicine.org/-/media/institute-excellence- education/documents/quality_criteria_in_research.pdf
Laerd Statistics. (2014). Kruskal-Wallis H Test in SPSS Statistics | Procedure, output and interpretation of the output using a relevant example. Laerd.com. https://statistics.laerd.com/spss-tutorials/kruskal-wallis-h-test-using-spss-statistics.ph
Liang, L., Hua, R., Tang, S., Li, C., & Xie, W. (2021). Low-to-Moderate Alcohol Intake Associated with Lower Risk of Incidental Depressive Symptoms: A Pooled Analysis of Three Intercontinental Cohort Studies. Journal of Affective Disorders, 286, 49–57. https://doi.org/10.1016/j.jad.2021.02.050
Li, Y., Zhang, C., Ding, S., Li, J., Li, L., Kang, Y., Dong, X., Wan, Z., Luo, Y., Cheng, A. S., Xie, J., & Duan, Y. (2022). Physical activity, smoking, alcohol consumption and depressive symptoms among young, early mature and late mature people: A cross- sectional study of 76,223 in China. Journal of Affective Disorders, 299, 60–66. https://doi-org.mci.idm.oclc.org/10.1016/j.jad.2021.11.054
Maisch, B. (2024). Alcohol consumption—None is better than a little. Herz, 49(6), 409 419. https://doi.org/10.1007/s00059-024-05280-z
Park, S. K., Oh, C.-M., Kim, E., Ryoo, J.-H., & Jung, J. Y. (2024). The longitudinal analysis for the association between smoking and the risk of depressive symptoms. BMC Psychiatry, 24(1). https://doi-org.mci.idm.oclc.org/10.1186/s12888-024-05828-7
Qi, P., Huang, M., & Zhu, H. (2024). Association between alcohol drinking frequency and depression among adults in the United States: A cross-sectional study. BMC Psychiatry, 24(1), 836. https://doi.org/10.1186/s12888-024-06296-9
Statista. (2025, January 3). Per capita alcohol consumption worldwide 2020, by country. Statista. https://www.statista.com/forecasts/1148811/per-capita-alcohol-consumption-by-country
Su, Z., Yang, X., Hou, J., Liu, S., Wang, Y., & Chen, Z. (2024). Gender differences in the co- occurrence of anxiety and depressive symptoms among early adolescents: A network approach. Journal of Psychiatric Research, 179, 300–305. https://doi- org.mci.idm.oclc.org/10.1016/j.jpsychires.2024.09.024
World Health Organization. (2023a, March 31). Depressive disorder (depression). World Health Organization; World Health Organization. https://www.who.int/news-room/fact- sheets/detail/depression
World Health Organization. (2023b, July 31). Tobacco. World Health Organization. https://www.who.int/news-room/fact-sheets/detail/tobacco
World Population Review. (2023). Depression Rates By Country 2023. World Population Review. https://worldpopulationreview.com/country-rankings/depression-rates-by- country
Wu, Z., Yue, Q., Zhao, Z., Wen, J., Tang, L., Zhong, Z., Yang, J., Yuan, Y., & Zhang, X. (2023). A cross-sectional study of smoking and depression among US adults: NHANES (2005– 2018). Frontiers in Public Health, 11, 1081706. https://doi.org/10.3389/fpubh.2023.1081706
Yue, Y., Hong, L., Guo, L., Gao, X., Deng, J., Huang, J., Huang, G., & Lu, C. (2015). Gender differences in the association between cigarette smoking, alcohol consumption and depressive symptoms: a cross-sectional study among Chinese adolescents. Scientific Reports, 17959. https://doi-org.mci.idm.oclc.org/10.1038/srep17959
Zhan, W., V. Shaboltas, A., & V. Skochilov, R. (2012). Gender Differences in the Relationship between Alcohol Use and Depressive Symptoms in St. Petersburg, Russia. Journal of Addiction Research & Therapy, 03(02). https://doi.org/10.4172/2155-6105.1000124
Zhao, R., Wang, J., Lou, J., Liu, M., Deng, J., Huang, D., & Fang, H. (2024). The effect of education level on depressive symptoms in Chinese older adults–parallel mediating effects of economic security level and subjective memory ability. BMC Geriatrics, 24(1). https://doi-org.mci.idm.oclc.org/10.1186/s12877-024-05233-5