1 Introduction & Background

Depression is a prevalent global health issue, experienced by 4–10% of the global population (Chapman et al., 2022). Currently, around 280 million people (3.8%) are affected globally (WHO, 2023), making depression one of the top contributors to the global health burden (Santomauro et al., 2021).

Although cost-efficient treatments exist across countries and stigma is decreasing, treatment challenges remain common (Ferrari et al., 2024). These include delayed diagnosis, non-individualized treatments, and inadequate response to treatment (Chapman et al., 2022).

In the UK, 15–30% of individuals still suffer from depression after two or more treatments (Chapman et al., 2022). This is relevant as depression affects education, employment, relationships, and imposes economic and healthcare costs (OECD & European Commission, 2024; Statista, 2024).

1.1 Hypotheses

  • H1. Depression prevalence increases with experienced discrimination based on sexuality.
  • H2. Depression prevalence increases with experienced discrimination based on skin color or race.
  • H3. Depression prevalence decreases with age.
  • H4. Depression prevalence is higher among females compared to males.

2 Methods

The present work aimed to investigate depression in a British population, as 15-30% of individuals do not recover from depression after two or more treatments (Chapman et al., 2022) and therefore a greater understanding of potential contributing factors is crucial for improving recovery outcomes.

2.1 Dependent Variable

Our dependent Variable is “Depression”, which was measured using the Centre for Epidemiological Studies Depression Scale (CES-D8), an 8-item scale.

2.2 Independent Variables

Discrimination based on the respondent’s sexuality (nominal: “marked”, “not marked”), discrimination based on colour or race (nominal: “marked”, “not marked”), age (ratio scale: 15-90), and gender (nominal: “male”, “female”) were considered as potential factors influencing depression.

2.3 Data Analysis

To assess the internal consistency of the depression scale, Cronbach’s alpha was calculated, typically ranging from 0 to 1 (Döring & Bortz, 2016), although it can sometimes be negative (Bühner, 2005).

#  Calculation of Cronbach's alpha (using df_uk) to check internal consistency ("reliability") of depression items
cronbach.alpha(df_uk[,c("d20", "d21", "d22", "d23", "d24", "d25", "d26", "d27")], na.rm=T)
## 
## Cronbach's alpha for the 'df_uk[, c("d20", "d21", "d22", "d23", "d24", "d25", "d26", "d27")]' data-set
## 
## Items: 8
## Sample units: 1684
## alpha: NA

The degree of agreement between items was high, with an alpha of NA, well above the recommended threshold of 0.7 (Hair, 2010), indicating good internal consistency of the depression scale (Osburn, 2000).

3 Results

In this section, the sample characteristics are presented, followed by the main analysis results.

3.1 Sample Description

The dataset was subset to include only participants from the United Kingdom (UK), resulting in an initial sample size of n = 1684. However, the final sample consisted of n = 0 valid respondents, aged 15 to 90 years, for the depression variable in the UK sample.

# Prepare Data
# In this section, "agea" (age) is converted into numeric, only participants with non-missing values (age and depression) are selected, and age groups are created. Further, intermediate depression values are rounded to whole numbers (1–4).

# Convert age to numeric
df_uk$age = as.numeric(as.character(df_uk[,"agea"]))
# calculate depression (make sure this is in df_uk)
df_uk$depression = rowSums(df_uk[, c("d20", "d21", "d22", "d23", "d24", "d25", "d26", "d27")], na.rm = TRUE) / 8

# Filter to complete cases
df_uk_complete = df_uk[!is.na(df_uk$age) & !is.na(df_uk$depression), ]

# Create age groups in df_uk_complete
# Select only participants with non-missing age and depression (49 missing depression, 37 missing age)
df_uk_complete$age_group = cut(
  df_uk_complete$age,
  breaks = c(15, 24, 34, 44, 54, 64, 74, 90),
  labels = c("15–24", "25–34", "35–44", "45–54", "55–64", "65–74", "75+"),
  include.lowest = TRUE
)
# Create depression_round 
# Round depression scores to whole numbers (1-4) and name it "df_uk_complete"
# Meaning: 1 = no or very mild symptoms; 2 = mild symptoms; 3 = moderate symptoms; 4 =
df_uk_complete$depression_round = round(df_uk_complete$depression)

3.2 Frequency Tables and Graphs

As for sample descriptions, in the pure sciences, tables are typically used to present data (N. Mevenkamp, personal communication, June 12, 2025). Both absolute and relative frequencies are presented in the form of a table. Furthermore, visualization was chosen to clearly show group differences.

3.2.1 Absolute Frequency Table

Age was categorized into groups and depression scores were rounded to whole numbers from 1 to 4, corresponding to symptom severity levels as follows:

  • 1 = no or very mild symptoms
  • 2 = mild symptoms
  • 3 = moderate symptoms
  • 4 = severe symptoms

This categorization was used to examine how frequently each level of depression occurs within each age group (see Table 1).

# Analyze depression by age group
# How frequently does each level of depression occur in each age group?
table_absolute = table(df_uk_complete$age_group, df_uk_complete$depression_round)

# Add column sums only (bottom row)
column_sums = addmargins(table_absolute, margin = 1) # 1 = row

# Name is "Total"
rownames(column_sums)[nrow(column_sums)] = "Total" # Distribution of depression across age groups

# Show the table 
scroll_box(
  kable_styling(
    kable(
      column_sums,
      col.names = c(
        "Age groups",
        "1 = no or very mild symptoms",
        "2 = mild symptoms",
        "3 = moderate symptoms",
        "4 = severe symptoms"
      ),
      caption = "Table 1: Absolute frequencies of depression by age group (N = 1635)",
      align = c("l", "r", "r", "r", "r", "r"), 
      row.names = TRUE  
    ),
    full_width = TRUE,
    font_size = 13,
    bootstrap_options = c("hover", "condensed")
  ),
  height = "300px"
)
Table 1: Absolute frequencies of depression by age group (N = 1635)
Age groups 1 = no or very mild symptoms 2 = mild symptoms 3 = moderate symptoms 4 = severe symptoms
15–24 1 1 86 9 0
25–34 0 4 211 20 0
35–44 2 11 237 12 0
45–54 1 6 216 18 1
55–64 0 5 240 12 0
65–74 1 6 261 11 0
75+ 1 7 252 15 0
Total 6 40 1503 97 1
3.2.1.1 Graph Showing Absolute Frequency Distribution

A grouped bar chart was created to visually represent the cross-tabulation of participants by age group and depression severity levels (1–4). This visualization shows the number of individuals in each age group and how depression levels are distributed across these groups. It highlights the differences in counts of depression severity within each age category, allowing for an intuitive comparison of depression patterns by age.

# Grouped bar chart – creating a visual representation of a cross-tabulation
# How many people are in each age group
# How depression levels (1–4) are distributed across these age groups
# Shows differences in counts of depression severity by age.

ggplot(df_uk_complete, aes(x = age_group, fill = as.factor(depression_round))) +
  geom_bar(position = "dodge") +
  scale_fill_manual(values = c("1" = "#F8766D", "2" = "#7CAE00", "3" = "#00BFC5", "4" = "#C77CFF")) +
  labs(
    title = "Figure 1: Depression Levels (1-4) by Age Group (Absolute Counts)",
    x = "Age Group",
    y = "Count",
    fill = "Depression Score"
  ) +
  theme_minimal()

3.2.2 Relative Frequency Table (per age group)

Table 2 displays the relative percentage frequencies of depression severity across different age groups, providing insight into how symptom levels vary by age

# Relative frequencies (row-wise %)
table_relative = prop.table(table_absolute, margin = 1)

# Round and convert to percentages
table_relative_percentages = round(table_relative * 100, 1)

# Show table
scroll_box(
  kable_styling(
    kable(
      table_relative_percentages,
      col.names = c(
        "Age groups", 
        "1 = no or very mild symptoms", 
        "2 = mild symptoms", 
        "3 = moderate symptoms", 
        "4 = severe symptoms"
      ),
      caption = "Table 2: Relative (%) frequencies of depression by age group",
      align = c("l", "r", "r", "r", "r") # 5 columns total
    ),
    full_width = TRUE,
    font_size = 13,
    bootstrap_options = c("hover", "condensed")
  ),
  height = "300px"
)
Table 2: Relative (%) frequencies of depression by age group
Age groups 1 = no or very mild symptoms 2 = mild symptoms 3 = moderate symptoms 4 = severe symptoms
15–24 1.0 1.0 88.7 9.3 0.0
25–34 0.0 1.7 89.8 8.5 0.0
35–44 0.8 4.2 90.5 4.6 0.0
45–54 0.4 2.5 89.3 7.4 0.4
55–64 0.0 1.9 93.4 4.7 0.0
65–74 0.4 2.2 93.5 3.9 0.0
75+ 0.4 2.5 91.6 5.5 0.0
3.2.2.1 Graph Showing Relative Frequency Distribution

A stacked bar chart was used to visualize the proportion of each depression severity level within different age groups. This chart allows for a relative comparison of how depression severity varies across age categories, highlighting patterns in the distribution of depression levels by age.

# Stacked bar chart: visualizes the proportion of each depression level within each age group.
# compares how depression severity varies relatively across age categories
ggplot(df_uk_complete, aes(x = age_group, fill = as.factor(depression_round))) +
  geom_bar(position = "fill") +  # relative frequencies
  scale_y_continuous(labels = percent) +
  scale_fill_manual(values = c("1" = "#F8766D", "2" = "#7CAE00", "3" = "#00BFC4", "4" = "#C77CFF")) +
  labs(
    title = "Figure 2: Depression levels by age group (relative)",
    x = "Age Group",
    y = "Percentage",
    fill = "Depression Score"
  ) +
  theme_minimal()

3.2.3 Frequency Table of Participants by Gender

The sample includes counts of participants by gender, providing an overview of the gender distribution and balance within the dataset.

# How many people of each gender are in the sample? 
# How is the distribution and balance of gender in the data?
# Create frequency table
gender_table = table(df_uk$gndr)

# Convert to data frame
gender_df_uk = as.data.frame(gender_table)
names(gender_df_uk) = c("Gender", "Count")

# Show table
scroll_box(
  kable_styling(
    kable(
      gender_df_uk,
      caption = "Table 3: Frequency distribution of Gender",
      align = c("l", "r")
    ),
    full_width = TRUE,      
    font_size = 13,
    bootstrap_options = c("striped", "hover", "condensed", "responsive"),
    position = "left"         
  ),
  height = "140px"
)
Table 3: Frequency distribution of Gender
Gender Count
Male 824
Female 860
3.2.3.1 Graph Showing Frequency Distribution of Participants by Gender

A univariate frequency distribution visualization was used to display the counts of each category. The gender variable (gndr), as a nominal scale, was ranked by frequency to show the relative prevalence of each group.

# Univariate frequency distribution visualization
# Ranking for nominal scale (gndr) 
ggplot(df_uk, aes(gndr)) +
  geom_bar(aes(fill = gndr)) +
  scale_fill_manual(values = c("Female" = "orchid", "Male" = "cornflowerblue")) +
  labs(
    title = "Figure 3: Gender distribution", 
    subtitle = "ESS Round11",
    x = "Gender",
    y = "Count",
    caption = "Valentina Lanser",
    fill = "Gender"
  ) +
  theme_minimal()

3.2.3.2 Graph Showing Gender Distribution within each Depression Score

A bar chart was created to visualize the cross-tabulation of gender and depression scores, showing the distribution of genders within each depression level and highlighting any gender differences in depression prevalence.

# Create the visual representation of a cross tabulation - bivariate analysis
# Bar chart
# Show gender distribution within each depression score
# Are there certain depression levels which are more prevalent among one gender?

# Keep only valid values (1 to 4)
df_uk_validvalues=df_uk[df_uk$depression %in% 1:4, ]

# "depression" is numeric - therefore, convert to factor and set the order
df_uk_validvalues$depression = factor(df_uk_validvalues$depression, levels = c(1, 2, 3, 4))

# Create graph
ggplot(df_uk_validvalues, aes(depression)) +
  geom_bar(aes(fill = gndr), position = "fill", width = 0.6) +
  scale_y_continuous(labels = scales::percent) +
  scale_fill_manual(values = c("cornflowerblue", "orchid")) +
  labs(
    title = "Figure 4: Depression by Gender",
    subtitle = "ESS Round11",
    x = "Depression Level",
    y = "",
    fill = "Gender"
  ) +
  theme_minimal()

3.2.4 Descriptive Statistics of Depression

This section shows basic statistics like averages and counts for the main variables, including the depression score.

3.2.4.1 Summary Statistics
# data.frame to make a table out of it
summary_df = data.frame(
  Statistic = c("Minimum", "Maximum", "Median", "Mean"),
  Value = c(1, 4, 1.875, 1.952)
)
# table
kable_styling(
kable(summary_df,
      col.names = c("Statistic", "Value"),
      caption = "Table 4: Summary Statistics of Depression Scores")
)
Table 4: Summary Statistics of Depression Scores
Statistic Value
Minimum 1.000
Maximum 4.000
Median 1.875
Mean 1.952

The depression scale has a range from 1 (lowest depression level) to 4 (highest depression level). Based on the data, we observe the following: At least one individual answered all items with a score of 1 (indicating the lowest possible depression level), and at least one individual answered all items with a score of 4 (indicating the highest possible depression level). The majority of participants report low to moderate depression levels, as indicated by the median (1.875) and mean (1.952) scores, which are relatively low. This suggests that most participants fall within the lower half of the depression scale.

3.2.4.2 Frequency Distribution of Depression

To provide an overview of the depression score patterns in the ESS UK sample, Table 5 presents their frequency distribution.

# Frequency distribution of the new variable (depression)
# table(df_uk$depression)
freq_table = as.data.frame(table(df_uk$depression))

# Calculate percentages
freq_table$Percentage = round(100 * freq_table$Freq / sum(freq_table$Freq), 1)

# Rename columns
colnames(freq_table) = c("Depression Score", "Frequency", "Percentage (%)")

# Create and style the table 
kable_output = kable(
  freq_table,
  caption = "Table 5: Frequency Distribution of Depression Scores (ESS UK Sample)",
  align = c("l", "r", "r")
)

kable_styling(
  kable_output,
  full_width = TRUE,
  position = "left",
  bootstrap_options = c("striped", "hover", "condensed", "responsive")
)
Table 5: Frequency Distribution of Depression Scores (ESS UK Sample)
Depression Score Frequency Percentage (%)
1 11 0.7
1.125 1 0.1
1.25 6 0.4
1.375 17 1.0
1.5 110 6.7
1.625 178 10.9
1.75 297 18.2
1.875 304 18.6
2 221 13.5
2.125 176 10.8
2.25 97 5.9
2.375 72 4.4
2.5 48 2.9
2.625 25 1.5
2.75 29 1.8
2.875 13 0.8
3 10 0.6
3.125 6 0.4
3.25 9 0.6
3.375 3 0.2
3.75 1 0.1
4 1 0.1
3.2.4.3 Create Likert Table and Likert Plot

Table 6 and Figure 5 present the distribution of responses to Likert-scale items regarding emotional well-being during the past week, providing both a numerical summary and a visual overview of the data.

# create formatted table
kable_styling(kable(likert_table,
                    caption = "Table 6: Distribution of Answers Regarding Depression (ESS Round 11, Based on Data From the United Kingdom)"
                    )
              )
Table 6: Distribution of Answers Regarding Depression (ESS Round 11, Based on Data From the United Kingdom)
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 often during the last week participants felt depressed 64.9 29.1 4.6 1.5 1.426 39981
How often during the last week participants felt that everything they did was an effort 48.4 38.4 9.8 3.4 1.682 39983
How often during the last week participants’s sleep was restless 43.9 39.9 11.6 4.6 1.770 40017
How often during the last week participants were happy 4.0 23.5 48.9 23.6 2.920 39890
How often during the last week participants felt lonely 68.1 24.3 5.3 2.3 1.417 39983
How often during the last week participants enjoyed life 5.3 24.8 44.8 25.0 2.895 39878
How often during the last week participants felt sad 52.5 41.1 4.9 1.6 1.555 39981
How often during the last week participants could not get going (felt lethargic and lacked motivation) 55.7 36.1 6.2 2.0 1.546 39949 
# create basic plot 
# plot(likert(summary=likert_table[,1:4])) # limit to columns 1:4 to skip mean and count
# plot(likert(likert_df)
# give a title
p = plot(likert(likert_df))
p + ggtitle("Figure 5: Likert Plot of Depression-Related Items")

3.3 Multivariate Regression (based on UK data) - Unweighted

A linear regression was conducted to predict depression scores based on dscrsexMarked (reported or experienced discrimination due to sex), dscrrceMarked (discrimination due to race), female, and age.

# Fixed intercept, fixed slope = this is a basic regression model
# Model (dscrsex, dscrrce, female, age) and show extended summary
# first: convert "agea" (age) into numeric
df_uk$age = as.numeric(as.character(df_uk[,"agea"]))
model = lm(depression ~ dscrsex + dscrrce + female + age, data=df_uk)
summary(model)
## 
## Call:
## lm(formula = depression ~ dscrsex + dscrrce + female + age, data = df_uk)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -0.99037 -0.23155 -0.04297  0.14787  2.01499 
## 
## Coefficients:
##                 Estimate Std. Error t value Pr(>|t|)    
## (Intercept)    1.9283717  0.0283416  68.040  < 2e-16 ***
## dscrsexMarked  0.0749941  0.0588679   1.274    0.203    
## dscrrceMarked  0.0329009  0.0478649   0.687    0.492    
## female         0.0727095  0.0174997   4.155 3.43e-05 ***
## age           -0.0003151  0.0004655  -0.677    0.499    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.3501 on 1600 degrees of freedom
##   (79 observations deleted due to missingness)
## Multiple R-squared:  0.01243,    Adjusted R-squared:  0.009964 
## F-statistic: 5.036 on 4 and 1600 DF,  p-value: 0.0004926

3.4 Multivariate Regression (based on UK data) - Weighted

# Model (dscrsex, dscrrce, female, age) and show extended summary
model = lm(depression ~ dscrsex + dscrrce + female + age, data=df_uk, weights = anweight)
summary(model)
## 
## Call:
## lm(formula = depression ~ dscrsex + dscrrce + female + age, data = df_uk, 
##     weights = anweight)
## 
## Weighted Residuals:
##     Min      1Q  Median      3Q     Max 
## -1.9166 -0.3651 -0.0824  0.2929  3.8764 
## 
## Coefficients:
##                 Estimate Std. Error t value Pr(>|t|)    
## (Intercept)    1.9340542  0.0247308  78.204  < 2e-16 ***
## dscrsexMarked  0.1054133  0.0532040   1.981   0.0477 *  
## dscrrceMarked  0.0700548  0.0431887   1.622   0.1050    
## female         0.0729406  0.0169170   4.312 1.72e-05 ***
## age           -0.0006411  0.0004398  -1.458   0.1452    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.6176 on 1600 degrees of freedom
##   (79 observations deleted due to missingness)
## Multiple R-squared:  0.01805,    Adjusted R-squared:  0.0156 
## F-statistic: 7.354 on 4 and 1600 DF,  p-value: 7.229e-06

  • A linear regression was conducted to predict depression scores based on dscrsexMarked (reported or experienced discrimination because of one’s sex), dscrrceMarked (discrimination due to race), female, and age.

  • The intercept was estimated at 1.934 (p < 2e-16, highly significant).

  • dscrsexMarked had a coefficient estimate of 0.105 (p = 0.048), indicating a small but statistically significant positive association with depression scores.

  • dscrrceMarked had a coefficient of 0.070 (p = 0.105), which was not statistically significant, suggesting no clear relationship with depression in this model.

  • female was associated with an increase of 0.073 in depression scores (p = 1.7e-05), indicating females tend to have slightly higher depression scores compared to males.

  • age showed a non-significant negative association with depression scores (estimate = -0.0006, p = 0.145).

  • The overall model explained approximately 1.8% of the variance in depression scores (Multiple R-squared = 0.018), suggesting that other factors not included here also contribute to depression.

3.4.1 Predictors of Clinically Significant Depression

In this work, the cut-off score of 9 (out of 24, considering 8 items scored 0–3) has been considered to identify whether individuals show a clinical level of depression. This choice is based on findings by Briggs et al. (2018), stating that, firstly, a cut-off score of 9 (based on the sum of the scores from all 8 items on the CES-D-8) identifies clinically significant depressive symptoms precisely. Secondly, at a cut-off score of 9, both sensitivity (98%) and specificity (83%) were high. Although the specificity is higher, for example at a cut-off score of 11 (96%), the sensitivity with 95% is lower. Regarding the question of whether an individual shows clinical-level depression, it can be argued that to detect as many real depression cases as possible (high sensitivity) while also keeping false positives reasonably low (good specificity), a cut-off score of 9 is a meaningful choice. Results show that out of 1,635 valid responses from UK participants, 490 individuals (30.0%) met the clinical threshold for depression, while 1,145 (70.0%) did not.

# Create tables
# Frequencies
freq_table = table(df_uk_complete$clinical_depression)
# Percentages
prop_table = round(prop.table(freq_table) * 100, 1)
# Combine into one table
result_table = data.frame(
  Category = names(freq_table),
  Frequency = as.vector(freq_table),
  Percent = as.vector(prop_table)
)

kable(result_table, caption = "Table 7: Frequencies and Percentages of Clinical Depression in the UK Sample - yes (1); no (0)")
Table 7: Frequencies and Percentages of Clinical Depression in the UK Sample - yes (1); no (0)
Category Frequency Percent
0 1145 70
1 490 30

3.4.2 Hierarchical Modeling

As literature, such as by Tsimpida et al. (2024), reports, mental health inequalities (i.e., depression) are not equally distributed and are widening more rapidly in certain areas within England. To explore whether the effects of discrimination, gender, and age on depression vary across different regions within the UK, a multilevel (hierarchical) analysis using the NUTS-1 regional classification has been conducted. The hypotheses are as follows:

  • H1. The effect of discrimination based on an individual’s sexuality on depression does not differ significantly across regions in the UK.
  • H2. The effect of gender on depression does not differ significantly across regions in the UK.
  • H3. The effect of discrimination based on an individual’s skin colour or race on depression does not differ significantly across regions in the UK.
  • H4. The effect of age on depression does not differ significantly across regions in the UK.

A null model has been created to determine if significant variation in depression scores exists between regions (whether the grouping variable at level 2 (or higher) has a significant effect on the intercept of the level 1 dependent variable. This step helps to decide whether multilevel modeling is necessary or if a simpler single-level regression is sufficient (Garson, 2020).

The model results indicated an intraclass correlation coefficient (ICC) of approximately 0.024 (2.4%). According to commonly accepted guidelines (e.g., Koo & Li, 2016), an ICC below 0.05 suggests minimal clustering and negligible regional effects on depression scores.

Therefore, regional differences account for only a small fraction of the variance in depression scores, implying that the impact of region on depression is limited in this sample. Based on this, it appears that hierarchical modeling is not essential, and single-level regression models are appropriate for this purpose.

# Hypothesis: 
# H1. The effect of discrimination based on an individual's sexuality (dscrsex) on depression does not           differ significantly across regions in the UK.
# H2. The effect of gender on depression does not differ significantly across regions in the UK.
# H3. The effect of discrimination based on an individual's skin colour or race on depression does not           differ significantly across regions in the UK.
# H4. The effect of age on depression does not differ significantly across regions in the UK.


# UK Analysis Using NUTS-1 Regional Classification (variable: region)
# Restrict to UK NUTS-1 Regions and Drop Unused Factor Levels
df_uk$region = droplevels(df_uk$region)

# Null Model (Random Intercept Only)
# Step 1: Determine whether multilevel modeling is necessary or if a single-level analysis is sufficient.
# The null model tests whether there is significant variation in depression scores across regions.
model = lmer(depression ~ 1 + (1 | region), data = df_uk)
summary(model)
## Linear mixed model fit by REML ['lmerMod']
## Formula: depression ~ 1 + (1 | region)
##    Data: df_uk
## 
## REML criterion at convergence: 1238.8
## 
## Scaled residuals: 
##     Min      1Q  Median      3Q     Max 
## -2.7326 -0.5960 -0.1693  0.5033  5.8578 
## 
## Random effects:
##  Groups   Name        Variance Std.Dev.
##  region   (Intercept) 0.003014 0.0549  
##  Residual             0.123218 0.3510  
## Number of obs: 1635, groups:  region, 12
## 
## Fixed effects:
##             Estimate Std. Error t value
## (Intercept)  1.96093    0.01834   106.9
# Intraclass correlation coefficient (ICC)
vc = as.data.frame(VarCorr(model))
icc = vc$vcov[1] / sum(vc$vcov)
print(paste("ICC =", round(icc, 3)))
## [1] "ICC = 0.024"
# Interpretation: Only about 2.4% of the total variance in depression scores is attributable to differences between regions in the UK. According to common guidelines (e.g., Koo & Li, 2016), an ICC below 0.05 is considered negligible or indicative of low clustering effects. This suggests that regional group membership has a minimal impact on depression in this dataset. Therefore, the initial decision is that a hierarchical (multilevel) model is not necessary, and simpler single-level models are likely sufficient.

4 Discussion and Conclusion

The results indicate that discrimination based on sex or race, gender, and age explain only a small portion of the variance in depression scores within the UK sample. Additionally, regional differences in depression were minimal, suggesting that other factors may play a more significant role.

Given these findings, it is important to explore additional determinants of depression such as loneliness, social isolation, and inequalities in access to healthcare services, particularly among marginalized groups. Understanding these factors could provide valuable insights for improving mental health interventions and support in the UK (see, e.g., Kirkbridge, 2024).

Future explorations should therefore consider a broader range of social and structural variables to better capture the complex causes of depression and inform effective treatment strategies.

5 References

Briggs, R., Carey, D., O’Halloran, A. M., Kenny, R. A., & Kennelly, S. P. (2018). Validation of the 8-item Centre for Epidemiological Studies Depression Scale in a cohort of community-dwelling older people: Data from The Irish Longitudinal Study on Ageing (TILDA). European Geriatric Medicine, 9(1), 121–126. https://doi.org/10.1007/s41999-017-0016-0

Bühner, M. (2005). Einführung in die Test- und Fragebogenkonstruktion (Nachdr.). Pearson Studium.

Chapman, N., Browning, M., Baghurst, D., Hotopf, M., Willis, D., Haylock, S., Zakaria, S., Speechley, J., Withey, J., Brooks, E., Chan, F., Pappa, S., Geddes, J., Insole, L., Mohammed, Z., Kessler, D., Jones, P. B., Mansoori, P., & Difficult to Treat Depression Research Priority Setting Group. (2022). Setting national research priorities for difficult-to-treat depression in the UK between 2021-2026. Journal of Global Health, 12, 09004. https://doi.org/10.7189/jogh.12.09004

Döring, N., & Bortz, J. (with Pöschl-Günther, S.). (2016). Forschungsmethoden und Evaluation in den Sozial- und Humanwissenschaften (5. vollst. überarb. Aufl). Springer. 13

Ferrari, A. J., Santomauro, D. F., Aali, A., Abate, Y. H., Abbafati, C., Abbastabar, H., Abd ElHafeez, S., Abdelmasseh, M., Abd-Elsalam, S., Abdollahi, A., Abdullahi, A., Abegaz, K. H., Abeldaño Zuñiga, R. A., Aboagye, R. G., Abolhassani, H., Abreu, L. G., Abualruz, H., Abu-Gharbieh, E., Abu-Rmeileh, N. M., … Murray, C. J. L. (2024). Global incidence, prevalence, years lived with disability (YLDs), disability adjusted life-years (DALYs), and healthy life expectancy (HALE) for 371 diseases and injuries in 204 countries and territories and 811 subnational locations, 1990 2021: A systematic analysis for the Global Burden of Disease Study 2021. The Lancet, 403(10440), 2133–2161. https://doi.org/10.1016/S0140-6736(24)00757-8

Garson, G. (2020). The null model. In The Null Model (Vol. 0, pp. 57-100). SAGE Publications, Inc., https://doi.org/10.4135/9781544319315.n3

Kirkbride, J. B., Anglin, D. M., Colman, I., Dykxhoorn, J., Jones, P. B., Patalay, P., Pitman, A., Soneson, E., Steare, T., Wright, T., & Griffiths, S. L. (2024). The social determinants of mental health and disorder: evidence, prevention and recommendations. World psychiatry : official journal of the World Psychiatric Association (WPA), 23(1), 58–90. https://doi.org/10.1002/wps.21160

Koo, T. K., & Li, M. Y. (2016). A Guideline of Selecting and Reporting Intraclass Correlation Coefficients for Reliability Research. Journal of Chiropractic Medicine, 15(2), 155–163. https://doi.org/10.1016/j.jcm.2016.02.012

OECD & European Commission. (2024). Health at a Glance: Europe 2024: State of Health in the EU Cycle. OECD. https://doi.org/10.1787/b3704e14-en

Santomauro, D. F., Mantilla Herrera, A. M., Shadid, J., Zheng, P., Ashbaugh, C., Pigott, D. M., Abbafati, C., Adolph, C., Amlag, J. O., Aravkin, A. Y., Bang-Jensen, B. L., Bertolacci, G. J., Bloom, S. S., Castellano, R., Castro, E., Chakrabarti, S.,

Tsimpida, D., Tsakiridi, A., Daras, K., Corcoran, R., & Gabbay, M. (2024). Unravelling the dynamics of mental health inequalities in England: A 12-year nationwide longitudinal spatial analysis of recorded depression prevalence. SSM - population health, 26, 101669. https://doi.org/10.1016/j.ssmph.2024.101669

Vinokur, A. D., Price, R. H., & Caplan, R. D. (1996). Hard times and hurtful partners: How financial strain affects depression and relationship satisfaction of unemployed persons and their spouses. Journal of Personality and Social Psychology, 71(1), 166–179. https://doi.org/10.1037/0022-3514.71.1.166

WHO. (2023, March 31). Depressive disorder (depression). https://www.who.int/news room/fact-sheets/detail/depression