Digitalization is increasing the prevalence of anxiety and depression worldwide

By Pooja Toshniwal PahariaReviewed by Benedette Cuffari, M.Sc.Apr 11 2023

In a recent study published in the journal PLoS ONE, researchers evaluate the impact of digitalization on a societal level by using a linguistic big-data approach.

Study: Increasing digitalization is associated with anxiety and depression: A Google Ngram analysis. Image Credit: insta_photos / Shutterstock.com

Background

The prevalence rates of depression and anxiety are rising worldwide, especially after the outbreak of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the pathogen responsible for the coronavirus disease 2019 (COVID-19) pandemic. Efforts taken for improving mental health emphasize in-person therapies or providing remote health interventions through digitalized platforms.

The identification of associated risk factors, aggravating factors, and characterization of the underlying processes of mental dysfunction could improve the efficacy of treatment and, as a result, the standard of care provided to patients. Previous studies have evaluated society-level risk factors restricted to socioeconomic status, unemployment, social capital, climatic change, and migration, with most of these studies depending on self-documented data for analysis.

About the study

In the present study, researchers utilized data retrieved from the Google Books Ngram Viewer (Google Ngram) to investigate the digital revolution's impact on depression and anxiety by reviewing the historically disjointed works of literature on anxiety, depression, and digitalization.

The usage frequencies of the words ‘anxiety,’ 'depression,' and 'digitalization' over the previous 50 years up to 2019 were retrieved and adjusted from the Google Ngram digital data repository, which comprises eight million books, or 6% of all books published. The word frequencies for the 'religion' world list were also assessed as controls.

The analysis data was from six languages, including German, British English, Russian, Spanish, Italian, and French. Lists of depression and anxiety words were retrieved from the International Classification of Diseases eighth revision (ICD-8), ICD-9, ICD-10, and ICD-11 codes. The team derived lists of digitalization words from Wolf and Bartelheimer and Brewster and Murray-Smith, spanning over 50 years.

Using the Younes and Reips method sets comprising 26 and 33 words that most precisely represented depression and anxiety, and digitalization, respectively, were generated. Next, Roget's Thesaurus online dictionary was used to search for synonyms to exclude those with a broader linguistic meaning.

By using the method, 85 words across the Spanish, German, French, Italian, and Russian languages were obtained, with cross-validations from 73 inhabitants who were fluent in English. Terms having double meanings were excluded from the analysis, following which the two final lists comprised 59 terms. Word inflections were analyzed for quantifying the word lists utilizing the '_INF' tag for every word and comparatively evaluating the word inflection frequencies.

The team calculated the correlation coefficients between the word frequencies in years and percentages, as well as the summed-up z-scores across all languages and word lists.

Results

Depression and anxiety word frequencies were significantly correlated, as were digitalization and anxiety word frequencies. No statistically significant correlation was observed for the control construct' religion' word frequencies over the previous 50 years. The word frequencies for religion and depression were negatively correlated.

Strongly positive correlations were observed for word frequencies across languages between anxiety and years, depression and years, and digitalization and years; however, no statistically significant correlations between religion and years were observed. Similar correlations were observed for depression and anxiety words and negative correlations for religion using Spanish, German, Italian, or French words exclusively.

English and Russian words showed marginally different correlations for depression and anxiety; however, these results were significantly negative for the correlation between digitalization and religion.

Conclusions

Taken together, depression, digitalization, and anxiety frequencies jointly elevated in the previous 50 years between 1970 and 2019 as compared to the control religion word list, with anxiety and depression strongly correlated with each other.

These findings highlight the effect of digitalization on depression and anxiety, thereby adding to social science data by describing the relatively increased frequency of the use of digitalization, depression, and anxiety terms. These observations also provide further support for the simultaneous occurrence of depression and anxiety.

Notably, the researchers devised a novel approach for investigating the causes and prevalence of mental health disorders by refining current linguistic-type big data methods. Thus, the methodology used here could aid in developing different methods to assess self-reported data in social psychology and clinical research.

The study results contribute to understanding the impact of societal alterations on the mental well-being of individuals. While individuals may benefit from digital health interventions in terms of their increased accessibility, increasing digitalization may degrade the mental well-being of individuals. Thus, governments and other health authorities must only adopt digitalization after weighing its risk-benefit ratio.