You are here

  • Home
  • Archive
  • Volume 27, Issue 1
  • Prescriptions of antidepressants and anxiolytics in France 2012–2022 and changes with the COVID-19 pandemic: interrupted time series analysis

Article Text

Article menu

Download PDFPDF

Original research
Prescriptions of antidepressants and anxiolytics in France 2012–2022 and changes with the COVID-19 pandemic: interrupted time series analysis
  1. David De Bandt1,2,
  2. Sarah R Haile3,
  3. Louise Devillers1,2,
  4. Bastien Bourrion1,2,
  5. Dominik Menges3
  1. 1General Practice, Versailles Saint-Quentin-en-Yvelines University Faculty of Science, Montigny le Bretonneux, France
  2. 2Center for Research in Epidemiology and Population Health, The French National Institute of Health and Medical Research, INSERM U1018 Equipe 11, Villejuif, Île-de-France, France
  3. 3Epidemiology, Biostatistics and Prevention Institute, University of Zurich, Zurich, Switzerland
  1. Correspondence to Dr David De Bandt, General practice, Versailles Saint-Quentin-en-Yvelines University Faculty of Science, Montigny le Bretonneux 78180, France; david.debandt{at}gmail.com

Abstract

Background Depression and anxiety have increased in prevalence since the start of the COVID-19 pandemic.

Objective To evaluate the consumption of antidepressants and anxiolytics from 2012 to 2022 and the pandemic’s potential impact in France.

Methods We conducted an interrupted time series analysis of routine drug sales data (Medic'AM) from all French outpatient pharmacies from 2012 to 2022. We investigated trends in defined daily doses of antidepressants and anxiolytics sold per 1000 inhabitants (DDD/TID) and related expenditures before and after pandemic onset and in relation with stringency of pandemic mitigation measures. Analyses were performed descriptively and using segmented linear regression, autoregressive and autoregressive integrated moving average models.

Findings From 2012 to 2019, overall monthly antidepressant sales increased (+0.02 DDD/TID) while monthly anxiolytic sales decreased (−0.07 DDD/TID). With pandemic onset, there was a relevant and persisting trend increase (+0.20 DDD/TID per month) for antidepressant sales overall, with an estimated excess of 112.6 DDD/TID sold from May 2020 until December 2022. Anxiolytic sales were elevated from February 2020 throughout the pandemic but returned to expected levels by December 2022, with an estimated excess of 33.8 DDD/TID. There was no evident association between stringency and antidepressant or anxiolytic sales.

Conclusions This study showed a protracted trend increase in the consumption of antidepressants since pandemic onset, while increases in anxiolytic consumption were temporary.

Clinical implications We provide evidence that the COVID-19 pandemic may have had long-lasting consequences on the prevalence and treatment of depression and anxiety disorders, requiring further actions by researchers and policy-makers to address this potential public mental health crisis.

  • Anxiety disorders
  • Depression & mood disorders
  • COVID-19
  • Data Interpretation, Statistical

Data availability statement

Data are available in a public, open access repository: https://assurance-maladie.ameli.fr/etudes-et-donnees/open-data-depenses-sante-soins-ville

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

https://doi.org/10.1136/bmjment-2024-301026

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    WHAT IS ALREADY KNOWN ON THIS TOPIC

    • Increases in diagnoses of depression and anxiety as well as increases in sales of antidepressant and anxiolytic drugs with the onset of the COVID-19 pandemic were observed in various countries.

    WHAT THIS STUDY ADDS

    • This study showed that while antidepressant sales increased and anxiolytic sales decreased from 2012 to 2019, there was a relevant increase in trend for antidepressant sales with the COVID-19 pandemic, overall and particularly for first-line treatments. Anxiolytic sales temporarily increased from 2020 to 2022 but returned to expected levels, and the increase in trend for antidepressant sales was sustained until the end of 2022. In contrast to earlier studies, this study demonstrated a protracted increase in antidepressant sales, pointing towards a sustained issue in mental healthcare.

    HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

    • Antidepressant and anxiolytic sales relevantly exceeded the expected trends during the COVID-19 pandemic, which warrants further investigation and attention by public health policymakers to address its causes.

    Introduction

    Mental health importantly affects overall well-being, and depression and anxiety have become a major public health concern. In 2019, the WHO estimated that 280 million and 301 million individuals were suffering from depression and anxiety worldwide,1 respectively, and increases in the prevalence of depression and anxiety have been reported for various countries over the past years.2 3 While non-pharmacological interventions such as cognitive-behavioural therapy are an important part of treatment for these conditions, management often includes the use of antidepressant and anxiolytic drugs.4 5 Antidepressant and anxiolytic prescriptions have evolved dynamically over time, reflecting changes in societal attitudes, healthcare policies, clinical practice guidelines and prevailing patterns of anxiety and depression. Overall, the consumption of antidepressants has increased, while anxiolytic use has decreased over the last decade.6 7

    The COVID-19 pandemic profoundly disrupted societies worldwide and affected healthcare delivery and mental well-being in the population,8 as nations grappled with stringent public health measures such as lockdowns and physical distancing. While the WHO declared an end to the global health emergency in May 2023,9 its longer term implications on mental health and the prevalence of depression and anxiety merit careful investigation. Several studies from various countries have shown increases in depression and anxiety diagnoses and prescriptions of antidepressants and anxiolytics in the general population early after pandemic onset.8 10–18 This indicated a substantial acute impact of the pandemic on mental health. However, it remains unclear whether this trend persisted throughout the pandemic, and whether greater stringency of pandemic mitigation measures was associated with a higher incidence of depression and anxiety.

    This study aimed to describe the consumption of antidepressants and anxiolytics from 2012 until 2022 in France and to evaluate the potential impact of the COVID-19 pandemic. Specific objectives were to describe antidepressant and anxiolytic sales over this timeframe, assess whether previously reported short-term changes at pandemic onset persisted and evaluate the association of drug sales with the stringency of pandemic mitigation measures.

    Materials and methods

    French healthcare system

    The French healthcare system is based on a statutory public health insurance with complementary private insurance, covering a large proportion of health services for the sociodemographically diverse population.19 Mental health services are widely available, although hospital-centred and accessible only through referral by a general practitioner. While insurance coverage includes most mental health services, consultations with psychologists were not reimbursed until 2022 and France has been affected by a significant shortage of mental health professionals in recent years.19–21

    Data source

    We used routine drug sales data from all outpatient pharmacies in France from 1 January 2012 until 31 December 2022, derived from the Medic'AM database.22 Data were publicly available by month and by individual presentation of each drug, that is, pharmacological compound as defined by Anatomical Therapeutic Chemical (ATC) class, dosage, package size and producer.23 They included drugs prescribed by licensed French physicians and reimbursed by health insurances, while drugs supplied by hospitals, care facilities for dependent elderly people without pharmacy or hospitalisation-at-home care were not included. We combined monthly data sets and restricted the data to antidepressants and anxiolytics (ATC3 classes N06A and N05B, respectively).

    Outcome definition

    The outcomes of interest of this study were drug sales defined as the number of defined daily doses (DDDs) sold per 1000 inhabitants of France per day (DDD/TID).24 DDD/TID is a standard measure for the quantity of a drug based on its average daily maintenance dose, standardised to the underlying population size. This allows to combine individual drugs with different dosages in one measure for pharmacoepidemiologic analyses. We first determined DDDs (ie, assumed average maintenance doses per day) for each individual drug (ie, ATC5 class) based on the WHO reference database.23 From this and the number of packages sold, per unit dosages, and package sizes, we calculated the number of DDDs sold in each month for each individual drug. We then calculated the DDD/TID in each month for each individual drug and drug class. In the absence of monthly population data, we used the estimated population size of France on 1 January of the respective year for calculations.25 We further evaluated health system expenditures of related drug sales defined as the total amount reimbursed by health insurances (see online supplemental material).

    Explanatory factors

    One objective was to investigate the potential impact of the COVID-19 pandemic on antidepressant and anxiolytic sales. Since the first SARS-CoV-2 infection in France was diagnosed on 24 January 2020, we assumed the first month of the pandemic to be February 2020 in analyses.26 A priori, we assumed that any effects of the pandemic on depression and antidepressant prescriptions would have occurred with a delay of approximately 3 months after pandemic onset and after lifting of the first lockdown in May 2020, consistent with evidence from France.27 We assumed that relevant depressive symptoms would require time to develop and that increases in new diagnoses and prescriptions for exacerbated pre-existing conditions could have occurred only with improved access to mental health services after the lockdown. Simultaneously, we assumed that due to the acute threat of the pandemic, any effects on anxiety or sleeping problems, and consequently, anxiolytics prescriptions, would have occurred without delay. We further hypothesised that there may be a positive correlation between the stringency of pandemic mitigation measures and antidepressant and anxiolytic prescriptions (higher stringency associated with more prescriptions). We used monthly averages of the Oxford COVID-19 Government Response Tracker Stringency Index for France to evaluate such a potential correlation.28

    Statistical analyses

    We descriptively analysed the data for antidepressant and anxiolytic sales and report aggregate data for each year, overall and by individual drug. We limited reporting for individual drugs to those considered most relevant, which we defined as being sold at quantities >0.5 DDD/TID between 2018 and 2022 (excluding etifoxine since its use was restricted by the European Medicines Agency in 202229). Overall estimates always include all drugs in the respective drug class.

    We used an interrupted time series (ITS) approach to evaluate changes in antidepressant and anxiolytic sales with pandemic onset. This quasi-experimental approach assumes that there is an underlying historical trend, which has been altered by one or more events at specific points in time, allowing to estimate the counterfactual had the event not occurred.30 In our study, we assumed changes to take effect in May 2020 for antidepressants (lag of 3 months) and February 2020 for anxiolytics (no lag), as outlined above. We further assumed no relevant waning of any potential pandemic effects on mental health until December 2022. We thoroughly evaluated the data concerning the key assumptions of ITS analyses (see online supplemental material).

    We used segmented linear regression models, autoregressive models and autoregressive integrated moving average (ARIMA) models to estimate monthly changes in drug sales prior to pandemic onset (prepandemic trend), changes with pandemic onset (immediate level change) and changes in monthly drug sales trends (change in trend) after pandemic onset (see online supplemental material for details on analytical approaches and their assumptions).31–33 Presented results are based on autoregressive models, allowing a straightforward interpretation while accounting for autocorrelation and seasonality. We conducted several sensitivity analyses using different model specifications to account for non-stationarity, autocorrelation and seasonality, and assuming different lag times for the onset of any potential effects of the pandemic. To estimate prepandemic trends, immediate level changes and changes in trend of individual drug sales, we fitted separate segmented linear regression, autoregressive and ARIMA models for each drug in exploratory analyses. Of note, the results cannot be extrapolated beyond the timeframe covered by the data.

    Based on observed and predicted outcomes, we estimated the total amount of excess DDD/TID of antidepressants and anxiolytics sold between pandemic onset and December 2022 as the sum of differences at each month. Using a Monte Carlo simulation approach with 1 000 000 iterations, we sampled predicted outcomes from a normal distribution based on the 95% prediction interval (PI) and calculated the sum of monthly differences with the observed outcomes over the timeframe for each iteration. Based on this, we estimated the mean and 2.5% and 97.5% simulation percentiles of the excess DDD/TID for antidepressants and anxiolytics from May 2020 and February 2020 until December 2022, respectively.

    We interpreted the results in terms of their strength of statistical evidence and did not use a significance threshold for reported unadjusted p values.34 We performed all analyses in R (V.4.2.2), using the forecast (V.8.19), lmtest (V.0.9–40) and Hmisc (V.4.7–2) packages.

    Patient and public involvement

    No patients or members from the public have been involved in the conceptualisation and conduct of this pharmacoepidemiological study based on routine healthcare data.

    Results

    Antidepressant sales 2012–2022

    After an initial decline from 2012 until 2013, there was a consistent increase in overall antidepressant sales up to 2019, ranging from 47.0 DDD/TID in 2013 to 49.1 DDD/TID in 2019 (figure 1, online supplemental table 1). From 2020 to 2022 during the COVID-19 pandemic, antidepressant sales increased to 55.2 DDD/TID in 2022. The most frequently sold antidepressants in 2022 were escitalopram (11 DDD/TID), fluoxetine (5.9 DDD/TID), paroxetine (9.8 DDD/TID), sertraline (5.8 DDD/TID) and venlafaxine (7.8 DDD/TID). Sales trends from 2012 until 2022 showed relevant increases for these drugs (fluoxetine +25.5%, paroxetine +29.3%, sertraline +102.7%, venlafaxine +38.5%), except for escitalopram demonstrating a generally decreasing trend (−12.4%) from 2012 until 2020 with a subsequent increase (+2.4%) through 2021 and 2022 (online supplemental figure 2). Vortioxetine has shown an increase in sales up to 2022 (2.2 DDD/TID) since its market introduction in 2016. Health system expenditures for antidepressants strongly decreased in the years from 2012 until 2016, with subsequently slower declines until 2020 (online supplemental figures 1 and 3, online supplemental tables 2 and 3).

    Figure 1
    Figure 1

    Antidepressant (A) and anxiolytic (B) drugs sold in pharmacies in France from 2012 until 2022 in defined daily doses per 1000 inhabitants per day (DDD/TID). Detailed results for drug sales and health system expenditures are reported in the online supplemental material.

    Anxiolytic sales 2012–2022

    Overall anxiolytic sales declined from 44.9 DDD/TID in 2012 to 36.7 DDD/TID in 2022, with a temporary increase from this trend in 2020 (39.4 DDD/TID) and in 2021 (38.8 DDD/TID; figure 1, online supplemental table 1). Alprazolam (8.4 DDD/TID), oxazepam (8.2 DDD/TID), bromazepam (5.2 DDD/TID), lorazepam (4.5 DDD/TID) and diazepam (4.0 DDD/TID) were the most sold anxiolytics in 2022. From 2012 until 2022, sales of alprazolam remained relatively stable (+2.3%), while there was a decrease for bromazepam (−37.0%) and lorazepam (−31.6%), and an increase for diazepam (+60.6%) and oxazepam (+23.7%; online supplemental figure 4). Expenditures for anxiolytics declined relevantly from 2012 until 2016 for all individual drugs, with two episodes of increased expenditures from 2018 to 2019 and from 2021 to 2022 (online supplemental figures 1 and 5, online supplemental tables 2 and 3).

    Impact of the COVID-19 pandemic

    In ITS analyses of overall antidepressant sales, there was strong evidence for a slightly increasing trend before the pandemic (+0.02 DDD/TID per month, 95% CI 0.01 to 0.03), no evidence for an immediate change in May 2020 (+0.42 DDD/TID, 95% CI −0.39 to 1.24), and very strong evidence for an increase in trend of antidepressant sales after May 2020 (+0.20 DDD/TID per month, 95% CI 0.15 to 0.24; figure 2, table 1). Overall, we estimated that an excess of 112.6 DDD/TID (95% PI 88.3 to 136.8) of antidepressants was sold over the period from May 2020 until December 2022. Effect size, strength of evidence for the change in trend and estimated excess sales were consistent across sensitivity analyses using different models, although estimated excess sales were slightly lower (82.5 DDD/TID, 95% PI 57.2 to 107.8) with the ARIMA model (online supplemental table 4, online supplemental figure 7). Sensitivity analyses assuming different lag times gave similar results (online supplemental table 5). For individual antidepressant drugs, there was very strong evidence for a prepandemic decrease in sales for citalopram, clomipramine and escitalopram and an increase for duloxetine, fluoxetine, mianserin, mirtazapine, paroxetine, sertraline, venlafaxine and vortioxetine (table 1, online supplemental figure 8). Furthermore, there was very strong evidence for an increase in trend for citalopram, fluoxetine, mianserin, paroxetine, sertraline and venlafaxine compared with the prepandemic trend. Results from sensitivity analyses resulted in similar findings (online supplemental table 6).

    Figure 2
    Figure 2

    Interrupted time series analysis of the potential impact of the COVID-19 pandemic on antidepressant (A) and anxiolytic (B) drug sales in France from 2012 until 2022 in defined daily doses per 1000 inhabitants per day (DDD/TID). The analyses assumed a pre-pandemic trend, an immediate level change and a change in the trend with the onset of the pandemic. Straight lines represent results based on segmented linear regression models, serrated lines represent results based on autoregressive models (see online supplemental material for further details and sensitivity analyses). The analyses assumed a delay of 3 months for any potential effects on antidepressant drug sales (ie, onset in May 2020) and no delay for any potential effects on anxiolytic drug sales (ie, onset in February 2020). The shaded area in green represents the duration of the COVID-19 pandemic from February 2020 until August 2022 when all pandemic mitigation measures were lifted in France. The shaded area around the regression lines represents the 95% CI based on the segmented linear regression models.

    View this table:
    Table 1

    Prepandemic trend, immediate level change and change in trend of DDD/TID for antidepressants overall and for individual drugs, based on autoregressive models

    For anxiolytics, there was very strong evidence for a decreasing prepandemic trend (−0.07 DDD/TID per month, 95% CI −0.08 to −0.06), very strong evidence for an immediate level change in February 2020 (+1.59 DDD/TID, 95% CI 0.80 to 2.38), and some evidence for a change in trend after February 2020 (−0.04 DDD/TID per month, 95% CI −0.08 to 0.01) overall (figure 2, table 2). The estimated excess for anxiolytic sales was 33.8 DDD/TID (95% PI 12.9 to 54.8) from February 2020 until December 2022. Results were comparable across all sensitivity analyses, including the ARIMA model, although most models did not provide evidence for a change in trend after pandemic onset (online supplemental tables 7 and 8, online supplemental figure 10). When evaluating individual anxiolytic drugs, there was very strong evidence for a prepandemic decline in sales for bromazepam, clobazam, clorazepate, hydroxyzine, lorazepam and prazepam, and for an increase for clotiazepam, diazepam and oxazepam (table 2, (online supplemental figure 11). There was very strong evidence for an increase in trend in February 2020 for sales of bromazepam, clorazepate, lorazepam and prazepam with respect to the prepandemic trend. Sensitivity analyses showed similar results, although the strength of evidence for changes in trends was lower with ARIMA models (online supplemental table 9).

    View this table:
    Table 2

    Pre-pandemic trend, immediate level change and change in trend of DDD/TID for anxiolytics overall and for individual drugs, based on autoregressive models

    Stringency of pandemic mitigation measures

    When assessing any potential associations between the stringency of pandemic mitigation measures in France and antidepressant and anxiolytic sales over the timeframe from February 2020 until December 2022, there was no indication of a relevant correlation with stringency in the same or preceding months for both drug classes (figure 3). Peaks in stringency in spring/summer 2020 and fall/winter 2020, and the lifting of all restrictions in France in August 2022 were not clearly followed by changes in sales, providing no evidence for such an association.

    Figure 3
    Figure 3

    Stringency of pandemic mitigation measures from February 2020 until December 2022 in France and correlation with antidepressant (A) and anxiolytic (B) drug sales in defined daily doses per 1000 inhabitants per day (DDD/TID). Stringency was quantified using the Oxford stringency index.28 No clear correlation between increased or decreased stringency of pandemic measures was identified over the observed timeframe. All pandemic mitigation measures were lifted in August 2022 in France.

    Discussion

    Main findings

    In this ITS analysis of French outpatient pharmacy sales data, we found an increase in overall antidepressant sales and a decrease in overall anxiolytic sales over the period from 2012 to 2019. With onset of the COVID-19 pandemic, we observed a relevant and persisting increase of antidepressant sales trends compared with the prepandemic trend. We estimated that an excess of about 113 DDD/TID of antidepressants was sold from May 2020 until December 2022 compared with what would have been expected. Meanwhile, we found an immediate increase in anxiolytic sales in February 2020, but no long-lasting change in trend compared with before the pandemic. While anxiolytic sales were elevated from February 2020 until December 2022 with an estimated excess of 34 DDD/TID, they returned to what would have been expected based on the prepandemic trend by December 2022. We observed no clear association between antidepressant and anxiolytic sales and stringency of pandemic mitigation measures.

    Evidence in context

    The observed increases in the antidepressant sales trend from 2012 to 2019 can likely be explained by an increasing prevalence of depression.2 3 This may be driven by increased diagnosis and treatment of depressive disorders and by changes in the population structure, as depression is more prevalent in the growing elderly and migrant populations.35 Antidepressants demonstrating an increasing prepandemic trend were duloxetine, fluoxetine, mianserin, mirtazapine, paroxetine, sertraline, venlafaxine and vortioxetine, which are commonly prescribed as first-line treatment in depressive or anxiety disorders.36 Meanwhile, sales of tricyclic antidepressants, which were not recommended as first-line treatment,37 and certain selective serotonin reuptake inhibitors with cardiac adverse effects such as escitalopram decreased.38

    Anxiolytic sales demonstrated a declining trend from 2012 to 2019 in line with other studies, as their utilisation—especially the prescription of benzodiazepines—was increasingly discouraged in clinical practice guidelines due to their risk for adverse effects and dependence.6 7 This decline was observed for most anxiolytics, with the exception of alprazolam, clotiazepam, diazepam and oxazepam. Those drugs are recommended first-line drugs for short-term treatment of acute crises of panic disorders and generalised anxiety disorders, while antidepressants such as selective serotonin reuptake inhibitors are recommended for their long-term treatment.37 39

    Our findings of an increase in trend of antidepressant sales with the COVID-19 pandemic are in accordance with several other studies conducted earlier during the pandemic and based on shorter observation timeframes, which have reported increases in the prevalence of depressive and anxiety disorders alongside increased sales or consumption of psychotropic drugs.8 10–18 One early study from the United Kingdom argued that temporary effects of the pandemic could be responsible for these increases and suggested that early increases of antidepressants and anxiolytics sales may be due to stockpiling effects because of the lockdown.10 Potential stockpiling was also reported in several Scandinavian countries.16 While sales of both drug classes spiked in March 2020 likely for this reason in our study (figure 2), increases in sales were sustained for almost 3 years suggesting a long-term impact. One study from Portugal found that there was a sustained decrease in antidepressant prescriptions over the first year after pandemic onset.40 Another study from Spain found a larger relative increase in anxiolytic prescriptions compared with antidepressants, which contrasts with our findings of a larger relative increase in antidepressant sales.11 Finally, a study from Croatia reported that observed increases were not statistically significantly different from prepandemic trends based on an ITS of annual data (likely resulting in lower precision than monthly data).18 Methodological differences and shorter observation periods may explain at least some of the differences with our study, but specific healthcare systems, provision of mental health services, and cultures also need to be considered.

    There is evidence that quarantine and isolation, social distancing measures and the uncertainties of the pandemic contributed to increases in depressive and anxiety disorders, which may be especially pronounced for individuals who already had a medical condition prior to the pandemic.41 42 Several studies have also reported increases in social anxiety in the population with the pandemic.43 This may be partly reflected by our observation of a temporary increase in anxiolytic sales which returned to expected levels by the end of 2022. However, we observed no association between antidepressant and anxiolytic sales and stringency, and the increase in trend of antidepressant sales persisted until the end of 2022 when stringency was greatly reduced. The longer timeframe of this study, thus, provides arguments for a real and sustained impact of the pandemic on antidepressant sales, making explanations other than short-term changes more likely.

    In our study, we observed sales trend increases primarily for recommended first-line antidepressants including fluoxetine, mianserin, paroxetine, sertraline, venlafaxine and citalopram.36 37 The results suggest an overall increasing prevalence of depressive and anxiety disorders due to the pandemic, in line with other studies.8 However, antidepressant sales are only a proxy for the prevalence and several reasons may have contributed to this observation: First, the pandemic and its consequences on the healthcare workforce have severely affected the provision of health services in general practice and outpatient mental healthcare in France.20 21 Before the pandemic, these sectors had already experienced important challenges by shortages of medical personnel.19–21 As non-pharmacological supportive treatments are time-consuming, require personal contact and consultations with psychologists were not fully covered by the French general social security system,19 newly diagnosed (incident) cases of depression and anxiety may consequently have been treated more frequently using medication than prior to the pandemic. Additionally, pre-existing conditions previously managed non-pharmacologically may have had to be managed with antidepressants or anxiolytics. Second, pre-existing psychiatric conditions may have exacerbated with the pandemic, leading to an increase in antidepressant prescriptions for individuals already suffering from mental health issues prior to pandemic onset. Third, the sustained increases in the trend of antidepressant sales may also be partially due to an increased prevalence of anxiety disorders, for which long-term treatment with antidepressants is recommended.37 39 Fourth, potential long-term effects of SARS-CoV-2 infection, also termed post-COVID-19 conditions or Long COVID, may have additionally contributed to an increased prevalence of depressive and anxiety disorders.44 45 And last, other concurrent events such as global warming, economic crises and international geopolitical events could have increased the prevalence of depression and anxiety in France. All these reasons can explain the sustained increase in trend of antidepressant sales until the end of 2022 and the lack of an association between antidepressant and anxiolytic sales and stringency.

    Strengths and limitations

    This study presents a detailed analysis of routine healthcare data on antidepressant and anxiolytic consumption over the last decade. The used ITS approach is a strong quasi-experimental design allowing to assess temporal changes caused by impactful events while controlling for prior trends.32 Meanwhile, several limitations need to be considered. First, ITS analyses assume that there are no relevant changes in the underlying population. While it is likely that migration and changes in age distributions have occurred over the observed timeframe, we consider it unlikely that such would have been responsible for the observed trends after pandemic onset. Second, the data set did not include drugs prescribed in the inpatient setting. Despite limitations in inpatient healthcare provision during the pandemic, we did not consider this a relevant issue since the majority of antidepressant and anxiolytic drugs are expected to be prescribed in outpatient settings. Third, the study used DDD/TID as a primary outcome measure, which may not necessarily represent changes in the prevalence of individuals being prescribed the drugs (ie, the prevalence of depression or anxiety disorders) or the number of prescriptions per individual. However, we consider it a strong proxy for the number of prescriptions and the prevalence of corresponding mental health conditions. Fourth, the applied ITS approaches bear several assumptions regarding seasonality, autocorrelation and trends over time. However, we performed various sensitivity analyses, which resulted in similar estimates and conclusions. Finally, our analysis covered the entire population of France and was not designed to evaluate regional or urban–rural differences. While in line with several other studies, it is unclear to what extent the findings are generalisable to other countries in Europe and beyond.

    Clinical implications

    The sustained increase in the trend of antidepressant consumption since the onset of the COVID-19 pandemic in France is concerning and warrants further investigation regarding its specific causes, as it suggests a protracted problem due to increases in the prevalence of depression and anxiety disorders and/or bottlenecks in the provision of non-pharmacological mental health services. This also calls for the attention of policymakers, since urgent public health measures to mitigate the underlying issues may be required. Given that increases in anxiety and depression have also been reported for other human crises,46–48 the findings from this study and others may be valuable for policymakers during future pandemics or similarly disruptive events to anticipate increases in the need of mental health services and to allocate resources accordingly to support the mental health workforce and those affected by mental health conditions.

    Data availability statement

    Data are available in a public, open access repository: https://assurance-maladie.ameli.fr/etudes-et-donnees/open-data-depenses-sante-soins-ville

    Ethics statements

    Patient consent for publication

    Ethics approval

    Not applicable.

    References

    1. ↵
      1. Charlson F,
      2. van Ommeren M,
      3. Flaxman A, et al
      . New WHO prevalence estimates of mental disorders in conflict settings: a systematic review and meta-analysis. Lancet 2019;394:240–8. doi:10.1016/S0140-6736(19)30934-1
      OpenUrlCrossRefPubMed
    2. ↵
      1. Walrave R,
      2. Beerten SG,
      3. Mamouris P, et al
      . Trends in the epidemiology of depression and comorbidities from 2000 to 2019 in Belgium. BMC Prim Care 2022;23:163. doi:10.1186/s12875-022-01769-w
    3. ↵
      1. Dobson KG,
      2. Vigod SN,
      3. Mustard C, et al
      . Trends in the prevalence of depression and anxiety disorders among Canadian working-age adults between 2000 and 2016. Health Rep 2020;31:12–23. doi:10.25318/82-003-x202001200002-eng
      OpenUrlPubMed
    4. ↵
      1. Arroll B,
      2. Elley CR,
      3. Fishman T, et al
      . Antidepressants versus placebo for depression in primary care. Cochrane Database Syst Rev 2009;2009:CD007954. doi:10.1002/14651858.CD007954
    5. ↵
      1. Breilmann J,
      2. Girlanda F,
      3. Guaiana G, et al
      . Benzodiazepines versus placebo for panic disorder in adults. Cochrane Database Syst Rev 2019;3:CD010677. doi:10.1002/14651858.CD010677.pub2
    6. ↵
      1. Alabaku O,
      2. Yang A,
      3. Tharmarajah S, et al
      . Global trends in antidepressant, atypical antipsychotic, and benzodiazepine use: a cross-sectional analysis of 64 countries. PLOS ONE 2023;18:e0284389. doi:10.1371/journal.pone.0284389
    7. ↵
      1. Højlund M,
      2. Gudmundsson LS,
      3. Andersen JH, et al
      . Use of benzodiazepines and benzodiazepine-related drugs in the Nordic countries between 2000 and 2020. Basic Clin Pharmacol Toxicol 2023;132:60–70. doi:10.1111/bcpt.13811
      OpenUrl
    8. ↵
      1. Santomauro DF,
      2. Mantilla Herrera AM,
      3. Shadid J, et al
      . COVID-19 mental disorders collaborators. global prevalence and burden of depressive and anxiety disorders in 204 countries and territories in 2020 due to the COVID-19 pandemic. The Lancet 2021;398:1700–12. doi:10.1016/S0140-6736(21)02143-7
      OpenUrl
    9. ↵
      WHO director-general’s opening remarks at the media briefing. 2023. Available: https://www.who.int/director-general/speeches/detail/who-director-general-s-opening-remarks-at-the-media-briefing---5-may-2023 [Accessed 29 Jan 2024].
    10. ↵
      1. Maguire A,
      2. Kent L,
      3. O’Neill S, et al
      . Impact of the COVID-19 pandemic on psychotropic medication uptake: time-series analysis of a population-wide cohort. Br J Psychiatry 2022;221:748–57. doi:10.1192/bjp.2022.112
      OpenUrl
    11. ↵
      1. González-López MDC,
      2. Díaz-Calvo V,
      3. Ruíz-González C, et al
      . Consumption of psychiatric drugs in primary care during the COVID-19 pandemic. Int J Environ Res Public Health 2022;19:4782. doi:10.3390/ijerph19084782
    12. ↵
      1. Benistand P,
      2. Vorilhon P,
      3. Laporte C, et al
      . Effect of the COVID-19 pandemic on the psychotropic drug consumption. Front Psychiatry 2022;13:1020023. doi:10.3389/fpsyt.2022.1020023
    13. ↵
      1. Frangou S,
      2. Travis-Lumer Y,
      3. Kodesh A, et al
      . Increased incident rates of antidepressant use during the COVID-19 pandemic: interrupted time-series analysis of a nationally representative sample. Psychol Med 2023;53:4943–51. doi:10.1017/S0033291722001891
      OpenUrl
    14. ↵
      1. Tscharke BJ,
      2. Hollingworth S,
      3. van Driel ML, et al
      . The impact of COVID-19 on antidepressant sales and prescription dispensing in Australia. Aust N Z J Psychiatry 2022;56:871–2. doi:10.1177/00048674211068396
      OpenUrl
    15. ↵
      1. Del Fiol F de S,
      2. Bergamaschi C de C,
      3. Lopes LC, et al
      . Sales trends of psychotropic drugs in the COVID-19 pandemic: a national database study in Brazil. Front Pharmacol 2023;14:1131357. doi:10.3389/fphar.2023.1131357
    16. ↵
      1. Tiger M,
      2. Wesselhoeft R,
      3. Karlsson P, et al
      . Utilization of antidepressants, anxiolytics, and hypnotics during the COVID-19 pandemic in Scandinavia. J Affect Disord 2023;323:292–8. doi:10.1016/j.jad.2022.11.068
      OpenUrl
    17. ↵
      1. Bliddal M,
      2. Rasmussen L,
      3. Andersen JH, et al
      . Psychotropic medication use and psychiatric disorders during the COVID-19 pandemic among Danish children, adolescents, and young adults. JAMA Psychiatry 2023;80:176–80. doi:10.1001/jamapsychiatry.2022.4165
      OpenUrl
    18. ↵
      1. Vukićević T,
      2. Draganić P,
      3. Škribulja M, et al
      . Consumption of psychotropic drugs in Croatia before and during the COVID-19 pandemic: a 10-year longitudinal study (2012-2021). Soc Psychiatry Psychiatr Epidemiol 2023. doi:10.1007/s00127-023-02574-1 [Epub ahead of print 17 Oct 2023].
    19. ↵
      1. European Health Observatory on Health Systems and Policies
      . France: health system review 2023. 2023. Available: https://eurohealthobservatory.who.int/publications/i/france-health-system-review-2023 [Accessed 29 Jan 2024].
    20. ↵
      1. Direction de la recherche, des études, de l’évaluation et des statistiques
      . Les Problèmes de Santé Mentale, en augmentation en 2020, Impactent Fortement L’Activité des Généralistes. Available: https://drees.solidarites-sante.gouv.fr/communique-de-presse/les-problemes-de-sante-mentale-en-augmentation-en-2020-impactent-fortement [Accessed 29 Jan 2024].
    21. ↵
      1. Coldefy M,
      2. Gandré C
      . Atlas de la Santé Mentale en France. Ouvrage Irdes, No.7. 2020. Available: https://www.irdes.fr/recherche/2020/ouvrage-007-atlas-de-la-sante-mentale-en-france.html [Accessed 29 Jan 2024].
    22. ↵
      1. L’Assurance Maladie
      . Médicaments Délivrés par LES pharmacies de Ville par Classe ATC - Medic’AM - 2012 À 2022 (Série Labellisée). 2023. Available: https://assurance-maladie.ameli.fr/etudes-et-donnees/medicaments-classe-atc-medicam [Accessed 29 Jan 2024].
    23. ↵
      1. World Health Organisation Collaborating Centre for Drug Statistics Methodology
      . ATC/DDD index. Available: https://www.whocc.no/atc_ddd_index/ [Accessed 29 Jan 2024].
    24. ↵
      1. Rasmussen L,
      2. Wettermark B,
      3. Steinke D, et al
      . Core concepts in pharmacoepidemiology: measures of drug utilization based on individual-level drug dispensing data. Pharmacoepidemiol Drug Saf 2022;31:1015–26. doi:10.1002/pds.5490
      OpenUrl
    25. ↵
      1. Institut national de la statistique et des études économiques (Insee)
      . Population au 1ER Janvier. Available: https://www.insee.fr/fr/statistiques/5225246 [Accessed 29 Jan 2024].
    26. ↵
      1. Bernard Stoecklin S,
      2. Rolland P,
      3. Silue Y, et al
      . First cases of Coronavirus disease 2019 (COVID-19) in France: surveillance, investigations and control measures. Euro Surveill 2020;25:2000094. doi:10.2807/1560-7917.ES.2020.25.6.2000094
    27. ↵
      1. Direction de la recherche, des études, de l’évaluation et des statistiques
      . Confinement Du Printemps 2020: une Hausse des syndromes Dépressifs, surtout Chez LES 15-24 ANS. Résultats Issus de la 1Re vague de L’Enquête Epicov et Comparaison avec LES Enquêtes de Santé Européennes (EHIS) de 2014 et 2019. 2021. Available: https://drees.solidarites-sante.gouv.fr/publications/etudes-et-resultats/confinement-du-printemps-2020-une-hausse-des-syndromes-depressifs [Accessed 29 Jan 2024].
    28. ↵
      1. Hale T,
      2. Angrist N,
      3. Goldszmidt R, et al
      . A global panel database of pandemic policies (Oxford COVID-19 government response Tracker). Nat Hum Behav 2021;5:529–38. doi:10.1038/s41562-021-01079-8
      OpenUrl
    29. ↵
      1. European Medicines Agency (EMA)
      . Etifoxine-containing medicinal products. 2021. Available: https://www.ema.europa.eu/en/medicines/human/referrals/etifoxine-containing-medicinal-products [Accessed 29 Jan 2024].
    30. ↵
      1. Lopez Bernal J,
      2. Soumerai S,
      3. Gasparrini A
      . A methodological framework for model selection in interrupted time series studies. Journal of Clinical Epidemiology 2018;103:82–91. doi:10.1016/j.jclinepi.2018.05.026
      OpenUrlCrossRefPubMed
    31. ↵
      1. Lagarde M
      . How to do (or not to do) … assessing the impact of a policy change with routine longitudinal data. Health Policy Plan 2012;27:76–83. doi:10.1093/heapol/czr004
      OpenUrlCrossRefPubMedWeb of Science
    32. ↵
      1. Bernal JL,
      2. Cummins S,
      3. Gasparrini A
      . Interrupted time series regression for the evaluation of public health interventions: a tutorial. Int J Epidemiol 2017;46:348–55. doi:10.1093/ije/dyw098
      OpenUrlCrossRefPubMed
    33. ↵
      1. Schaffer AL,
      2. Dobbins TA,
      3. Pearson S-A
      . Interrupted time series analysis using autoregressive integrated moving average (ARIMA) models: a guide for evaluating large-scale health interventions. BMC Med Res Methodol 2021;21:58. doi:10.1186/s12874-021-01235-8
    34. ↵
      1. Muff S,
      2. Nilsen EB,
      3. O’Hara RB, et al
      . Rewriting results sections in the language of evidence. Trends Ecol Evol 2022;37:203–10. doi:10.1016/j.tree.2021.10.009
      OpenUrl
    35. ↵
      1. Léon C,
      2. du Roscoät E,
      3. Beck F
      . Prévalence des Épisodes Dépressifs en France Chez LES 18-85 ANS: Résultats Du Baromètre Santé 2021. Santé Publique FR. 2023. Available: https://www.santepubliquefrance.fr/import/prevalence-des-episodes-depressifs-en-france-chez-les-18-85-ans-resultats-du-barometre-sante-2021 [Accessed 29 Jan 2024].
    36. ↵
      1. Haute Authorité de Santé
      . Épisode Dépressif Caractérisé de L’Adulte: Prise en charge en Premier Recours. Available: https://www.has-sante.fr/jcms/c_1739917/fr/episode-depressif-caracterise-de-l-adulte-prise-en-charge-en-premier-recours [Accessed 29 Jan 2024].
    37. ↵
      1. Caisse Primaire d’Assurance Maladie
      . Traitement des troubles Anxieux. 2022. Available: https://www.ameli.fr/assure/sante/themes/troubles-anxieux-anxiete/traitement [Accessed 29 Jan 2024].
    38. ↵
      1. Beach SR,
      2. Kostis WJ,
      3. Celano CM, et al
      . Meta-analysis of selective serotonin reuptake inhibitor-associated QTc prolongation. J Clin Psychiatry 2014;75:e441–9. doi:10.4088/JCP.13r08672
      OpenUrlCrossRefPubMed
    39. ↵
      1. National Institute for Health and Care Excellence (NICE)
      . Generalised anxiety disorder and panic disorder in adults: management – guidance. 2020. Available: https://www.nice.org.uk/guidance/cg113 [Accessed 29 Jan 2024].
    40. ↵
      1. Estrela M,
      2. Silva TM,
      3. Gomes ER, et al
      . Prescription of anxiolytics, sedatives, hypnotics and antidepressants in outpatient, universal care during the COVID-19 pandemic in Portugal: a nationwide, interrupted time-series approach. J Epidemiol Community Health 2022;76:335–40. doi:10.1136/jech-2021-216732
      OpenUrlAbstract/FREE Full Text
    41. ↵
      1. Domenghino A,
      2. Aschmann HE,
      3. Ballouz T, et al
      . Mental health of individuals infected with SARS-Cov-2 during mandated isolation and compliance with recommendations—a population-based cohort study. PLOS ONE 2022;17:e0264655. doi:10.1371/journal.pone.0264655
    42. ↵
      1. Coulaud P-J,
      2. Jesson J,
      3. Bolduc N, et al
      . Association between level of compliance with COVID-19 public health measures and depressive symptoms: a cross-sectional survey of young adults in Canada and France. PLOS ONE 2023;18:e0289547. doi:10.1371/journal.pone.0289547
    43. ↵
      1. Kindred R,
      2. Bates GW
      . The influence of the COVID-19 pandemic on social anxiety: a systematic review. Int J Environ Res Public Health 2023;20:2362. doi:10.3390/ijerph20032362
    44. ↵
      1. Taquet M,
      2. Sillett R,
      3. Zhu L, et al
      . Neurological and psychiatric risk Trajectories after SARS-CoV-2 infection: an analysis of 2-year retrospective cohort studies including 1 284 437 patients. Lancet Psychiatry 2022;9:815–27. doi:10.1016/S2215-0366(22)00260-7
      OpenUrl
    45. ↵
      1. Ballouz T,
      2. Menges D,
      3. Anagnostopoulos A, et al
      . Recovery and symptom trajectories up to two years after SARS-CoV-2 infection: population based, longitudinal cohort study. BMJ 2023;381:e074425. doi:10.1136/bmj-2022-074425
    46. ↵
      1. van Griensven F,
      2. Chakkraband MLS,
      3. Thienkrua W, et al
      . Mental health problems among adults in tsunami-affected areas in Southern Thailand. JAMA 2006;296:537–48. doi:10.1001/jama.296.5.537
      OpenUrlCrossRefPubMedWeb of Science
    47. ↵
      1. Usher K,
      2. Brown LH,
      3. Buettner P, et al
      . Rate of prescription of antidepressant and anxiolytic drugs after cyclone Yasi in North Queensland. Prehosp Disaster Med 2012;27:519–23. doi:10.1017/S1049023X12001392
      OpenUrlPubMed
    48. ↵
      1. Chang S-S,
      2. Stuckler D,
      3. Yip P, et al
      . Impact of 2008 global economic crisis on suicide: time trend study in 54 countries. BMJ 2013;347:f5239. doi:10.1136/bmj.f5239

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Footnotes

    • Contributors Conceptualisation: DDB, DM; methodology: DDB, SRH, DM; validation: DDB, DM; formal analysis: DM; investigation: DDB, SRH, LD, BB, DM; data curation: DDB, DM; visualisation: DDB, SRH, DM; project administration: DDB, DM; writing—original draft: DDB, DM; writing—review and editing: DDB, SRH, LD, BB, DM. DDB and DM act as guarantors for the study.

    • Funding This research project did not receive any study-specific funding. DM received salary funding by the University of Zurich Postdoc Grant, grant no FK-22–053, via the University of Zurich (Switzerland). DDB and LD received salary funding as Chef de Clinique en Médecine Générale (CCUMG), and BB as Maître de Conférence Universitaire en Médecine Générale (MCUMG), from the University of Versailles-Saint-Quentin-en-Yvelines (France). DDB, LB and BB received fees through their work as a general practitioner. DDB is guarantor

    • Competing interests None declared.

    • Provenance and peer review Not commissioned; internally peer-reviewed.

    • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.