Opioid-containing antitussives in Switzerland: a descriptive cross-sectional time-series analysis of pharmacy sales 2013–2022

DOI: https://doi.org/https://doi.org/10.57187/s.4188

Introduction

Acute respiratory tract infections, including acute coughs, are among the most frequent conditions pharmacists encounter in their daily practice [1]. Medications containing codeine, codeine derivatives or dextromethorphan, hereinafter “opioid-containing antitussives”, are used to symptomatically treat a dry, irritative cough – despite a relative lack of evidence of their efficacy [2].

In the late 2000s, concerns started emerging in the USA about the misuse of opioid-containing antitussives as recreational drugs by teenagers, a trend which subsequently spread to Europe, including Switzerland [3, 4]. A survey conducted in 2018 among 1180 Swiss adolescents found that one in six 20-year-olds had used codeine cough syrup recreationally [5], a number which might even be an underestimate, as found in later hair analyses of the survey participants [6]. Risks of codeine misuse include respiratory depression and dependence. Dextromethorphan overdose can lead to seizures and coma [7, 8].

In 2019, the Swiss Therapeutic Products Act was updated and included a revision of the Swiss drug dispensing categories. All medications had previously been classified into categories A–E based on the required dispensing authority, with category A being the most restrictive (single dispensing requiring a medical prescription) and category E being the least restrictive (over-the-counter [OTC] without medical advice). Category C (medications dispensed by pharmacists) was abolished in the new classification, and all medications in this category underwent a risk-based reclassification to either category D (over-the-counter dispensing with specialist advice) or category B (dispensing requiring a medical prescription). More specifically, drugs reclassified from category C to category B received a special status, in that these can be dispensed by pharmacists after a documented consultation. Opioid-containing antitussives, which were previously in category C, fell into this so-called B+ dispensing category due to their risk of misuse [9, 10].

Despite regular media attention and this significant regulation change, an investigation of the utilisation of opioid-containing antitussives in Switzerland has not been conducted to date. This study aims to close this gap by describing the sales trends of opioid-containing antitussives in Switzerland between 2011 and 2022.

Methods

Study design and data sources

We conducted a descriptive interrupted time-series analysis using repeated cross-sectional data on drug sales in Switzerland. Data was derived from the IQVIA MIDAS^® pharmaceutical sales database for the period between quarter 1 (Q1) 2013 and Q3 2022. IQVIA MIDAS^® is a dataset that tracks sales volumes and values across 94 countries and over 1.6 million drug products, reflecting estimates of real-world activity. This includes hospital and retail pharmacy channels but does not include information about individual facilities. For our analysis, drug sales within the retail channel, which includes pharmacies and self-dispensing physicians (i.e. physicians who own their own dispensary), were used as an indicator of community use. This data included the year, calendar quarter (Q), Anatomical Therapeutic Chemical (ATC) code of the medication and the total number of standardised units sold. A standardised drug unit is defined as a single tablet/capsule, patch, vial or 5 ml of oral liquid. The data exhibited no missing values. To validate the data, we performed a cross-check of the included products and the products approved and distributed on the Swiss market and ensured that sales numbers were within a reasonable range.

We included all sales and variables for the following ATC codes in the category R05 cough and cold preparations: R05DA04 codeine, R05DA09 dextromethorphan, R05FA02 opium derivatives and expectorants. Products under the ATC code R05FA02 containing the non-narcotic opioid noscapine were excluded from our analysis as their regulatory status did not change. The ATC code R05DA20 opium alkaloids and derivatives was not included as it included only one product and would thus reveal information outside the scope of the licensing agreement with IQVIA MIDAS. While dextromethorphan-containing antitussives do not primarily act at opioid receptors, we included products under R05DA09 dextromethorphan in our analysis as they pose the same concerns regarding misuse as opioid-containing antitussives and were thus treated equally to codeine and codeine derivatives and combination products during the regulation change of 2019 [7, 11]. Combination cold medicines containing dextromethorphan are classified under the ATC code N02BE51 Paracetamol, combinations excl. psycholeptics and were outside the scope of the regulation change and of our analysis. In a post hoc analysis, the ATC code R05DB13 butamirate was added to the analysis to serve as a control, as butamirate-containing products were not affected by the regulation change.

Statistical analysis

We analysed quarterly sales data from Q1 2013 to Q3 2022. The quarterly sales were standardised to the Swiss population (rate per 100,000). Swiss population statistics for the years 2013 to 2022 were obtained from the Swiss Federal Statistical Office [12].

We conducted a descriptive time-series analysis to plot and visualise the development of quarterly sales of antitussives in Switzerland between Q1 2013 and Q3 2022. To assess time trends, the time series was decomposed with Seasonal and Trend decomposition using LOESS, resulting in a trend component which excludes the seasonal fluctuation and random noise from the data.

To assess fluctuations in sales, annual sales rates were calculated as the sum of the sales rates across all four quarters in the year for years with complete data (2013–2021). The percent change in sales rate was determined by comparing it to the previous year.

All analyses were conducted for the cumulative sales of all ATC codes and stratified by ATC code. To visualise the timing of the reclassification, we defined a range for each ATC class given that the roll-out of the reclassification occurred throughout 2019 and the date of upregulation for each antitussive medication varied (see appendix table S1). Additionally, we included the start and end of the COVID-19 pandemic in Switzerland as previous studies have shown the impact of COVID-19 on medication utilisation [13]. The COVID-19 pandemic start and end dates were determined as 28 February 2020 and 31 March 2022, respectively, in accordance with the proclamation and end of the “special situation” according to Article 6 of the Swiss Epidemics Act [14] as communicated in press releases by the Swiss Federal Council [15, 16].

No protocol has been preregistered for this descriptive study.

Statistical analyses were performed using R statistical software (version 4.0.4) [17]: data was plotted using ggplot2 [18] and the Seasonal and Trend decomposition using LOESS was performed with forecast [19].

Ethics approval

This study did not use any individual participant or patient data, and therefore is exempt from research ethics review. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines [20].

Results

Between Q1 2013 and Q3 2022, an estimated 324 million standard units of opioid-containing antitussives were sold. The majority of sales were for preparations containing dextromethorphan (an estimated 191 million standard units, 59%), followed by codeine preparations (27%), and opioid and expectorant combination products (14%) (see table 1).

The quarterly sales rate trends are plotted in figure 1 for the overall and individual ATC classes (appendix table S3 provides the corresponding rates). While there was variation in the sales rate, the decomposed trend line shows a sharp decrease in all sales (figure 1) at the end of 2019; however, the sales rate had nearly returned to baseline by the end of 2021. The sales of butamirate-containing antitussives showed a very similar trend as opioid-containing antitussives between 2020 and 2022; however, butamirate-containing antitussive sales increased further after 2022. The sales rates, decomposed trend and the corresponding graph can be found in appendix table S4 and appendix figure S2.

Figure 1Antitussive sales trends per substance class. The sales are units sold per 100,000 population. The green line shows the raw data, while the black line shows the trend component remaining after time-series decomposition with LOESS. Graphs including the seasonal and remainder components and the underlying data can be found in appendix figure S1. The period in which the products in the respective category were upregulated is marked in gray; the period of the COVID-19 pandemic is marked in orange. Author analysis based on IQVIA MIDAS^® quarterly volume sales data for Q1 2013 – Q3 2022, reflecting estimates of real-world activity. Copyright IQVIA. All rights reserved.

Table 2 provides the annual sales rate per 100,000 for years with complete data (2013–2021) and the percent change vs the previous year. Overall, antitussive sales fluctuated between 2012 and 2019; however, sales decreased substantially in 2020. In 2020, 30.5% less opioid-containing antitussives were sold than in 2019. A similar decline in 2020 was seen for all individual ATC codes with dextromethorphan sales decreasing the least (−27.8%) and opioid and expectorant sales decreasing the most (−39.9%) compared to the previous year.

Discussion

While the sales of opioid-containing antitussive medications were slowly decreasing prior to the reclassification that occurred throughout 2019, the sales rates decreased substantially in Q4 of 2019. Although we observed a decrease prior to the start of the COVID-19 pandemic in Switzerland, the substantial decrease in sales in 2020 is likely largely due to the effects of the pandemic rather than the reclassification.

Having only been completed in late 2019, the effect of the reclassification of opioid-containing antitussives on the 2019 sales rates is difficult to discern given the proximity to the start of the COVID-19 pandemic. However, we do note that the sales rate decreased by 3.4% in 2019 compared to 2018, and the quarterly data shows that most opioid-containing antitussives have not returned to 2018 levels. This may indicate that the reclassification has slightly curbed sales, particularly for codeine-containing antitussives.

The sales of antitussives containing dextromethorphan, or opioids combined with expectorants appear to be gradually increasing post-pandemic; however codeine-containing antitussives only saw a partial recovery of the sales rate. At the end of the study period in Q3 2022, sales were 37.2% lower than in Q4 of 2018. In 2015, the European Medicines Agency issued a contraindication for the use of codeine-containing antitussives in children younger than 12 years, as well as in patients with known ultrarapid metaboliser status in the CYP2D6 enzyme [21]. This contraindication was adopted in Switzerland in 2017 [22]. Therefore, it is also possible that combination products containing additional active ingredients for the symptomatic treatment of acute respiratory infections (e.g. butamirate) are increasingly favoured over codeine monopreparations with the aim of improving treatment efficacy, given that the antitussive properties of codeine do not seem to outperform placebo [23]. However, concerns and additional restrictions placed on codeine use have little impact on the sales of all opioid-containing antitussives cumulatively as dextromethorphan remains the most popularly used opioid-containing antitussive in Switzerland. While dextromethorphan does not act at opioid receptors, its misuse potential is similar to that of codeine [24] and dextromethorphan monopreparations have thus undergone the same regulatory changes as codeine-containing antitussives [11]. The media coverage on antitussive misuse and the ensuing regulation change in Switzerland was, however, mainly focused on codeine-containing antitussives [25–28]. This might suggest that the revision of the Therapeutic Products Act had a larger impact on codeine-containing than on dextromethorphan-containing antitussives. On the other hand, calls to the Swiss poisons information centre Tox Info Suisse for dextromethorphan poisonings have been decreasing since 2017 [29]. As data after 2020 is not available from this source, it remains unclear how the poisoning trends have been impacted by or evolved since the COVID-19 pandemic.

However, as identified, the estimation of the effect of the Therapeutic Products Act revision on the sales of opioid-containing antitussives is complicated by the start of the COVID-19 pandemic in February 2020, which affected drug sales through various mechanisms [13]. Firstly, people who contracted COVID-19 were obliged to isolate at home and were thus unlikely to seek treatment at a pharmacy. Therefore, as opioid-containing antitussives under the new Therapeutic Products Act can only be dispensed to the patient personally, non-opioid antitussives (from dispensing category D) would have been favoured when sold to caregivers. Secondly, due to the protective measures taken during the pandemic, other respiratory tract infections such as seasonal influenza had a very low incidence in the 2020/2021 season [30], likely also minimising the necessity for treating cough from respiratory tract infections other than COVID-19 and decreasing sales of opioid-containing antitussives. While the pandemic broadly affected supply chains for drugs which led to shortages, the effect on opioid-containing antitussives specifically was relatively small [31]. Lastly, a study by Gordon and colleagues published in March 2020 found that dextromethorphan increased the titres of SARS-CoV-2 in vitro [32]. The authors, and subsequently several news outlets, called for caution in the use of dextromethorphan-containing antitussives for the symptomatic treatment of COVID-19 [33–35]. To what extent this affected the sales of dextromethorphan-containing antitussives is unclear; however, in our study dextromethorphan-containing antitussives saw a smaller decrease (−27.8%) in 2020 compared to 2019 than codeine-containing antitussives (−31.7%) and opioid and expectorants (−39.9%). To better distinguish the effects of COVID-19 from those of the revision, we additionally analysed the sales trends of butamirate-containing antitussives which were not reclassified and thus unaffected by the regulation. To date, butamirate-containing antitussives are still available over-the-counter [36]. We found that butamirate-containing antitussive sales trends developed very similarly to the sales trends of opioid-containing antitussives during the COVID-19 pandemic in that there was a sharp drop during 2020 followed by an increase in 2021. However, while the sales of opioid-containing antitussives merely returned to pre-upregulation levels (Q4 2018 vs Q3 2022: −3.8%), the sales of butamirate-containing antitussives were 40.6% higher at the end of our observation period than in Q4 of 2018. On the one hand, this suggests that the drop in sales of opioid-containing antitussives in 2020 was indeed mainly driven by the circumstances of the pandemic; on the other hand, the following rise in butamirate sales might be a consequence of fewer sales of codeine-containing antitussives, assuming the overall demand for antitussives was not impacted long-term by the pandemic or the upregulation.

To date, this is the first analysis of opioid-containing antitussive sales in Switzerland. This study has notable strengths. The Swiss IQVIA MIDAS^® data covers pharmacies, self-dispensing physicians and drugstores; the coverage in Switzerland is estimated at 100% [37]. All sales and changes for pharmaceutical specialties to pharmacies, physicians’ practices and drugstores are recorded for the listed drugs. However, our database only gives information on standardised units sold and does not inform on the number of patients who acquired opioid-containing antitussives, the number of prescribers or dispensers, or their characteristics. Additionally, a comparison to other countries is difficult due to the variety of regulatory settings. While France experienced a similar upregulation of codeine-containing products from over-the-counter to prescription-only in 2017 [38], this affected not only antitussives but also codeine-containing analgesic products [39]. In France the upregulation led to a drop in overall codeine sales; however, we were unable to find any published data on the impact on antitussives specifically or on the dispensing of codeine products during the COVID-19 pandemic to compare it to our analysis [40]. Nevertheless, our results provide insight into the general trend observed in opioid-containing antitussive sales in Switzerland from Q1 2013 to Q3 2022.

Cumulatively, a decreasing trend was observed for the sales of opioid-containing antitussives between 2013 and 2019. It is likely that repeated media attention on cases of antitussive misuse has led to more cautious recommendation of opioid-containing antitussives in pharmacies, especially in a younger population. It does not appear that the revision of the Swiss Therapeutic Products Act in 2019 led to a further decrease in the sales of opioid-containing antitussives. However, as the COVID-19 pandemic occurred shortly following revision, longer data collection will be needed to comprehensively assess its effects on antitussive utilisation going forward.

Data availability statement

The IQVIA MIDAS^® pharmaceutical sales data analysed for this study was licensed from IQVIA. Underlying raw data will not be disclosed, only the data points set out in this paper. The code used in the analysis can be shared upon request to the corresponding author.

Acknowledgments

Data from this study was provided by the Leslie Dan Faculty of Pharmacy at the University of Toronto. The statements, findings, conclusions, views and opinions contained and expressed in this research article are based in part on data obtained under license from the following IQVIA Solutions Canada Inc. information service: IQVIA MIDAS^®. Copyright IQVIA. All Rights Reserved. The statements, findings, conclusions, views and opinions contained and expressed herein are not necessarily those of IQVIA Solutions Canada Inc. or any of its affiliated or subsidiary entities.

References

1. Eickhoff C, Hämmerlein A, Griese N, Schulz M. Nature and frequency of drug-related problems in self-medication (over-the-counter drugs) in daily community pharmacy practice in Germany. Pharmacoepidemiol Drug Saf. 2012 Mar;21(3):254–60. 10.1002/pds.2241 

2. Krüger K, Gehrke-Beck S, Holzinger F, Heintze C. S3-Leitlinie Akuter und chronischer Husten. AWMF online. 2021. Available from: https://register.awmf.org/de/leitlinien/detail/053-013 

3. Burns JM, Boyer EW. Antitussives and substance abuse. Subst Abuse Rehabil. 2013 Nov;4:75–82. 10.2147/SAR.S36761 

4. Bachmann A, Galgano L, Guillaume M. Jugendliche mit Medikamenten- und Mischkonsum: Situations- und Bedarfsanalyse. Infodrog. 2022. Available from: https://www.infodrog.ch/files/content/ff-de/Bericht_Jugendliche%20Mischkonsum%20Situationsanalsyse%20und%20Empfehlungen.pdf 

5. Quednow BB, Steinhoff A, Bechtiger L, Ribeaud D, Eisner M, Shanahan L. High Prevalence and Early Onsets: Legal and Illegal Substance Use in an Urban Cohort of Young Adults in Switzerland. Eur Addict Res. 2022;28(3):186–98. 10.1159/000520178 

6. Steinhoff A, Shanahan L, Bechtiger L, Zimmermann J, Ribeaud D, Eisner MP, et al. When Substance Use Is Underreported: Comparing Self-Reports and Hair Toxicology in an Urban Cohort of Young Adults. J Am Acad Child Adolesc Psychiatry. 2023 Jul;62(7):791–804. 10.1016/j.jaac.2022.11.011 

7. Journey JD, Agrawal S, Stern E. Dextromethorphan Toxicity. In: StatPearls. Treasure Island (FL): StatPearls Publishing; 2023., Available from https://www.ncbi.nlm.nih.gov/books/NBK538502/ 

8. Weier M, Weier N, O’Mara B. Over-the-Counter Medications and Their Misuse: A Focus on Codeine. In: Patel VB, Preedy VR, editors. Handbook of Substance Misuse and Addictions. Cham: Springer International Publishing; 2022. pp. 1–23. 

9. Die Bundesversammlung der Schweizerischen Eidgenossenschaft. Bundesgesetz über Arzneimittel und Medizinprodukte. 2002. Available from: https://www.fedlex.admin.ch/eli/cc/2001/422/de 

10. Bundesamt für Gesundheit Erleichterte Abgabe von Arzneimitteln der Liste B. https://www.bag.admin.ch/bag/de/home/medizin-und-forschung/heilmittel/abgabe-von-arzneimitteln.html 

11. Swissmedic. Revision des Heilmittelrechts – Liste der von der Abgabekategorie C in die Abgabekategorie B umgeteilten Arzneimittel. Swissmedic 2019. Available from: https://www.swissmedic.ch/swissmedic/de/home/news/mitteilungen/liste_abgabekategorie_c_abgabekategorie_bumgeteilten_am.html 

12. Bundesamt für Statistik. Struktur der ständigen Wohnbevölkerung nach Kanton, 1999-2022. BAS. 2023. Available from: https://www.bfs.admin.ch/asset/de/26565154 

13. Suda KJ, Kim KC, Hernandez I, Gellad WF, Rothenberger S, Campbell A, et al. The global impact of COVID-19 on drug purchases: A cross-sectional time series analysis. J Am Pharm Assoc (Wash DC). 2022;62(3):766–774.e6. 10.1016/j.japh.2021.12.014 

14. Bundesrat. Bundesgesetz über die Bekämpfung übertragbarer Krankheiten des Menschen. 2012. Available from: https://www.fedlex.admin.ch/eli/cc/2015/297/de 

15. Bundesrat. Coronavirus: Bundesrat verbietet grosse Veranstaltungen. 2020. Available from: https://www.admin.ch/gov/de/start/dokumentation/medienmitteilungen.msg-id-78289.html 

16. Bundesrat. Coronavirus: Rückkehr in die normale Lage und Planung der Übergangsphase bis Frühling 2023. 2022. Available from: https://www.admin.ch/gov/de/start/dokumentation/medienmitteilungen.msg-id-87801.html 

17. R Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2022. 

18. Wickham H. ggplot2: Elegant Graphics for Data Analysis. New York: Springer; 2016. 

19. Hyndman RJ, Khandakar Y. Automatic time series forecasting: the forecast package for {R}. J Stat Softw. 2008;27(3):1–22. 10.18637/jss.v027.i03 

20. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP; STROBE Initiative. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet. 2007 Oct;370(9596):1453–7. 10.1016/S0140-6736(07)61602-X 

21. European Medicines Agency. Codeine not to be used in children below 12 years for cough and cold. 2015. Available from: https://www.ema.europa.eu/en/news/codeine-not-be-used-children-below-12-years-cough-and-cold 

22. Swissmedic. Husten- und Erkältungsmittel mit Codein bzw. Dihydrocodein: Anpassungen der Arzneimittelinformationen. 2017. Available from: https://www.swissmedic.ch/swissmedic/de/home/humanarzneimittel/marktueberwachung/health-professional-communication--hpc-/archiv/hpc-husten-und-erkaeltungsmittel-mit-codein-bzw-dihydrocodein.html 

23. Morice A, Kardos P. Comprehensive evidence-based review on European antitussives. BMJ Open Respir Res. 2016 Aug;3(1):e000137. 10.1136/bmjresp-2016-000137 

24. Oh S, Agrawal S, Sabir S, Taylor A. Dextromethorphan. In: StatPearls. Treasure Island (FL): StatPearls Publishing; 2024., Available from https://www.ncbi.nlm.nih.gov/books/NBK538216/ 

25. Schnyder M. Medikamenten-Missbrauch - Hustensirup mit Codein: Was gilt nun ab Januar? In: Schweiz. Radio Fernseh. 2018. Available from: https://www.srf.ch/sendungen/kassensturz-espresso/medikamenten-missbrauch-hustensirup-mit-codein-was-gilt-nun-ab-januar 

26. Wermelinger R. Medikamenten-Missbrauch - Hustensaft als Partydroge: Kommt die Rezeptpflicht? In: Schweiz. Radio Fernseh. 2018. Available from: https://www.srf.ch/news/schweiz/medikamenten-missbrauch-hustensaft-als-partydroge-kommt-die-rezeptpflicht 

27. Roth S. Missbrauch: Mit Hustensirup zum Rausch. In: St Galler Tagblatt. 2018. Available from: https://www.tagblatt.ch/ostschweiz/appenzellerland/missbrauch-mit-hustensirup-zum-rausch-ld.606519 

28. 20 Minuten. Jugendliche dröhnen sich mit Hustensaft zu. In: 20 Minuten. 2018. Available from: https://www.20min.ch/story/jugendliche-droehnen-sich-mit-hustensaft-zu-190224395258 

29. Tox Info Suisse. Jahresbericht 2020. Availabe from: https://www.toxinfo.ch/customer/files/878/9211581_Tox_JB-2020_DE_Web.pdf 

30. Bundesamt für Gesundheit BAG. Infoportal übertragbare Krankheiten: Influenza (Grippe). Available from: https://idd.bag.admin.ch/diseases/influenza/statistic 

31. drugshortage.ch. Lieferengpässe von Medikamenten - Abgeschlossen. Available from: https://www.drugshortage.ch/index.php/abgeschlossen/ 

32. Gordon DE, Jang GM, Bouhaddou M, Xu J, Obernier K, White KM, et al. A SARS-CoV-2 protein interaction map reveals targets for drug repurposing. Nature. 2020 Jul;583(7816):459–68. 10.1038/s41586-020-2286-9 

33. Healy M. Why you should avoid some cough syrups if you think you’ve got the coronavirus. In: Los Angel. Times. 2020. Available from: https://www.latimes.com/science/story/2020-04-30/why-you-should-avoid-cough-syrup-if-you-think-youve-got-the-coronavirus 

34. Zimmer C. Old Drugs May Find a New Purpose: Fighting the Coronavirus. In: N. Y. Times. 2020. Available from: https://www.nytimes.com/2020/04/30/health/coronavirus-antiviral-drugs.html 

35. Hüttemann D. Nature-Studie: Neue Kandidaten gegen Coronaviren und Warnung vor Dextromethorphan. In: Pharm. Ztg. Online. 2020. Available from: https://www.pharmazeutische-zeitung.de/neue-kandidaten-gegen-coronaviren-und-warnung-vor-dextromethorphan-117321/ 

36. HCI Solutions AG Compendium. R05DB13 Butamirat. Available from: https://compendium.ch/de/register/atc/R05DB13 

37. Ju C, Wei L, Man KK, Wang Z, Ma TT, Chan AY, et al. Global, regional, and national trends in opioid analgesic consumption from 2015 to 2019: a longitudinal study. Lancet Public Health. 2022 Apr;7(4):e335–46. 10.1016/S2468-2667(22)00013-5 

38. Ministère du Travail de la Santé et des Solidarités. Agnès Buzyn décide d’inscrire la codéine et d’autres dérivés de l’opium à la liste des médicaments disponibles uniquement sur ordonnance. 2017. Available from: https://sante.gouv.fr/archives/archives-presse/archives-communiques-de-presse/article/agnes-buzyn-decide-d-inscrire-la-codeine-et-d-autres-derives-de-l-opium-a-la 

39. Natali I. Economic Opportunity and Opioid Regulation: the Case of Codeine in France. Toulouse School of Economics (TSE). 2024. Working paper 24-1563. Available from: https://www.tse-fr.eu/publications/economic-opportunity-and-opioid-regulation-case-codeine-france 

40. Richards GC, Aronson JK, MacKenna B, Goldacre B, Hobbs FD, Heneghan C. Sales of Over-the-Counter Products Containing Codeine in 31 Countries, 2013-2019: A Retrospective Observational Study. Drug Saf. 2022 Mar;45(3):237–47. 10.1007/s40264-021-01143-2 

Appendix

The appendix is available in the pdf version of the article at https://doi.org/10.57187/s.4188.