https://orcid.org/0000-0003-4120-1258
Figures
Abstract
Background
Antimicrobial resistance in community-associated infections is an increasing worldwide concern. In low-to-middle income countries, over-the-counter (OTC) sales of antimicrobials without medical prescription have been blamed for increasing consumption and resistance. We studied the impact of restriction of OTC sales of antimicrobials in Brazil (instituted in October 2010) on resistance trends of Escherichia coli from community-onset urinary tract infections.
Methods
We analyzed monthly resistance trend of Escherichia coli from community-onset urinary tract infections from 2005 through 2018. The data were submitted to interrupted time series analysis in both linear and Poisson regression models.
Citation: Rodrigues FS, Amat HRA, Fortaleza CMCB (2021) Impact of restriction of over-the-counter sales of antimicrobials on antimicrobial resistance in Escherichia coli from community-onset urinary tract infections in inner São Paulo State, Brazil. PLoS ONE 16(11): e0259632. https://doi.org/10.1371/journal.pone.0259632
Editor: Monica Cartelle Gestal, University of Georgia, UNITED STATES
Received: January 4, 2021; Accepted: October 22, 2021; Published: November 10, 2021
Copyright: © 2021 Rodrigues et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: Study database is included as a supplementary file.
Funding: HRAA received a student research grant from São Paulo State Research Foundation (FAPESP, grant #2018/17210-1), with CMCBF as her advisor. The funding agency did not played no role in study design, methodology, nor in the writing of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
List of abbreviations: COUTI, community-onset urinary tract infections; ITS, interrupted time series (analysis); SMX-TMP, sulfamethoxazole-trimethoprim
Background
In the past decade, the World Health Organization (WHO) recognized antimicrobial resistance as a major threat for public health worldwide. Countries were urged to develop their own action plans to prevent and control the emergence and spread of antimicrobial-resistant pathogens [1, 2]. This is a major challenge for low-to-middle income countries, which have experienced recent increases in consumption of antimicrobials [3]. In those countries, over-the-counter (OTC) sales account for a relevant proportion of excessive use of antimicrobial agents [4, 5]. Laws restricting or prohibiting OTC sales of antimicrobials have been issued, but their impact is currently not straightforward [6, 7].
The population drivers of antimicrobial resistance are complex. A vast array of direct and herd effects interplay in the continuous process of emergence and spread of resistant bacterial strains [8]. Even though secular trends demonstrate clear association between the introduction of antimicrobials in clinical practice and the emergence and spread of resistance phenotypes, this association is not straightforward [9]. Therefore, the impact of regulations must be assessed not only regarding antimicrobial sales, but also focusing on their outcomes of interest, i.e., on the time trends of resistance among common pathogens.
A ban on non-prescription sales of antimicrobials in private drugstores was issued by Brazilian Sanitary Agency (ANVISA) in October 2010. The impact of this legal measure was uneven across the country. In fact, there was a significant reduction in sales of antimicrobials in pharmacies in the Southern and Southeastern states, which have a higher human development index (as compared to Northern, Northeastern and Midwest, and corresponds to the study zone. The effect was small in the less developed areas [10, 11]. Most importantly, the impact on antimicrobial resistance is uncertain in Brazil and in our region.
With that in mind, we use time series analysis to investigate the impact of ANVISA regulation on antimicrobial resistance in Escherichia coli isolated in urine from patients with community-onset urinary tract infection (COUTI) in inner São Paulo State, Brazil.
Methods
Ethical approval
This study was approved by the Committee for Ethics in Human Research in Botucatu Medical School (register: CAAE 30000014.1.0000.5411), city of Botucatu, São Paulo State, Brazil. The ethics committee waived the requirement for informed consent. Data from all patients was initially checked for duplications, and subsequently anonymized before final validation and statistical analyses.
Study setting and design
We conducted a natural experiment study based on ecological data on incidence and antimicrobial resistance among E. coli isolates from COUTI in patients living in the Botucatu Health Administrative Area (BHAA), inner São Paulo State, Brazil. That area comprises 13 municipalities located around Botucatu city (22°53′25″S, 48°27′19″W), with a total of 300,000 inhabitants. Microbiological tests from public primary care units and outpatient services are performed in the laboratory from Botucatu Medical School teaching hospital.
Operational procedures
We collected data on microbiological results from COUTI in the period from January 2005 through December 2018. Those cultures were obtained from patients with suspected COUTI, cared for in public primary care and outpatient facilities in the BHAA. We recorded resistance of E. coli isolates to several antimicrobials, mainly the most used for the treatment of COUTI in our country: cefazolin, ceftriaxone, ciprofloxacin, amikacin and sulfamethoxazole-trimethoprim (SMX-TMP). We chose to address those antibiotics based on their use in outpatients or inpatients therapy of COUTI. Antimicrobial resistant tests were performed in Microscan Autoscan system (Beckman Coulter, Pasadena, CA) and followed standards from the Clinical Laboratory Standards Institute (CSLI) [12]. Duplicate results (defined as those obtained within a 30-day period) and those in which a nosocomial origin was proved were excluded from the study.
Time series analysis
Time series of monthly incidence of pathogens and resistance patterns were submitted to interrupted time series analysis (ITS) through segmented regression in the itsa package of STATA 14 (Statacorp, College Station, TX), as previously described [13]. Briefly, we analyzed the impact of the OTC sales ban in October 2010 on the immediate and late trends of antimicrobial resistance. We complemented this analysis with Poisson Regression models, adjusted for time trends and seasons.
In order to further investigate seasonality, Box-Jenkins models were fit to time series of incidence and antimicrobial resistance. We used the following parameters for models: regular (1,0,1), seasonal (1,0,1), with seasonality in 12 months [14]. Those analyses were performed in NCSS 9 (LLC, Kaysville, UT).
Results
In the study period, there were 40,814 positive cultures from COUTI patients, of which 31,162 (76.4%) recovered E. coli. The cumulative incidence was 10,686 per 100,000 inhabitants. There was a continuous trend towards increase in the isolates (incidence rate ratio [IRR] for month, 1.004; 95% confidence interval, 1.003–1.005; P<0.001), with peaks in summer (IRR, 1.063; 95% 1.031–1.097; P<0.001, as compared to winter). Those findings imply an average increasing trend in rates of 0.4% per month, as well as an overall increase of 6.3% during summer. While the overall increase may be attributable to greater access to diagnostic tests, the seasonal findings are consistent and were corrected excluding long term trends. Coherently, we found significant coefficients (P<0.001) for autoregressive, moving average and–most importantly—seasonal parameters in Box Jenkins models.
Results from ITS analysis of resistance trends are presented in Fig 1 and Table 1 presents complementary season-adjusted Poisson regression models. Overall, immediate decrease in resistance was detected in all models only for cefazolin and amikacin. No beneficial impact on long-term trends was identified for the resistance to antimicrobials of interest.
TMP-SMX, Sulfamethoxazole-trimethoprim. Notice that the number of months with zero resistance to Amikacin limits the accuracy of ITS. *Immediate impact of the intervention (P<0.05). ** Impact of intervention on time trends (P<0.05).
In an opposite direction, there was immediate increase in resistance to ceftriaxone (ITS and Poisson models) and ciprofloxacin (Poisson regression), and an upward change in long-term trends of resistance to TMP-SMX (ITS and Poisson regression) and cefazolin (Poisson regression). The impact of seasons in Poisson multivariable models varied, and no resistance rate presented significant seasonal parameters in Box-Jenkins models.
Discussion
COUTI are unique among bacterial infections, because they are by far the ones for which cultures are most frequently collected. Therefore, they are expected to provide a reliable proxy of antimicrobial resistance trends in the community setting [15]. Our study aimed to analyze the impact of restriction of OTC sales of antimicrobials based on trends in resistance in E. coli, the most frequent uropathogen.
Even though the effect of the restriction on sales of antimicrobials in São Paulo state was relevant [11, 12], we found little to no impact on antimicrobial resistance in COUTI E. coli. Trends of resistance to ciprofloxacin were not impacted. Resistance to ceftriaxone accelerated after 2010, while the decreasing trend in TMP-SMX resistance slowed. Immediate or long-term changes after the sales restriction was noticed only for first generation cephalosporins and aminoglycosides, which are not “first line” therapeutic choices for COUTI [16].
Disappointing as our findings may be, they are evidently not an argument against restricting OTC sales of antimicrobials. Instead, they point out that restriction was not sufficient for stopping increasing trends in resistance to widely used therapeutic agents. Presumably, there are other drivers of the emergence and spread of resistance. Regarding E. coli and other Enterobacteriaceae, the fecal carriage of resistant strains in community settings [17], arising from food chain and eventually spread by international travel [18] may be among those drivers. One health approaches have demonstrated the importance of antimicrobial use in agriculture and the dispersion of those drugs in the environment in the causal chain of resistance. Policies that do not address those drivers from a wide ecological perspective are therefore likely to fail [19].
Our study was limited by its small geographic range and relatively short-term analysis. Also, it addressed only one pathogen from a single clinical syndrome, albeit the most frequently identified in the community setting. Finally, it did not include in its analysis the antimicrobial resistance arising and spreading within healthcare-settings. In those settings, other aspects such as hospital connectiveness and patient sharing may play a role in the dissemination of resistance phenotypes [20]. However, there are also strengths in our study. Antimicrobial resistance rates evolve and will obviously not change from one day to another, so our long-term analysis seemed appropriate. We analyzed results from more than 30,000 nonduplicate urine cultures collected over 14 years, using the appropriate ITS approach for natural experiments [13].
Conclusion
In conclusion, we found that legislation of OTC sales of antimicrobials in drugstores did not present a relevant impact on the resistance of E. coli to antimicrobials widely used in clinical practice. Therefore, wider interventions should be consider for prevention and control of antimicrobial resistance in community settings.
References
- 1.
World Health Organization (WHO). Antimicrobial resistance. Draft global action plan on antimicrobial resistance. Geneva: WHO, 2015. Available from: http://apps.who.int/gb/ebwha/pdf_files/WHA68/A68_20-en.pdf
- 2. Munkholm L, Rubin O. The global governance of antimicrobial resistance: a cross-country study of alignment between the global action plan and national action plans. Global Health. 2020;16:109. pmid:33176810
- 3. Klein EY, Van Boeckel TP, Martinez EM, Pant S, Gandra S, Levin SA, et al. Global increase and geographic convergence in antibiotic consumption between 2000 and 2015. Proc Natl Acad Sci U S A. 2018;115:E3463–E3470. pmid:29581252
- 4. Sakeena MHF, Bennett AA, McLachlan AJ. Non-prescription sales of antimicrobial agents at community pharmacies in developing countries: a systematic review. Int J Antimicrob Agents. 2018;52:771–782. pmid:30312654
- 5. Auta A, Hadi MA, Oga E, Adewuyi EO, Abdu-Aguye SN, Adeloye D, et al. Global access to antibiotics without prescription in community pharmacies: A systematic review and meta-analysis. J Infect. 2019;78:8–18. pmid:29981773
- 6. Wirtz VJ, Herrera-Patino JJ, Santa-Ana-Tellez Y, Dreser A, Elseviers M, Vander Stichele RH. Analysing policy interventions to prohibit over-the-counter antibiotic sales in four Latin American countries. Trop Med Int Health 2013;18:665–73. pmid:23551290
- 7. Jacobs TG, Robertson J, van den Ham HA, Iwamoto K, Bak Pedersen H, Mantel-Teeuwisse AK. Assessing the impact of law enforcement to reduce over-the-counter (OTC) sales of antibiotics in low- and middle-income countries; a systematic literature review. BMC Health Serv Res. 2019;19:536. pmid:31366363
- 8. Lipsitch M, Samore MH. Antimicrobial use and antimicrobial resistance: a population perspective. Emerg Infect Dis. 2002;8:347–54. pmid:11971765
- 9. Aminov R. History of antimicrobial drug discovery: Major classes and health impact. Biochem Pharmacol. 2017;133:4–19. pmid:27720719
- 10. Moura ML, Boszczowski I, Mortari N, Barrozo LV, Chiaravalloti Neto F, Lobo RD, et al. The Impact of Restricting Over-the-Counter Sales of Antimicrobial Drugs: Preliminary Analysis of National Data. Medicine (Baltimore). 2015;94:e1605. pmid:26402824
- 11. Kliemann BS, Levin AS, Moura ML, Boszczowski I, Lewis JJ. Socioeconomic Determinants of Antibiotic Consumption in the State of São Paulo, Brazil: The Effect of Restricting Over-The-Counter Sales. PLoS One. 2016;11:e0167885. pmid:27941993
- 12.
CLSI. Performance stantards for antimicrobial susceptibility tests. CLSI supplement M100. Wayne, PA: Clinical and Laboratory Standards Institute, 2020.
- 13. Linden A. Conducting interrupted time-series analysis for single- and multiple-group comparisons. The Stata Journal 2015; 15:408–500.
- 14. Helfenstein U. Box-Jenkins modelling of some viral infectious diseases. Stat Med. 1986;5:37–47. pmid:3961314
- 15. Martínez EP, van Rosmalen J, Bustillos R, Natsch S, Mouton JW, Verbon A. Trends, seasonality and the association between outpatient antibiotic use and antimicrobial resistance among urinary bacteria in the Netherlands. J Antimicrob Chemother. 2020;75:2314–2325. pmid:32417922
- 16. Kang CI, Kim J, Park DW, Kim BN, Ha US, Lee SJ, et al. Clinical Practice Guidelines for the Antibiotic Treatment of Community-Acquired Urinary Tract Infections. Infect Chemother. 2018;50:67–100. pmid:29637759
- 17. Woerther PL, Burdet C, Chachaty E, Andremont A. Trends in human fecal carriage of extended-spectrum β-lactamases in the community: toward the globalization of CTX-M. Clin Microbiol Rev. 2013; 326:744–58. pmid:24092853
- 18. Ruppé E, Andremont A, Armand-Lefèvre L. Digestive tract colonization by multidrug-resistant Enterobacteriaceae in travellers: An update. Travel Med Infect Dis. 2018;21:28–35. pmid:29155322
- 19. Wallinga D, Rayner G, Lang T. Antimicrobial resistance and biological governance: explanations for policy failure. Public Health. 2015;129:1314–25. pmid:26454427
- 20. David S, Reuter S, Harris SR, Glasner C, Feltwell T, Argimon S, et al. Epidemic of carbapenem-resistant Klebsiella pneumoniae in Europe is driven by nosocomial spread. Nat Microbiol. 2019;4:1919–1929. pmid:31358985