You are here

  • Home
  • Archive
  • Volume 93, Issue 8
  • The antibiotic management of gonorrhoea in Ontario, Canada following multiple changes in guidelines: an interrupted time-series analysis

Article Text

Article menu

Download PDFPDF

Health services research
Original article
The antibiotic management of gonorrhoea in Ontario, Canada following multiple changes in guidelines: an interrupted time-series analysis
  1. Catherine Dickson1,
  2. Monica Taljaard2,
  3. Dara Spatz Friedman3,
  4. Gila Metz4,
  5. Tom Wong5,
  6. Jeremy M Grimshaw6
  1. 1 Medical Resident,School of Epidemiology, Public Health and Preventive Medicine, University of Ottawa, Alta Vista Campus Room, Ottawa, Canada
  2. 2 Department of Clinical Epidemiology Program, Ottawa Hospital Research Institute, The Ottawa Hospital, Civic Campus, Ottawa, Canada
  3. 3 Department of Epidemiologist, Ottawa Public Health, Ontario, Canada
  4. 4 Medical Director for Sexual Health, Ottawa Public Health, Ottawa, Canada
  5. 5 Office of Population and Public Health, First Nations and Inuit Health Branch, Health Canada, Ottawa, Canada
  6. 6 Department of Clinical Epidemiology Program, Ottawa Hospital Research Institute, Center for Practice Changing Research Building, The Ottawa Hospital, General Campus, Ottawa, Canada
  1. Correspondence to Dr Catherine Dickson, Medical Resident, School of Epidemiology, Public Health and Preventive Medicine, University of Ottawa, Alta Vista Campus Room, Ottawa, ON K1N 6N5, Canada; catherinedickson01{at}gmail.com

Abstract

Objective This study assessed adherence with first-line gonorrhoea treatment recommendations in Ontario, Canada, following recent guideline changes due to antibiotic resistance.

Methods We used interrupted times-series analyses to analyse treatment data for cases of uncomplicated gonorrhoea reported in Ontario, Canada, between January 2006 and May 2014. We assessed adherence with first-line treatment according to the guidelines in place at the time and the use of specific antibiotics over time. We used the introduction of new recommendations in the Canadian Guidelines for Sexually Transmitted Infections in 2008 and 2011 and the release of the province of Ontario’s Guidelines for the Treatment and Management of Gonococcal Infections in Ontario in 2013 as interruptions in the time-series analysis.

Results Overall, 34 287 gonorrhoea cases were reported between 1 January 2006 and 31 May 2014. Treatment data were available for 32 312 (94.2%). Our analysis included 32 272 (94.1%) cases without either a conjunctival or disseminated infection. Following the release of the 2011 recommendations, adherence with first-line recommendations immediately decreased to below 30%. Adherence slowly increased but did not reach baseline levels before the 2013 guidelines were released. Following release of the 2013 guidelines, adherence again decreased; adherence is slowly recovering but by May 2014, was only approximately 60%.

Conclusions Due to concerns about antibiotic resistance, gonorrhoea treatment guidelines need to be updated regularly and rapidly adopted in practice. Our study showed poor adherence following dissemination of updated guidelines. Over a year after the latest Ontario guidelines were released, 40% of patients did not receive first-line treatment, putting them at risk of treatment failure and potentially promoting further drug resistance. Greater attention should be devoted to dissemination and implementation of new guidelines.

  • Sexually Transmitted Diseases

https://doi.org/10.1136/sextrans-2017-053224

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.

    Introduction

    With current trends in gonorrhoea antimicrobial resistance, we may soon face untreatable multidrug-resistant infections.1 The WHO highlights ‘effective prevention and control of gonococcal infections’ including ‘appropriate treatment regimens’ as key in slowing the progression of antimicrobial resistance.2 Clinical guidelines assist clinicians to select appropriate treatment regimens. Due to rapidly evolving resistance patterns, gonorrhoea treatment recommendations have changed frequently. For example, the Canadian Guidelines on Sexually Transmitted Infections’ first-line gonorrhoea treatment changed twice within 4 years: first with the removal of fluoroquinolones in 2008,3 then in 2011 with the doubling of recommended doses of cefixime and ceftriaxone, the first-line recommendation that ceftriaxone be used to treat men who have sex with men as well as pharyngeal infections, and the addition of combination gonorrhoea therapy with azithromycin for all cases.4 In Ontario, Canada’s most populated province, the 2013 Guidelines for the Treatment and Management of Gonococcal Infections in Ontario introduced a third change, making intramuscular ceftriaxone plus azithromycin the only first-line treatment.5 Table 1 summarises each new guideline’s treatment recommendations.

    View this table:
    Table 1

    A summary of changes in recommendations on the treatment of gonorrhoea in Canada and Ontario from 2004 to the present

    In Ontario, patients with gonorrhoea are seen and managed in a variety of settings including sexual health and STI clinics and community primary care settings that infrequently manage STIs. While antibiotic treatment for gonorrhoea is provided free by public health in Ontario, patients may need to pay for the antibiotic treatment out of pocket if seen in primary care. With frequently changing recommendations, rapid uptake of new guidelines is a concern, especially among clinicians who rarely treat gonorrhoea. Past research has shown that best practice uptake in gonorrhoea management is limited.6–13 Few studies have looked at STI guideline adherence in Canada and none are recent.14 15

    The interrupted time-series (ITS) design is a powerful quasi-experimental approach that can be used to evaluate the impact of an intervention such as a policy change in real-world settings. In this design, outcomes are measured repeatedly over time, both before and after the intervention is introduced. A major strength of the ITS design is that it can account for underlying secular trends in order to measure the incremental effect of an intervention.16 Segmented regression analysis is typically used to analyse ITS data. The regression model is specified to compare change between pre-intervention and postintervention, and can be used to assess both immediate changes (a sudden change in level following the intervention) and gradual changes (a change in slope).16 17

    The purpose of this study was to use interrupted time-series analysis to evaluate changes in adherence to first-line gonorrhoea treatment recommendations in Ontario. Our main hypotheses were that practice would lag behind guideline changes and that changes in adherence would vary with guideline releases, based on the scope of the guideline changes. For example, in 2008, only clinicians prescribing fluoroquinolones would need to change their behaviour; however, in 2011, all clinicians would need to double prescribed doses of cephalosporins and prescribe azithromycin combination therapy, and in 2013, clinicians who have not already done so would need to begin prescribing ceftriaxone for all gonorrhoea cases.

    Methods

    Dataset

    We analysed data from Ontario’s Integrated Public Health Information System (iPHIS). Local Ontario public health units use iPHIS to submit information on reportable diseases to the Ontario Ministry of Health and Long Term Care. In Ontario, all physicians and laboratories are mandated by law to report all diagnosed cases of gonorrhoea to local public health. Our dataset consisted of all reported gonorrhoea cases in Ontario between 1 January 2006 and 31 May 2014 and included sex, infection site(s), treatment, the date the public health was notified, and whether the patient self-identified as having sex with members of the same sex.

    We excluded cases without treatment data and those with either a disseminated or conjunctival infection as treatment recommendations for these infections differ from those for uncomplicated infections in other sites.4

    We classified the prescribed antibiotics into drug families (summarised in online supplementary file S1).

    Supplementary file 1

    For each patient, we defined a dichotomous indicator for whether or not the patient had received first-line treatment according to the clinical guidelines of the time. Following the release of the Ontario guidelines in 2013, we considered the provincial guidelines, as the guidelines of the time. Online supplementary file S2 summarises the criteria used to define adherence to first-line treatment recommendations.

    Supplementary file 2

    Reporting dates were grouped into biweekly time intervals. For each interval, we calculated the aggregate percentage of cases receiving first-line treatment and specific antibiotics. We considered a guideline to be in effect as of the first biweekly interval in the month following the guideline’s release.

    Interrupted time-series analyses

    We conducted an interrupted time-series analysis using a segmented autoregressive linear regression model to analyse changes in the percentage of cases receiving first-line treatment over time.17 We segmented the model following the 2008, 2011 and 2013 guideline updates by including a fixed term representing a change in intercept and a fixed term representing a change in slope to correspond with the timing of each guideline update. This allowed us to evaluate both sudden and gradual changes in level of guideline adherence or rate of uptake of current recommendations, after new recommendations were introduced. We tested each model for stationarity using the Engle-Granger co-integration test. We tested for autocorrelation using the Durbin-Watson test. Where autocorrelation was detected, we included the autoregressive parameters in the model. We conducted the analyses using SAS V. 9.4.

    As secondary analyses, using the same modelling strategy, we conducted separate interrupted times-series analyses for cefixime, ceftriaxone and each drug family that had been a first-line gonorrhoea treatment in the past 10 years.

    For simplicity and interpretability, all analyses were conducted on the percentage scale with changes expressed as absolute differences in percentage. We examined the fit of the models using log-likelihood ratio tests, histograms of residuals, normal probability plots and partial autocorrelation function plots. To account for potential non-linear trends over time and/or deviations from the normal assumption, we repeated the analyses on the log-odds (logit) scale, with changes expressed as relative differences.

    Ethics approval for this study was obtained from the Ottawa Health Science Network Research Ethics Boards (approval number OHSN-REB 20140074–01H).

    Results

    Overall, 34 287 gonorrhoea cases were reported between 1 January 2006 and 31 May 2014. Treatment data were available for 32 312 (94.2%) cases. Our analysis included 32 272 (94.1%) cases without either a conjunctival or disseminated infection. The monthly number of cases increased gradually over time (135 cases in the first biweekly period; 224 in the final biweekly period). Online  Supplementary file 3 shows the monthly distribution of cases.

    Supplementary file 3

    Reported cases of gonorrhoea in Ontario by month from January 2006 to May 2014.

    Adherence to first-line treatment recommendations

    Figure 1 shows the observed and model-based adherence rates. The results of the segmented regression analysis are presented in table 2. Initially, approximately 90% of cases received guideline-recommended first-line treatment; this remained relatively stable until a small reduction in use of first-line treatment, which did not improve with time, prior to the 2011 guideline change. After the 2011 update to the Canadian recommendations, there was a sudden absolute decline in first-line treatment to 21.9% (a 61.4% absolute decrease over that expected, 95% CI: 56.3 to 66.5, p<0.0001) followed by a gradual increase in uptake over time (p<0.0001), reaching 58.3% just prior to the 2013 release of the Ontario guidelines. Following the introduction of the 2013 Ontario guidelines, use of first-line treatment decreased to 42.2% (a 12.5% absolute decrease over that expected, 95% CI: 6.6 to 18.4, p<0.0001) and then increased over time, although at a slower rate than before (p=0.0042). By the end of the study, adherence to first-line recommended treatment had reached 59.1%, substantially lower than at the beginning of the observation period.

    Figure 1
    Figure 1

    Percentage of patients with gonorrhoea in Ontario receiving first-line treatment according to the treatment guideline of the time over time from 1 January 2006 to 31 May 2014.

    View this table:
    Table 2

    Segmented regression analysis of percent adherence to guidelines of the time for gonorrhoea treatment, showing regression coefficient estimates, standard errors and p values

    Specific antibiotic use

    Figure 2 presents the observed and model-based cefixime and ceftriaxone prescription patterns. The results of the segmented regression analyses are presented in online supplementary file 4. Until 2011, the relative frequencies of cefixime and ceftriaxone use remained around 85% and below 5%, respectively. After the 2011 guideline change, cefixime use dropped from 85.3% to 79.0% (a 5.8% absolute decrease over that expected, 95% CI: 1.4 to 10.3, p=0.002) while ceftriaxone use increased from 8.9% to 14.5% (a 4.9% absolute increase over that expected, 95% CI: 1.4 to 8.4, p=0.007). Following these recommendations, the relative frequency of cefixime use decreased (p<0.0001) and the relative frequency of ceftriaxone use increased over time (p<0.0001). With the introduction of the 2013 Ontario recommendations, an immediate drop in cefixime use from 61.1% to 39.1% (a 21.5% absolute decrease over that expected, 95% CI: 17.0 to 26.0, p<0.0001) and a corresponding increase in ceftriaxone use from 36.3% to 58.9% (a 22.1% absolute increase over that expected, 95% CI: 17.8 to 26.4, p<0.0001) were observed. Following the new recommendations, ceftriaxone use continued to increase and cefixime use continued to decrease at levels not different than those following the introduction of the 2011 Canadian recommendations. By May 2014, 71.6% of cases were treated with ceftriaxone and 26.3% of cases were treated with cefixime, suggesting that most patients receiving non-first-line treatment were treated with cefixime.

    Supplementary file 4

    In early 2011, when a decrease in adherence to first-line treatment recommendations not related to a new guideline was noted, a small decline in cefixime use and corresponding rise in ceftriaxone use were observed.

    Figure 2
    Figure 2

    Percentage of patients with gonorrhoea in Ontario receiving cefixime and ceftriaxone over time from 1 January 2006 to 31 May 2014.

    Use of main antibiotic families

    Online supplementary file S5 shows the observed and model-based use of antibiotic families over time, while online supplementary file S6 shows the results of the segmented regression analyses. Initial cephalosporin (including cefixime and ceftriaxone) use was 87.5% and increased gradually with no changes in trend corresponding to any guideline change. The rate of macrolide use (mainly azithromycin) increased with both the introduction of the 2011 Canadian guideline change and the 2013 Ontario guideline. The rate of fluoroquinolone use (not a first-line option since 2008) started at 12.4% and gradually decreased with a small number of cases (2.4%) still receiving fluoroquinolones in 2014.

    Supplementary file 5

    Percentage of gonorrhoea patients in Ontario receiving cephalosporins, macrolides, and fluoroquinolones over time from January 1, 2006 to May 31, 2014.

    Supplementary file 6

    Initial rate of macrolide use (including azithromycin) was 74.6%, likely because treatment for chlamydia was recommended for people with gonorrhoea unless a negative test result is available. Following the release of the 2011 update that recommended combination therapy for all cases with azithromycin, a sudden increase in macrolide use from 69.7% to 76.2% (a 6.2% absolute increase over that expected, 95% CI: 3.4 to 9.0, p<0.0001) was observed followed by increased uptake over time. Following the introduction of the 2013 Ontario guidelines, another sudden increase in uptake from 86.2% to 90.5% (a 4.3% absolute increase over that expected, 95% CI: 0.7 to 8.0, p=0.022) was noted followed by a continued rise.

    Discussion

    This study showed that adherence to first-line treatment recommendations for gonorrhoea in Ontario, Canada, decreased following guideline changes; these effects were greater following major changes in recommendations.

    Our study was unique in its longitudinal analysis of clinician behaviour across multiple changes in gonorrhoea treatment guidelines. Other studies have undertaken cross-sectional analysis of adherence to guideline first-line treatment recommendations,8 9 11 examined changes in guideline adherence over time7 or compared behaviour before and after a single guideline change.6 10 12 Two studies used interrupted time-series analyses to assess fluoroquinolone use before and after the US guidelines stopped recommending its use.6 10 Studies that compared adherence patterns between practitioner groups found that guideline adherence varied by geographical location,6 7 11 type of practice (eg, specialised STI or genitourinary medicine clinics are more likely to adhere to guidelines)7 9 11–13 and number of gonorrhoea cases treated at a location or by a provider.8 12 US studies on ceftriaxone use found higher guideline adherence than our study.7 8 11 12

    Our study has several limitations. The iPHIS database was designed for administrative purposes and only includes information reported to public health. Treatment information may be incomplete if a patient is re-treated without notifying public health. Data are collected and entered locally, and methods may differ between health units. A second limitation is that we did not consider treatment dates when determining whether a patient had received first-line treatment. Some cases received multiple courses of treatment and were considered as having received the first-line treatment if they received the appropriate dose(s) of antibiotics at any point. The proportions of cases receiving first-line treatment would have been lower if we had only considered those who initially received the recommended treatment or who received both combination therapy drugs at the same time as having received first-line treatment. Third, we assumed that cases not treated according to first-line treatment recommendations were treated by clinicians not following guidelines; it is possible that in some cases, the treating clinician was selecting an alternative treatment informed by a culture and sensitivity analysis. Finally, for simplicity of interpretation, we conducted analyses using an additive model and assuming a linear trend over time. Inspection of histograms of residuals and normal probability plots revealed no major departures from the modelling assumptions, and sensitivity analyses conducted on the log-odds scale (not presented) did not change the substantive conclusions of our analyses.

    To examine other factors coinciding with guideline changes that might influence physician behaviour, we searched Pubmed and Canadian Newstand, looking for articles in major medical journals and news stories on gonorrhoea in the 3 months before and after each recommendation was released. Publications found in both medical journals and national or Ontario-based newspapers highlighted antimicrobial resistance trends and likely would have, if anything, promoted guideline use (the publications are summarised in supplementary file S7). We similarly investigated the decline in guideline adherence and increase in ceftriaxone use in early 2011 that was not related to a guideline change. We searched the same databases for publications between 1 November 2010 and 30 June 2011. Case reports of treatment failure with cephalosporins in Europe and Japan18–21 were published during this time. In addition, cases of cefixime treatment failure in Toronto were reported in early 2011.22 Some clinicians may have been aware of these cases and preemptively changed their prescribing behaviour.

    Supplementary file 7

    To date, Canadian and Ontario STI guidelines have largely been disseminated through passive strategies, such as journal publications, webinars, conference presentations, web materials and mobile applications. The relatively slow uptake of new Canadian and Ontario treatment recommendations seen in this study suggests the need for more effective dissemination and implementation strategies. Active dissemination strategies, such as educational outreach, working with opinion leaders, electronic reminders, and audit and feedback, tailored to address the barriers faced by the targeted clinicians have been shown to be effective in promoting changes in practice.23 24 Such interventions should ideally be designed to allow for rigorous evaluation of their impact, for example, by using parallel arm cluster randomised controlled designs, or stepped wedge designs with staggered implementation of the intervention across an entire health system.24 25

    Ineffective gonorrhoea treatment has broad public health implications, through permitting transmission of infection, promoting drug resistance and putting individuals at risk of complications of infection. As gonorrhoea treatment recommendations will continue to change in response to antimicrobial resistance patterns, guideline developers and public health systems should consider ways to enhance uptake. This could also be relevant to developers of guidelines for other infections with similarly rapidly evolving resistance patterns. As STIs are managed in a variety of settings, future research could explore the determinants of recommendation adherence following the introduction of new guidelines in order to identify targets for enhanced dissemination efforts.

    Key messages

    • Appropriate treatment of gonorrhoea infections is important to keep up with changes in antibiotic resistance patterns and to avoid treatment failure.

    • After major guideline changes, the proportion of patients receiving first-line treatment for gonorrhoea infections decreased dramatically followed by a gradual uptake over time.

    • After a guideline change, there is the risk that older potentially less effective regimens will be used.

    • Active guidelines dissemination and implementation strategies may accelerate the uptake of new recommendations.

    References

    1. ↵
      2. Barbee LA
      . Preparing for an era of untreatable gonorrhea. Curr Opin Infect Dis 2014;27:282–7.doi:10.1097/QCO.0000000000000058
      OpenUrlCrossRefPubMed
    2. ↵
      World Health Organization. Global action plan to control the spread and impact of antimicrobial resistance in Neisseria gonorrhoeae. Geneva, Switzerland: World Health Organization, 2012. http://www.who.int/reproductivehealth/publications/rtis/9789241503501/en/ (accessed 30 Jun 2017).
    3. ↵
      Public Health Agency of Canada. Canadian guidelines on sexually transmitted infections. Ottawa, Canada: Her Majesty the Queen in Right of Canada, 2010.
    4. ↵
      Public Health Agency of Canada. Canadian guidelines on sexually transmitted infections. Ottawa, Canada: Her Majesty the Queen in Right of Canada, 2013. http://www.phac-aspc.gc.ca/std-mts/sti-its/cgsti-ldcits/index-eng.php (accessed 30 Jun 2017).
    5. ↵
      Ontario Agency for Health Protection and Promotion (Public Health Ontario). Guidelines for the testing and treatment of gonorrhea in Ontario, 2013. Toronto, Canada: Queen’s Printer for Ontario, 2013. http://www.publichealthontario.ca/en/BrowseByTopic/InfectiousDiseases/Pages/Gonorrhea-Guideline.aspx#.UeA0UuC9yZd (accessed 30June 2017).
    6. ↵
      2. Dowell D ,
      3. Tian LH ,
      4. Stover JA , et al
      . Changes in fluoroquinolone use for gonorrhea following publication of revised treatment guidelines. Am J Public Health 2012;102:148–55.doi:10.2105/AJPH.2011.300283
      OpenUrlCrossRefPubMed
    7. ↵
      2. Kerani RP ,
      3. Stenger MR ,
      4. Weinstock H , et al
      . Gonorrhea treatment practices in the STD Surveillance Network, 2010-2012. Sex Transm Dis 2015;42:6–12.doi:10.1097/OLQ.0000000000000217
      OpenUrl
    8. ↵
      2. Anschuetz G ,
      3. Asbel L ,
      4. Salmon ME , et al
      . Use of first-line treatment for Neisseria gonorrhoeae after treatment guideline changes. Sex Transm Dis 2014;41:64–6.doi:10.1097/OLQ.0000000000000061
      OpenUrl
    9. ↵
      2. Lechtenberg RJ ,
      3. Samuel MC ,
      4. Bernstein KT , et al
      . Variation in adherence to the treatment guidelines for Neisseria gonorrhoeae by clinical practice setting, California, 2009 to 2011. Sex Transm Dis 2014;41:338–44.doi:10.1097/OLQ.0000000000000113
      OpenUrl
    10. ↵
      2. Owusu-Edusei K ,
      3. Carroll DS ,
      4. Gift TL
      . Examining fluoroquinolone claims among gonorrhea-associated prescription drug claims, 2000-2010. Am J Prev Med 2015;49:761–4.doi:10.1016/j.amepre.2015.04.031
      OpenUrl
    11. ↵
      2. Swails J ,
      3. Smock L ,
      4. Hsu K
      . Provider characteristics associated with guideline-nonadherent gonorrhea treatment, Massachusetts, 2010. Sex Transm Dis 2014;41:133–6.doi:10.1097/OLQ.0000000000000073
      OpenUrl
    12. ↵
      2. Tabidze IL ,
      3. Nicholson TF ,
      4. Mikati T , et al
      . Adherence to Centers for Disease Control and Prevention gonococcal treatment guidelines among Chicago health care providers, 2011-2012. Sex Transm Dis 2015;42:422–8.doi:10.1097/OLQ.0000000000000310
      OpenUrlCrossRefPubMed
    13. ↵
      2. Wetten S ,
      3. Mohammed H ,
      4. Yung M , et al
      . Diagnosis and treatment of chlamydia and gonorrhoea in general practice in England 2000-2011: a population-based study using data from the UK clinical practice research datalink. BMJ Open 2015;5:e007776.doi:10.1136/bmjopen-2015-007776
    14. ↵
      2. Moses S ,
      3. Elliott L
      . Sexually transmitted diseases in Manitoba: evaluation of physician treatment practices, STD drug utilization, and compliance with screening and treatment guidelines. Sex Transm Dis 2002;29:840–6.
      OpenUrlPubMed
    15. ↵
      2. Sellors JW ,
      3. Landis SJ ,
      4. Pickard L , et al
      . Factors associated with appropriate physician management of sexually transmitted diseases in an urban Canadian center. Sex Transm Dis 1997;24:393–7.doi:10.1097/00007435-199708000-00002
      OpenUrlPubMed
    16. ↵
      2. Shadish WR ,
      3. Cook TD ,
      4. Campbell DT
      . Experimental and quasi-experimental designs for generalized causal inference. Boston: Wadsworth Cengage Learning, 2002.
    17. ↵
      2. Wagner AK ,
      3. Soumerai SB ,
      4. Zhang F , et al
      . Segmented regression analysis of interrupted time series studies in medication use research. J Clin Pharm Ther 2002;27:299–309.doi:10.1046/j.1365-2710.2002.00430.x
      OpenUrlCrossRefPubMedWeb of Science
    18. ↵
      2. Ison CA ,
      3. Hussey J ,
      4. Sankar KN , et al
      . Gonorrhoea treatment failures to cefixime and azithromycin in England, 2010. Euro Surveill Bull Eur Sur Mal Transm Eur Commun Dis Bull 2011;16:19833.
      OpenUrl
    19. ↵
      2. Ohnishi M ,
      3. Saika T ,
      4. Hoshina S , et al
      . Ceftriaxone-resistant Neisseria gonorrhoeae, Japan. Emerg Infect Dis 2011;17:148–9.doi:10.3201/eid1701.100397
      OpenUrlCrossRefPubMed
    20. ↵
      2. Unemo M ,
      3. Golparian D ,
      4. Syversen G , et al
      . Two cases of verified clinical failures using internationally recommended first-line cefixime for gonorrhoea treatment, Norway, 2010. Euro Surveill 2010;15:19721.
      OpenUrlPubMed
    21. ↵
      2. Golparian D ,
      3. Hellmark B ,
      4. Fredlund H , et al
      . Emergence, spread and characteristics of Neisseria gonorrhoeae isolates with in vitro decreased susceptibility and resistance to extended-spectrum cephalosporins in Sweden. Sex Transm Infect 2010;86:454–60.doi:10.1136/sti.2010.045377
      OpenUrlAbstract/FREE Full Text
    22. ↵
      2. Allen VG ,
      3. Mitterni L ,
      4. Seah C , et al
      . Neisseria gonorrhoeae treatment failure and susceptibility to cefixime in Toronto, Canada. JAMA 2013;309:163–70.doi:10.1001/jama.2012.176575
      OpenUrlCrossRefPubMedWeb of Science
    23. ↵
      2. French SD ,
      3. Green SE ,
      4. O’Connor DA , et al
      . Developing theory-informed behaviour change interventions to implement evidence into practice: a systematic approach using the theoretical domains framework. Implement Sci 2012;7:38.doi:10.1186/1748-5908-7-38
      OpenUrlCrossRefPubMed
    24. ↵
      2. Grimshaw JM ,
      3. Eccles MP ,
      4. Lavis JN , et al
      . Knowledge translation of research findings. Implementation Science 2012;7.doi:10.1186/1748-5908-7-50
    25. ↵
      2. Hemming K ,
      3. Haines TP ,
      4. Chilton PJ , et al
      . The stepped wedge cluster randomised trial: rationale, design, analysis, and reporting. BMJ 2015;350:h391.doi:10.1136/bmj.h391
      OpenUrlFREE Full Text
    26. Public Health Agency of Canada. Canadian guidelines on sexually transmitted infections. Ottawa, Canada: Her Majesty the Queen in Right of Canada, 2006.

    Footnotes

    • Contributors CD developed the research protocol, worked with Public Health Ontario to obtain the data sets, conducted the data analysis and conducted the majority of the writing of the manuscript. MT provided guidance on the statistical analyses and the interpretation of the results, and revised the manuscript critically for important intellectual content. GM provided guidance on the planning of the manuscript, and revised the manuscript critically for important intellectual content. DSF provided guidance on the planning of the manuscript, assisted with data access, and revised the manuscript critically for important intellectual content. TW provided guidance on the planning of the manuscript, and revised the manuscript critically for important intellectual content. JG provided guidance on the planning of the manuscript, and revised the manuscript critically for important intellectual content.

    • Funding No funding was received for this study. JMG holds a Canada Research Chair for Health Knowledge Transfer and Uptake.

    • Competing interests None declared.

    • Ethics approval Ottawa Health Science Network Research Ethics Board (OHSN-REB) – IRB00002616, protocol #: 20140074-01H.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Data sharing statement No additional data is available.