Recommendations for antibiotic prescriptions for upper respiratory symptoms in children in Trinidad and Tobago: GRADE-ADOLOPMENT APPROACH

Rajeev Peeyush Nagassar¹, Elson Robinson², Venkata Ramana Vedula ¹, Soujanya Vastrad¹, Aruna Singh², Angelie Lochan¹, Desiree Baird², Madhura Manjunath², Rampiaree Mangaroo², Stanley Giddings³, Samantha Seecharan², Druv Sookraj², Vishwanath Andy Partapsingh⁴, Edwin Bolastig⁵, Vanessa Elias⁵, Ludovic Reveiz⁵, Nancy Santesso⁶.

Affiliations

1. The Eastern Regional Health Authority, Trinidad and Tobago
2. The North West Regional Health Authority, Trinidad and Tobago
3. The University of the West Indies, St Augustine
4. The Ministry of Health, Trinidad and Tobago
5. The Pan American Health Organization
6. Department of Health Research Methods, Evidence and Impact, McMaster University, Canada

Corresponding Author:
Dr Rajeev P. Nagassar
Sangre Grande Hospital
The Eastern Regional Health Authority, Trinidad and Tobago
Email: [email protected]

DOI: 10.48107/CMJ.2020.05.004

Copyright: This is an open-access article under the terms of the Creative Commons Attribution License which permits use, distribution, and reproduction in any medium, provided the original work is properly cited.

ABSTRACT

Background:  In Trinidad and Tobago, 22,329 and 18,594 cases of acute respiratory infections (ARI) in children less than 5 years were recorded for 2016 and 2017 respectively. Often, antimicrobials are over prescribed without proof of bacterial infection. Recommendations for management of ARIs are needed.

Methods: The GRADE-ADOLOPMENT approach was used to formulate recommendations. We established a guideline panel who met in person or by web conferences. We prioritised recommendation questions and searched for guidelines and systematic reviews from 2010 to July 2018 in several medical databases and guideline producer websites. We also searched for patients’ values and preferences, acceptability, resources, and feasibility studies specific to the Caribbean or Trinidad and Tobago. We summarised the evidence in evidence-to-decision frameworks and formulated recommendations by consensus.

Results: The guideline panel developed recommendations including: 1. For children 5 years old or younger who present with fever and respiratory symptoms not suggestive of serious illness, we recommend to either not prescribe antibiotics or to provide a delayed prescription (48 hours later) of amoxicillin or clarithromycin (when children have a history of allergy to penicillin) rather than an immediate prescription; 2. For children with symptoms suggestive of serious illness, we recommend immediate prescription of amoxicillin or clarithromycin (when children have a history of allergy to penicillin); and, 3. We suggest 7-10 days of treatment, depending on the suspected illness and antibiotic used.

Conclusion: Guidelines aid medical practitioners, patients and supply chain managers. This guideline will form the backbone of the 1st national standardisation of treatment using the GRADE-ADOLOPMENT approach.

BACKGROUND

The Epidemiology Division of the Ministry of Health (MOH), Trinidad and Tobago, recorded 22,329 and 18,594 cases of acute respiratory infections (ARIs) in children less than 5 years for 2016 and 2017 respectively. In 2012, the number of ARI cases were estimated at 23,834. Given that the population under 5 years was approximately 97,300 in 2012, this means, the prevalence of ARI was 24 %.¹ This is quite high and thus a major public health issue. These ARIs were primarily upper respiratory tract (URTIs) and some lower respiratory tract infections (LRTIs) not requiring hospitalisation and as a result be treated in the community. Most URTIs do not last very long and are self-limiting. Typical URTIs and their average duration have been summarised as: 1 ½ weeks for the common cold; 4 days for acute otitis media; 1 week for acute sore throat, acute pharyngitis, and acute tonsillitis; 2 ½ weeks for acute rhinosinusitis; and, 3 weeks for acute cough and acute bronchitis.² Most of these infections were viral and not bacterial. Therefore, antibiotics were not needed in many of these cases, especially in the primary care cases.^3,4

Despite antibiotics being unnecessary, antibiotics are perceived to be necessary and are often taken or are over-prescribed. A survey of caregivers in Trinidad and Tobago found that approximately 40% administered antibiotics to their children for cough or cold episodes without visiting a health care professional.⁴ A survey of physicians in general practice in Trinidad, found that antibiotics, particularly amoxicillin and co-amoxiclav, are over prescribed in paediatric URTI.³ In addition, a study of over 5,000 children admitted to hospital in Trinidad found that antimicrobial agents were prescribed to 36.4%. Over 30% of these children were admitted for respiratory illnesses.⁵ The authors reported that in most cases, antimicrobials were prescribed without proof of a bacterial infection, potentially worsening the global problem of antimicrobial resistance.⁵ In fact, resistance patterns to blood isolates from hospital, community and outpatients in east Trinidad were measured. It was found that for Escherichia coli, Klebsiella pneumonia, and Proteus mirabilis: Imipenem, meropenem, ertapenem showed greater than 80% sensitivity, respectively, while other antibiotics showed greater than 20% resistance. For Pseudomonas aeruginosa: ceftazidime, ciprofloxacin, gentamicin, levofloxacin and tazobactam/piperacillin showed 100%, 80%, 80%, 100% and 100% sensitivity respectively, while other antibiotics including amoxicillin and co-amoxiclav showed greater than 20% resistance.⁶ This highlights the high levels of resistance in the community and hospital.

In Trinidad and Tobago, given the public health impact of acute upper respiratory tract infections in children, the over-prescription, inappropriate use of antibiotics, and the risk of antibiotic resistance, a guideline addressing antibiotic prescribing is warranted. The utility of these guideline recommendations is thought to be at least three-fold. Firstly, to promote appropriate antibiotic prescribing. Secondly to decrease the overall burden of antimicrobial resistance and lastly to increase the supply chain efficiency and equitable access to these antibiotics. These guidelines will form part of the public health efforts to meet the sustainable development goals (SDG), addressing SDG 3 B and 3 C.⁷ It will also contribute to the strengthening of health technology, of which guidelines are a part.

METHODS

We developed these guidelines using the GRADE-ADOLOPMENT approach.⁸ Briefly, the Adolopment process is based on the use of evidence from previously published guidelines and local or regional evidence, to make recommendations that are relevant and applicable to the new guideline setting. The scope of this guideline was proposed by the Ministry of Health (MOH) Trinidad and Tobago and an expert guideline panel was established. The guideline panel included primary care physicians, paediatricians, pharmacists, nurses/district health visitors, a community representative, infectious diseases specialists, medical microbiologists, and supportive members including a methodologist and Pan American Health Organization (PAHO) staff. All members declared their interests using the MOH form to ensure a majority did not have interests that would have an impact on decisions. Meetings of the group took place in person and on secured web-conferencing systems hosted by PAHO. The guideline panel used the PICO (population, intervention, comparator, and outcome) format to develop recommendation questions and selected the questions after discussion (see Table 1). This guideline provides the recommendations for questions 1 and 2; while questions 3-6 will be covered in another guideline.

Table 1: Questions prioritised for making recommendations

*addressed in this guideline

For these questions, we conducted a search for recently published guidelines and systematic reviews from 2010 up to July 2018. We searched in the TRIP Medical database, Epistemonikos, and Google using broad search terms and variations of terms, including child, paediatric, pediatric, infant, and respiratory or fever. We conducted a search for primary studies about patients’ values and preferences, acceptability, resource use, and feasibility specific to the Caribbean islands or Trinidad and Tobago from 2007 to February 2018 in Embase and Medline. The citations were screened for relevance by one methodologist and confirmed by another. Guideline panel members were also asked to identify references.

For each question, we extracted the evidence from the relevant guideline and systematic reviews. When available, we used evidence tables in which the evidence was assessed using the GRADE (Grading of Recommendations, Assessment, Development and Evaluation) approach.⁹ The certainty of the evidence was classified as high, moderate, low, or very low. We prepared a summary of the evidence based on the evidence-to-decision framework.⁸ For both recommendation questions, the evidence for benefits and harms came from the published guidelines and systematic reviews. The evidence for patient values and preferences, acceptability, resources, and feasibility, came from the published literature for the Caribbean islands or from the observations of the guideline panel members. The guideline panel members formulated the recommendations by consensus. The strength of the recommendation was either strong or conditional (see Table 2 for the judgement and interpretation of strong and conditional recommendations). The final wording of the recommendations and remarks were approved by all members of the guideline panel. At least one public consultation was held, comments of which were integrated into the guideline. In addition, feedback was also sought from primary care practitioners, paediatricians and public interest groups.

Table 2: Judgement and interpretation of strong and conditional recommendations

RESULTS

In our search for guidelines 957 documents were screened. We identified 19 applicable guidelines and 103 systematic reviews. One NICE guideline, clinical guideline 69 entitled antibiotic prescribing for respiratory tract infection, was identified.² This clinical guideline was based on systematic reviews and provided adequate documentation in GRADE tables. It was therefore considered appropriate for adolopment.

We screened 423 documents from our search of systematic reviews; evidence from 11 systematic reviews were directly applicable to our PICO questions.^10-20 The search for studies with data specific to Trinidad and Tobago and the Caribbean region identified 203 studies. Of these we included information from 5 studies.^3-6,21,22 We also referred to local, Caribbean, unpublished, resistance data and data from the National Surveillance Unit. We developed three recommendations.

Recommendation 1

For children 6 months to 5 years who present with fever and respiratory symptoms not suggestive of serious illness, we recommend to either not prescribe antibiotics or to provide a delayed prescription (48 hours later) of amoxicillin or clarithromycin (when children have a history of allergy to penicillin) rather than an immediate prescription.  Strong recommendation, low to moderate certainty evidence for effects of interventions

Remarks: This recommendation applies to children 5 years old or younger with symptoms suggestive of non-serious illness including the common cold, acute otitis media, acute sore throat, acute pharyngitis, acute tonsillitis, acute rhinosinusitis, acute cough, or acute bronchitis.

Depending on severity, an immediate prescription may be the preferred option for children with bilateral acute otitis media when younger than 2 years, acute otitis media with otorrhoea, or acute sore throat/acute pharyngitis/acute tonsillitis when three or more Centor criteria are present.

When a delayed prescription or no prescription is provided, clinicians need to advise caregivers of the typical duration of the illness and to fill the prescription or visit a health care provider when symptoms continue or last longer. It will be important to provide a change management approach due to public beliefs in the benefits of immediate prescribing of antibiotics.

Recommendation 2

For children 6 months to 5 years old who present with fever and respiratory symptoms suggestive of serious illness, we recommend immediate prescription of amoxicillin or clarithromycin (when children have a history of allergy to penicillin). Strong recommendation, very low certainty evidence for effects of interventions

Remarks: For some children identified with serious illness, referral to tertiary or secondary centres Accident and Emergency Department (A and E) for immediate assessment and treatment may be necessary.

Children with fever or respiratory symptoms suggestive of serious illness or complications are described by [2] as children who:

• are systemically very unwell
• have pneumonia, mastoiditis, peritonsillar abscess, peritonsillar cellulitis, intraorbital and intracranial complications
• are at high risk of serious complications because of pre-existing comorbidity, including significant heart, lung, renal, liver or neuromuscular disease, immunosuppression, cystic fibrosis, and young children who were born prematurely

Evidence Summary

Immediate prescribing

Undifferentiated acute respiratory infections: Evidence from a Cochrane systematic review found that there are probably trivial benefits of immediate antibiotic treatment compared to no treatment for children with undifferentiated acute respiratory infections; 5 fewer children per 100 would develop acute otitis media, and 0 to 1 more develop pneumonia.¹⁰

Acute otitis media: Evidence reviewed in the NICE Guidance 2008 from a review and a meta-analysis of patient data of immediate antibiotic treatment for children aged between 6 months and 15 years, found that 15 children need to be treated in order to prevent one child from having some pain after 2 days. Children who took antibiotics were more at risk of having adverse events, such as nausea, diarrhoea and rash. In addition, analgesics were sub-optimally used to control pain.^12,15

Acute cough/acute bronchitis: Evidence summarised in the NICE Guidance from a systematic review found that antibiotics may reduce the durations of cough and productive cough, and on feeling ill, by approximately ½ a day.¹³ Evidence for bronchiolitis shows that in children that were admitted to hospital for bronchiolitis who received antibiotics, there may be small benefits of antibiotics (17 per 100 fewer using oxygen; 2 fewer PICU admissions; and 36 fewer readmissions). There will likely be little to no effect on symptom duration. Side effects of antibiotics were not reported but it is probable that side effects would be greater with treatment.²³

Acute sore throat/acute pharyngitis/acute tonsillitis:  A recent review of the literature found that antibiotics shorten the duration of pain symptoms by an average of about one day and can reduce the chance of rheumatic fever by more than two-thirds in communities where this complication is common.¹⁴ However, rheumatic fever is not common in Trinidad and Tobago.

Common cold: Antibiotics had little to no effect on reducing persistence of common cold symptoms and adult patients may experience adverse events from antibiotic use.²

Persistent nasal discharge: The evidence from the NICE guidance found that 8 children need to be treated for 1 cure (in children with nasal discharge greater than 10 days). No long-term benefits were reported.²

Acute rhinosinusitis: A recently published Cochrane review for adults clinically diagnosed with acute rhinosinusitis found that there are small benefits with antibiotic treatment: 5 more adults per 100 had clinical cure and 10 more adults had no purulent discharge after treatment. However, 5 more adults per 100 failed treatment, 7 more had diarrhoea, and 12 more had side effects with antibiotic treatment.¹¹

Delayed prescribing

Acute otitis media: Evidence reported in the NICE guidance found that a delayed prescribing strategy reduces the consumption of antibiotics by 63% compared with an immediate prescribing strategy. There were no differences between immediate or delayed prescribing to reduce the ‘severity’ of earache in children. Adults and children are 12% less likely to develop diarrhea when a delayed prescribing strategy is used, compared with an immediate prescribing strategy. Immediate prescribing reduces night disturbances in children with suspected acute otitis media compared with delayed.²

Acute cough/acute bronchitis: Evidence reported in the NICE guidance found no differences in symptom duration (cough) or the severity of symptoms among no prescription, delayed or immediate prescribing. Delayed and no prescribing strategies reduced the consumption of antibiotics for acute cough in adults and children (by 76% and 80% respectively). There is little to no difference in diarrhoea and fewer re-attendances within 1 month for acute cough in patients offered delayed or immediate prescribing compared to no prescription.²

Acute sore throat/acute pharyngitis/acute tonsillitis: Both no prescription and delayed prescribing reduces the consumption of antibiotics for sore throat in adults and children compared with an immediate prescription (by 13% and 31%, respectively). There is little to no difference in symptom resolution by 3 days and diarrhoea between no prescription, delayed or immediate prescribing in adults and children. There is also little to no difference in re-consultation for sore throat within 1 month between all strategies in adults and children.²

Common cold: Delayed prescribing reduced the consumption of antibiotics by 46% compared with immediate. The evidence suggests that there is little to no difference in temperature or symptoms in adults and children.²

Children at high risk of complications

The NICE guidance found two studies that showed contradictory results for whether fever and chest signs could identify children admitted to hospital before cough. The review also found studies of complications for otitis media: ear discharge (otorrhoea) or eardrum bulging may predict recurrent otitis media; and no evidence was found for mastoiditis. No evidence was found for predicting complications of the common cold or acute rhinosinusitis, but intraorbital and intracranial complications are serious complications of acute rhinosinusitis.² Although, published evidence is not available for Trinidad and Tobago specifically, it is likely that this research would be similar to the context in Trinidad and Tobago.

Resistance

The study by Nagassar et al of antibiotic resistance patterns in east Trinidad showed that resistance for antibiotics including amoxicillin and co-amoxiclav is greater than 20%.⁶ The guideline panel agreed that resistance to antibiotic treatments commonly used to treat respiratory symptoms and fever is occurring in Trinidad Tobago and the world. Global initiatives are in place to reduce the inappropriate use of antibiotics for viral infections.

Costs

A model conducted by NICE suggests that the least costly option is to adopt a delayed antibiotic strategy.² In the UK, the expected cost of a delayed prescription was £14 per patient compared with £16 and £45.50 for the no antibiotic and immediate antibiotic prescribing strategies respectively. One study conducted in Haiti reported that the availability of essential medicines is low and prices varied widely, with most medication regimens largely unaffordable to the public.²⁰ The authors indicated that using and including essential medications on the national formulary and working with organizations responsible for importing medications into Haiti could increase access. The guideline panel estimated that the cost of changing awareness, beliefs and behaviour to delay or not prescribe antibiotics will be greater initially. However, there are communication channels in place to provide this information to patients and clinicians. In the long term, the benefits, and reduction in prescriptions and risk of resistance would probably result in large cost savings.

Acceptability, Equity and Feasibility

In Trinidad and Tobago, approximately one third of the population is in rural communities. However, even in remote communities, people have access to hospitals, public and private pharmacies, and dispensers for a few hours a day. They could obtain medications if a delayed prescription or no prescription was provided. The guideline panel agreed that there may be some resistance in caregivers provided with no prescription or delayed prescription due to beliefs. Education and increased public awareness about the small to no benefits of antibiotics for non-serious illness related to fever and respiratory symptoms and about the harms (e.g., resistance) will likely be needed. Five studies in the review by NICE in the UK and USA found that caregiver/patient satisfaction with no prescription, delayed prescription or immediate prescriptions were generally above 70%.² The guideline panel agreed that some clinicians may fear the consequences of not providing a prescription and the illness worsens. However, the guideline panel also agreed that if children continue to feel unwell or worsen, caregivers will return for a visit or fill a delayed prescription and the children will not be missed. In addition, the panel agreed that clinicians would be able to refer children to tertiary or secondary centres if the illness was serious, for further assessment and treatment.

Overall justification

The guideline panel agreed that the benefits of immediate prescribing in non-serious illness are small, and there are also small numbers of harms to children. However, resistance is a critical harm which at a public health level results in a large harm to society. Not providing an immediate prescription, and instead providing a delayed prescription or no prescription is feasible to provide as clinics and dispensaries are accessible for most to all caregivers. Delayed or no prescription will probably be acceptable to caregivers and clinicians. However, initially, management of changing public knowledge and beliefs is necessary. Although, the costs of managing this change may be initially large, there will probably be large cost savings when delayed or no prescription strategies are implemented in Trinidad and Tobago. There is low to very low quality evidence for how to identify children at high risk of complications. However, the guideline panel agreed that children with serious illness or suspected serious complications should receive antibiotics immediately. They also agreed that clinical expertise would be useful to identify children at high risk.

Recommendation 3

Delayed or immediate antibiotic prescription of amoxicillin or clarithromycin (when allergy to amoxicillin) should be provided rather than other antibiotics. When prescribing amoxicillin or clarithromycin, we suggest a minimum of 7-10 days of treatment depending on the suspected illness and antibiotic used.  Strong recommendation, low to moderate certainty evidence of effects

Remarks: The clinical decision for 7 to 10 day treatment will be based on the suspected course of illness, or based on the clinical response after a short duration of treatment.

Evidence summary

Acute otitis media: A systematic review by Shekelle et al found no major differences in the effects of antibiotics and durations for otitis media.¹⁹

Acute sore throat/acute pharyngitis/acute tonsillitis: Evidence from two systematic reviews found that there is likely little to no difference in the clinical effects of different classes of antibiotics.^16,20 There is also little to no difference in adverse events for cephalosporins, macrolides or sulfonamides compared with phenoxymethylpenicillin.²⁰ The other systematic review found that a shorter course of late-generation (broader spectrum) antibiotics would likely result in more adverse events compared with a 10-day course of phenoxymethylpenicillin.¹⁶

Acute rhinosinusitis: There are inconsistent effects of antibiotics: one review found that more children may improve symptoms at 10 to 14 days follow-up with antibiotics compared with placebo; the other review found that different antibiotics likely do not increase cures or improve symptoms compared with placebo.^17,18 Organisms causing acute sinusitis that are resistant to phenoxymethylpenicillin are also likely to be resistant to amoxicillin.

 Acute cough/acute bronchitis: A systematic review with studies in adults, adolescents and children receiving a variety of antibiotics for different durations (5 to 10 days) found that symptoms may be reduced by half day with antibiotics. Fewer people may have a cough after treatment, but there is likely little difference in the number of people who are clinically improved with or without antibiotics. Adverse events were likely increased compared to no antibiotics. Few data were available to compare effects by duration of treatment, but there did not appear to be an association.¹³

Acceptability, feasibility, equity, costs

Both amoxicillin and clarithromycin are well-established and commonly used in Trinidad and Tobago for respiratory tract infections, which means that there is clinical experience with these drugs. Both drugs are available through the National Formulary and in the private sector. Amoxicillin, in particular, is more widely available across Trinidad and Tobago than clarithromycin and is the narrowest spectrum drug available. A study by Nagassar et al published in 2017, involving retrospective antibiotic prescribing practices in all age groups, indicated that co-amoxiclav was the second most commonly used antibiotic in the hospital setting, indicating a high level of acceptability to physicians, at least in east Trinidad.²² The oral formulation on Phenoxymethylpenicillin is Penicillin V and is available in tablets and suspension. Penicillin G is however, provided by injection and is less acceptable to provide and take. Penicillin V is typically used for sickle cell disease and its use should be limited to reduce the risk of increasing resistance in Trinidad and Tobago. Amoxicillin is the current practice for antibiotic treatment in children with acute otitis media in paediatric practice. However, some doctors may need to be encouraged to prescribe the narrower spectrum amoxicillin rather than the broader spectrum Co-amoxiclav. The cost of amoxicillin suspension is cheaper than clarithromycin suspension. It should be noted that co-amoxiclav suspension is similar in price to amoxicillin suspension. In a cost-effectiveness model produced by NICE, amoxicillin for 5 days was more costly and less effective.² Typically, additional costs are related to return visits.

Overall justification

The guideline panel agreed that there may be trivial differences in benefits and harms of different antibiotics when treatment is provided. However, resistance is a critical harm which at a public health level results in a large harm to society. Although, there may be less risk of increasing resistance to phenoxypenicillin, it is currently not available in oral form which is a barrier to its use. Amoxicillin and clarithromycin oral suspensions are available. The guideline panel agreed that amoxicillin would be preferred since it is cheaper and is the narrowest spectrum antibiotic available on the National Formulary. Co-amoxiclav is also commonly used, but it is a broader spectrum antibiotic that has a higher risk of creating resistance. There is low to very low certainty evidence in the differences in outcomes with treatment shorter or longer than 5 days. In most studies, antibiotics were provided for approximately 7-10 days, with costs incurred due to return visits when treatment was shorter.

DISCUSSION

This guideline presents recommendations made by a guideline panel with expertise in microbiology, infectious diseases, family medicine and public health, nursing, pharmaceuticals and paediatrics practicing in Trinidad and Tobago using the GRADE-ADOLOPMENT approach.⁸ The guideline panel used evidence from other published guidelines and systematic reviews to inform the recommendations. In particular, evidence synthesised in the NICE clinical guideline 69: Respiratory tract infections (self-limiting): prescribing antibiotics was used by the guideline panel.² Most importantly, however, the panel considered evidence specifically from Trinidad and Tobago and the Caribbean region. Their expert observations, and wider consultation from stakeholders contextualise the recommendations to ensure that the recommendations are applicable to Trinidad and Tobago. The current use of antibiotics, the availability of antibiotics, through the National Formulary, and the present prevalence of acute respiratory conditions in children less than or equal to five (5) years were considered.

These recommendations will have broad public health implications. This guideline will provide the first country specific guideline, using the GRADE-ADOLOPMENT approach, for the use of the first line antibiotics, amoxicillin and clarithromycin. There have been at least two journal articles recommending the development of guidelines for URTI from Trinidad and Tobago, including Mohan et al and Parimi et al.^3,4 This guideline can lead to more appropriate antibiotic use, less toxicity, minimization of costs for these pharmaceuticals, and a reduction in the development of resistance in bacteria. These benefits will be of national significance and will help address the national, regional and global public health issue of antibiotic resistance due to inappropriate antibiotic use.

With this guideline, we recognise that it will be necessary to increase the awareness in caregivers to reduce the demand for antibiotics for upper respiratory infections. Providing alternatives to caregivers to reduce the symptoms of upper respiratory infections would likely be helpful and more acceptable in Trinidad and Tobago. Future recommendations will address the effects and thus benefits or harms of alternatives to prescribing antibiotics, such as using honey, antihistamines, antipyretics or normal saline nebulisation. It will also look at the acceptability, feasibility, cost and equity in the use of these therapies. These future recommendations may rationalise the use of these alternatives and thus address the effect of overprescribing antibiotics on antibiotic resistance.

While these recommendations have been made in the context of the Trinidad and Tobago setting, there are still country-wide issues that will need to be addressed in order to ensure these guidelines can be implemented. It will be important that more research is done on the public health impact of inappropriate antibiotic use in respiratory tract and other types of infections. Public awareness and health promotion activities on a national scale will assist with making these guidelines more acceptable. In addition, there needs to be change management in addressing the need for the public, in Trinidad and Tobago, to understand that antibiotics are not the only answer to curing an infection. Change management should also target the medical profession, to promote the use of locally developed guidelines. Medical colleagues must understand the difference between a protocol and a properly developed guideline and the impact on policy. This change management could occur during Global Antibiotic Awareness week and in other public health campaigns to improve the uptake of this guideline.

Although attempts were made to use local existing literature to inform these recommendations, there are still gaps in this literature. In Trinidad and Tobago, additional research could include updated research on the epidemiology of respiratory tract infections, including upper respiratory and lower respiratory tract infections. Observational studies to delineate factors contributing to inappropriate antibiotic use, publication of local and regional resistance patterns, and local systematic reviews will also be useful. A culture of research has to be slowly developed in the medical community in Trinidad and Tobago.

In addition, the benefits of standard treatment guidelines to health care practitioners, supply managers and patients should be communicated. For medical practitioners, guidelines provide evidence-based guidance which can lead to improvements in diagnostic accuracy, effective and safe therapy. Guidelines can provide standardised information to give to patients, provide a basis for measuring, monitoring, evaluating, and improving performance and quality of care and support evidence, protection, or defence against malpractice. For patients the benefits are at least five-fold: improve the availability of medicines due to consistent usage patterns; enable consistent and predictable treatment from various levels of providers and locations; enhance treatment outcomes (cure/alleviation/reduced development of resistance/reduced toxicity); increase satisfaction with care; and, reduce health care costs. Ultimately, these benefits could have an impact on the national tendering process in Trinidad and Tobago. Subsequently, on the supply chain management of these antibiotics by improving availability of medicine, providing a standardised basis for quantifying, ordering, and pre-packaging medicines where possible, appropriate and applicable.²⁴ While guidelines can have an important impact, the process of developing guidelines can be seen as tedious. During public consultation of this guideline, issues with the GRADE-ADOLOPMENT approach were noted. This included the length of time needed to synthesise the evidence and formulate recommendations. In future, a culture of rigorous and practical guideline adoption/adaptation should be interwoven into the change management part of health system. Also, could be part of the undergraduate curriculum in medical and allied health sciences and incorporated in continuing medical education to assist with this change management. Lastly, this guideline will be reviewed and updated every two (2) years.

REFERENCES

1. UNICEF. Statistics. At a glance: Trinidad and Tobago. 2018 [Available from: https://www.unicef.org/infobycountry/trinidad_tobago_statistics.html].
2. NICE. NICE clinical guideline 69: Respiratory tract infections (self-limiting): prescribing antibiotics. Updated February 2014 [Available from: https://www.nice.org.uk/guidance/cg69].
3. Mohan S, Dharamraj K, Dindial R et al. Physician behaviour for antimicrobial prescribing for paediatric upper respiratory tract infections: a survey in general practice in Trinidad, West Indies. Annals of clinical microbiology and antimicrobials. 2004;3:11.
4. Parimi N, Pinto Pereira LM, Prabhakar P. Caregivers’ practices, knowledge and beliefs of antibiotics in paediatric upper respiratory tract infections in Trinidad and Tobago: a cross-sectional study. BMC family practice. 2004;5:28.
5. Orrett FA. Antimicrobial prescribing patterns at a rural hospital in Trinidad: evidence for intervention measures. African journal of medicine and medical sciences. 2001;30(3):161-4.
6. Nagassar RP B-NR. What are the Roles of Carbapenems, in an Institution Specific Epidemiological Antibiogram, in East Trinidad? West Indian Medical Journal.
7. United Nations. Sustainable Development Goal 3: Ensure healthy lives and promote well-being for all at all ages. [Available from: https://sustainabledevelopment.un.org/sdg3#targets]
8. Schunemann HJ, Wiercioch W, Brozek J et al. GRADE Evidence to Decision (EtD) frameworks for adoption, adaptation, and de novo development of trustworthy recommendations: GRADE-ADOLOPMENT. Journal of clinical epidemiology. 2017;81:101-10.
9. Guyatt G, Oxman AD, Akl EA, Kunz R et al. GRADE guidelines: 1. Introduction-GRADE evidence profiles and summary of findings tables. Journal of clinical epidemiology. 2011;64(4):383-94.
10. Alves Galvao MG, Rocha Crispino Santos MA, Alves da Cunha AJ. Antibiotics for preventing suppurative complications from undifferentiated acute respiratory infections in children under five years of age. The Cochrane database of systematic reviews. 2016;2:Cd007880.
11. Lemiengre MB, van Driel ML, Merenstein D et al. Antibiotics for acute rhinosinusitis in adults. The Cochrane database of systematic reviews. 2018;9:Cd006089.
12. Rovers MM, Glasziou P, Appelman CL et al. Antibiotics for acute otitis media: a meta-analysis with individual patient data. Lancet (London, England). 2006;368(9545):1429-35.
13. Smith SM, Fahey T, Smucny J et al. Antibiotics for acute bronchitis. The Cochrane database of systematic reviews. 2017;6:Cd000245.
14. Spinks A, Glasziou PP, Del Mar CB. Antibiotics for sore throat. The Cochrane database of systematic reviews. 2013(11):Cd000023.
15. Venekamp RP, Sanders SL, Glasziou PP, Del Mar CB et al. Antibiotics for acute otitis media in children. The Cochrane database of systematic reviews. 2015(6):Cd000219.
16. Altamimi S, Khalil A, Khalaiwi KA et al. Short-term late-generation antibiotics versus longer term penicillin for acute streptococcal pharyngitis in children. The Cochrane database of systematic reviews. 2012(8):Cd004872.
17. Cronin MJ, Khan S, Saeed S. The role of antibiotics in the treatment of acute rhinosinusitis in children: a systematic review. Archives of disease in childhood. 2013;98(4):299-303.
18. Falagas ME, Giannopoulou KP, Vardakas KZ et al. Comparison of antibiotics with placebo for treatment of acute sinusitis: a meta-analysis of randomised controlled trials. The Lancet Infectious diseases. 2008;8(9):543-52.
19. Shekelle PG, Takata G, Newberry SJ et al. Management of Acute Otitis Media: update. Evidence report/technology assessment. 2010(198):1-426.
20. van Driel ML, De Sutter AI, Habraken H et al. Different antibiotic treatments for group A streptococcal pharyngitis. The Cochrane database of systematic reviews. 2016;9:Cd004406.
21. Chahal HS, St Fort N, Bero L. Availability, prices and affordability of essential medicines in Haiti. Journal of global health. 2013;3(2):020405.
22. Nagassar RP HK, Daniel K, Bridgelal-Nagassar R. A Retrospective Study of Antibiotic Prescribing at the Sangre Grande Hospital, Trinidad, West Indies. West Indian Medical Journal.
23. NICE. NICE guideline 9: Bronchiolitis in children: diagnosis and management. 2015.
24. SIAPS. Developing, Implementing, and Monitoring the Use of StandardTreatment Guidelines: A SIAPS How-to Manual. Submitted to the US Agency for International Development by the Systems for Improved Access to Pharmaceuticals and Services (SIAPS) Program. . Arlington, VA: Management Sciences for Health.; 2015.