BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from
Brazilian medicinal plants to treat respiratory disease: systematic review and meta-analyses - Study Protocol
ForJournal: peerBMJ Open review only Manuscript ID: bmjopen-2014-005267
Article Type: Protocol
Date Submitted by the Author: 17-Mar-2014
Complete List of Authors: Lopes, Luciane; UNISO, Pharmacie Science; University of Sorocaba, Pharmaceutical Science Silva, Maria Carolina; University of Sorocaba, Pharmaceutical Science Motta, Cristiane; University of Sorocaba, Pharmaceutical Science Quirós, Antonio; Universidad de Córdoba, Biología Celular, Fisiología e Inmunología. Facultad de Medicina Biavatti, Maique; Federal University of Florianopolis, Pharmaceutical Department de Oliveira, Jardel; Secretaria Municipal de Saúde, Médico de Família e Comunidade, Especialista em Saúde da Família, Geriatria e Gerontologia Guyatt, Gordon; McMaster Unviersity, Department of Clinical Epidemiology and Biostatistics
Primary Subject http://bmjopen.bmj.com/ Complementary medicine Heading:
Pharmacology and therapeutics, Respiratory medicine, Complementary Secondary Subject Heading: medicine
COMPLEMENTARY MEDICINE, RESPIRATORY MEDICINE (see Thoracic Keywords: Medicine), THERAPEUTICS, Herbal medicine < THERAPEUTICS
on September 27, 2021 by guest. Protected copyright.
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 22 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 Brazilian medicinal plants to treat respiratory disease: 4 5 systematic review and meta-analyses- Study Protocol 6 7 Luciane C Lopes 8 Pharmaceutical Sciences Post graduate Course, University of Sorocaba, UNISO , SP, Brazil. 9 [email protected] 10 11 Maria Carolina O. Silva 12 Pharmaceutical Sciences Post graduate Course, University of Sorocaba, UNISO, SP, Brazil 13 [email protected] 14 15 Cristiane BergamashiFor Mottapeer review only 16 Pharmaceutical Sciences Post graduate Course, University of Sorocaba, UNISO, SP, Brazil 17 [email protected] 18 19 Antonio Macho Quirós 20 21 Biología Celular, Fisiología e Inmunología. Facultad de Medicina, Universidad de Córdoba, Spain 22 [email protected] 23 24 Maique Weber Biavatti 25 Pharmaceutical Department, Federal University of Florianopolis, SC, Brazil 26 [email protected] 27 28 Jardel Corrêa de Oliveira 29 Médico de Família e Comunidade, Especialista em Saúde da Família, Geriatria e Gerontologia, 30 Secretaria Municipal de Saúde, Florianópolis, SC, Brasil 31 32 [email protected]
33 http://bmjopen.bmj.com/ 34 Gordon Guyatt 35 Department of Clinical Epidemiology and Biostatistics, McMaster University, Canada. 36 [email protected] 37 38 Contact information for the corresponding author: Luciane Cruz Lopes, [email protected] 39 40 Rua Gomes Carneiro, 570 apto 141, Postal code: 13400-530 Piracicaba, SP, Brazil.
41 on September 27, 2021 by guest. Protected copyright. 42 Telephone: 55 19 978-18441 43 Fax: 55 15 21017074 44 45 46 Financial support: This project is funded from a grant by CAPES Research Programme 47 (grant) and is Support Program Graduate Education Institutions Private - PROSUP - 48 CAPES / UNISO 49 No conflict of interest 50 Word count: 3381 Figures: 0; tables: 0; references: 52 51 Additional file: 1 52 Keywords: Common cold, upper respiratory disease, Cough, Systematic review, 53 54 Meta-analysis, Brazilian medicinal plants 55 56 57 58 59 60 1 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 22 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 ABSTRACT 4 5 Introduction: Respiratory illness, often associated with cough, is frequent. In Brazil, 6 7 herbal medicines are often recommended as a first-line treatment for respiratory illness. 8 9 There exists uncertainty regarding the effectiveness of these treatments. No systematic 10 review has evaluated Brazilian medicinal plants to treat the common cold, lower and 11 12 upper respiratory tract infection and associated cough. 13 14 15 MethodsFor and analysis: peer We will conductreview a systematic reviewonly and if appropriate a 16 series of meta-analyses evaluating the safety and effectiveness of Brazilian medicinal 17 18 plants for the common cold and upper respiratory tract infection and associated cough. 19 20 We will acquire eligible randomized controlled trials and observational studies in adult 21 or pediatric patients with such illness treated by any form of Brazilian herbal compared 22 23 with placebo, no treatment, or an alternative therapy through a systematic search of 24 25 CINAHL, EMBASE, MEDLINE, AMED, Web of Science, Ovid Health star, Cochrane 26 Library, Pubmed and Scielo and the Cochrane Central Registry of Controlled Trials. 27 28 Teams of reviewers will, independently and in duplicate, screen titles and abstracts and 29 30 complete full text reviews to determine eligibility, and subsequently perform data 31 abstraction and assess risk of bias of eligible trials. When appropriate, we will conduct 32
33 meta-analyses to establish the effect of all reported therapies on patient-important http://bmjopen.bmj.com/ 34 35 outcomes. 36 37 38 Discussion: Our review will be the first to evaluate all Brazilian medicinal plants to 39 treat the common cold, upper respiratory tract infection and associated cough and 40
41 establish best estimates of the safety and effectiveness of treatments. Our review will on September 27, 2021 by guest. Protected copyright. 42 43 facilitate evidence-based management of patients in primary care and identify key areas 44 for future research. 45 46 47 Ethics and Dissemination: The systematic review will be published in a peer- 48 49 reviewed journal. Brief reports of review findings will be disseminated directly to 50 appropriate audiences via email and other modes of communication. The review will 51 52 guide healthcare practice and policy in Brazil. 53 54 Register Protocol: Prospero CRD42014007057 55 56 57 58 59 60 2 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 22 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 ARTICLE FOCUS: 5 Will a systematic review of Brazilian medicinal plants to treat the common cold, lower 6 and upper respiratory tract infection and associated cough reveal effectiveness of these 7 8 agents? 9 10 Do Brazilian medicinal plants lead to symptom improvement and cough control in 11 common cold, lower and upper respiratory tract infection in patients? 12 13 14 15 For peer review only 16 KEY MESSAGE: 17 This study will investigate the effects of all marketed Brazilian medicinal plants 18 indicated to use in respiratory disease: Ananas comosus, Echinacea purpurea Moench, 19 20 Eucalyptus globules, Glycyrrhiza glabra L., Hedera helix L., Malva sylvestris L., 21 22 Mentha spp* (M. piperita or M. villosa), Mikania spp* (M. glomerata or M. laevigata), 23 Pelargonium sidoides D.C., Petasites hybridus L., Pimpinella anisum L., Polygala 24 25 senega L., Psychotria ipecacuanha (Brot.) Stokes, Sambucus nigra L. 26 27 28 Outcomes of interest will include time to resolution of clinical symptoms and/or signs 29 (where clinical symptoms and signs include cough, sputum production or activity 30 31 limitations). Also of interest will be severity of symptoms prior to resolution. Other 32 important outcomes will include hospitalization rates and duration of hospital stay, and 33 http://bmjopen.bmj.com/ 34 days receiving antibiotics 35 36 37 38 STRENGTHS AND LIMITATIONS OF THIS STUDY 39 This will be the first systematic review to assess specifically the Brazilian medicinal 40 plants to treat the common cold, lower and upper respiratory tract infection and
41 on September 27, 2021 by guest. Protected copyright. 42 associated cough. The methods of the review are state-of-art, including explicit 43 eligibility criteria, a comprehensive search, independent duplicate assessment of 44 eligibility, and use of the GRADE approach to assessing confidence in estimates of 45 effect including independent duplicate assessment of risk of bias, precision, consistency, 46 47 directness and publication bias. We will make separate ratings for bodies of evidence 48 from randomized trials and bodies of evidence from observational studies. 49 50 Our results are likely to be limited by limitations in the primary studies include non- 51 randomized studies and randomized trials with a high risk of bias. Eligible studies will 52 53 differ substantially in study design and outcome measures. 54 55 56 57 INTRODUCTION 58 59 60 3 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 22 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 Respiratory illness is common and important 4 5 Respiratory illness, including the common cold and upper respiratory tract infection 6 7 (URTI), and associated cough, are frequent[1] and a major cause of morbidity, 8 9 especially in children and the elderly [2]. Although in most cases benign, respiratory 10 illness results in many consultations to primary care [3 4]. 11 12 13 Acute URTIs are the most common reason for people to seek medical care in the United 14 States [5] and at least one billion colds occur there per year, with a frequency of two to 15 For peer review only 16 six colds per person [6]. Symptoms of the common cold typically include a runny nose, 17 18 congestion, sneezing, weakened sense of taste and smell, scratchy throat and cough. 19 These start developing in the first three days following infection. Infants and young 20 21 children are more likely than adults and teens to also develop fever. Symptoms usually 22 23 abate within seven to 10 days but some colds last longer, especially in children, the 24 elderly and those with generally poor health [7]. The most common symptom of 25 26 respiratory tract infection, cough, is also the most common symptom presenting to 27 28 general practitioners [5 8]. 29 Cough may arise from at least three mechanisms. One is virus-induced postnasal drip. 30 31 Alternatively, it has been proposed that a viral upper respiratory tract infection produces 32
33 inflammatory mediators that result in an increase in the sensitivity of the afferent http://bmjopen.bmj.com/ 34 sensory nerves in the upper airway[9]. Third, the infection may spread from the upper 35 36 to the lower respiratory tract with a resulting bronchitis. This third mechanism is often 37 38 associated with sputum production. 39 40 Cough can be characterized based on time frame (ie, duration of cough), quality (eg, dry
41 or wet, brassy, or staccato), or suggested etiology (ie, specific and nonspecific)[10]. on September 27, 2021 by guest. Protected copyright. 42 43 Cough can be designated as acute (<3 weeks in duration), prolonged acute cough (3 to 8 44 45 weeks in duration) or chronic (> 8 weeks in duration) [11 12]. 46 In adults, although there is no prospective study of the causes of acute cough, it has long 47 48 been considered that the common cold is the single most common cause of acute cough 49 50 (ie, cough < 3 weeks in duration)[9]. 51 52 In most children acute coughing is usually due to a viral upper respiratory tract infection 53 54 such as a simple head cold with bronchitis or croup. Less often, but still common, 55 pathogens can involve the lower respiratory tract system causing bronchiolitis, 56 57 58 59 60 4 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 22 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 whooping cough, or pneumonia. Symptomatic URTI with cough in school children 4 5 typically occurs around 7–10 times per year[13]. 6 7 Non-herbal medication use to treat respiratory infection is common 8 9 Worldwide, the desire to reduce the symptom of cough is reflected in the billions of 10 11 dollars spent on over-the-counter (OTC) products, mostly cough and cold medications 12 13 (CCMs) [1]. The preparations are usually a combination of several medications 14 including antitussives, expectorants, antibiotic, antihistamines, decongestants and 15 For peer review only 16 antipyretics [14]. 17 18 19 Current systematic reviews addressing the use of CCMs show insufficient evidence to 20 decide whether OTC medications for cough beneficial[14]. It has been suggested that 21 22 zinc can inhibit viral growth[15]. As such, the treatment of cough and cold with zinc 23 24 was tested in several studies[16]. While some of them showed benefits, especially if 25 used within 24 h of the onset of common cold symptoms [17][18], others failed to show 26 27 the same effect[18]. At the present time, the use of zinc in children with cough and cold 28 29 is not recommended[1]. The available evidence does not support the use of high doses 30 of vitamin C for treating the common cold [19]. Likewise, there is insufficient evidence 31 32 to support the treatment of upper respiratory tract infections with antibiotics, and an
33 http://bmjopen.bmj.com/ 34 increased adverse effects associated with antibiotic use in adult patients [20]. 35 36 Antihistamines in monotherapy - in children as well as in adults - do not alleviate nasal 37 38 congestion, rhinorrhea and sneezing, or produce subjective improvement of the common 39 cold[21]. First generation antihistamines also cause more side-effects than placebo, in 40
41 particular they increase sedation in cold sufferers. Combinations of antihistamines with on September 27, 2021 by guest. Protected copyright. 42 43 decongestives are not effective in small children. In older children and adults most trials 44 show a beneficial effect on general recovery as well as on nasal symptoms. However, 45 46 the extent to which improvement is important to patients remains unclear [22]. 47 48 49 There are no effective licensed antiviral drugs for the common cold. Of the mucolytic 50 drugs available to treat acute upper and lower URTI, the cysteine derivatives (that is, 51 52 acetylcysteine and carbocysteine) are the most commonly prescribed in many European 53 54 [23 24] and African countries[25] and in Brazil [8 26] and seem to have a limited 55 efficacy and also appear to be safe in children older than two years. 56 57 58 59 60 5 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 22 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 Herbal medicines represent an alternative for treatment of respiratory illness 4 5 The synonyms of herbal medicines include herbal remedies, herbal medications, herbal 6 7 products, herbal preparations, medicinal herbs and phytopharmaceuticals. Patients and 8 9 physicians may be unaware that products with the same label differ appreciably in their 10 composition, mainly due to the use of variable plant material, extraction methods and 11 12 the addition of other components. 13 14 In high-income countries there is increasing public interest in, and use of, a wide range 15 For peer review only 16 of therapies that lie outside the main stream of traditional Western medical practice[27 17 18 28]. Complementary and alternative medicine (CAM) has grown rapidly over the last 19 two decades[27]. The Cochrane Collaboration defines complementary and alternative 20 21 medicine as a broad domain of healing resources that encompasses all health systems, 22 23 modalities, and practices and their accompanying theories and beliefs, other than those 24 intrinsic to the politically dominant health systems of a particular society or culture in a 25 26 given historical period. In the United States, approximately 38 percent of adults (about 4 27 28 in 10) and approximately 12 percent of children (about 1 in 9) are using some form of 29 CAM. 30 31 32 In Brazil, up to 25% of the $ 8 billion of revenues of the pharmaceutical industry, in
33 http://bmjopen.bmj.com/ 34 1996, may come from sales of drugs derived from plants [29]. United States and 35 Germany are among the largest consumers of Brazilian natural products. Between 1994 36 37 and 1998, imported, respectively, 1,521 and 1,466 tons plants that follow for these 38 39 countries under the generic label "Plant material of Brazil," according to Brazilian 40 Institute of Environment and Renewable Natural Resources[30].
41 on September 27, 2021 by guest. Protected copyright. 42 43 Herbal medicines have been widely used in cough[31]. Antitussives act either centrally 44 on the cough center of the brain or peripherally on the cough receptors in the respiratory 45 46 passages. The putative antitussive effect of many herbs may result from the content of 47 48 mucilage, which exerts protective and demulcent activity[31]. 49 50 Expectorant herbs containing saponins may reduce the surface tension of the secretions, 51 52 facilitating their separation from the mucous membranes. This induces reflex 53 54 stimulation which leads to an increase in the secretion of bronchial glands. Volatile-oil 55 type expectorant herbs exert a direct stimulatory effect on the bronchial glands by 56 57 means of local irritation. They may also have antibacterial activity. In colds and 58 59 60 6 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 22 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 influenza, herbs containing volatile oil can be used; also, volatile oils are ingredients of 4 5 syrups and liquids as well as external phytomedicines in the form of liniments, 6 ointments, and inhalations[32]. 7 8 9 Both limited research into possible mechanisms of action, and widespread use with 10 many testimonials to success, suggest that treatment with herbal medicines may have 11 12 great potential to treat respiratory diseases. The effectiveness of such treatment, 13 14 however, needs to be reviewed systematically and appraised critically to inform current 15 practice andFor direct future peer research. review only 16 17 18 In 2008, the Ministry of Health of Brazil has built a list (Renisus), with 71 species, pre- 19 selected by regions that alluded to its use for indications of use and according to the 20 21 categories of the International Classification of Diseases (ICD-10 ), with the potential to 22 23 advance the productive chain and generate products of interest to the public health 24 system in Brazil[33]. Moreover, in this same year, 36 species were recorded as herbal 25 26 medicines simplified registration (IN05) ie not need to prove through clinical trials the 27 28 effectiveness or safety record of its clinical indications only considered data from 29 tradition of popular use[34]. 30 31 32 A search in the database of the Brazilian Health Surveillance Agency, for herbal
33 http://bmjopen.bmj.com/ 34 medicines registered for commercialization indicated for the treatment of respiratory 35 diseases resulted in 15 species currently registered for this indication. Of these, six 36 37 belong to Renisus and nine to IN05. 38 39 40 Therefore, this study will investigate the effects of all marketed Brazilian medicinal
41 plants indicated to use in respiratory disease: Ananas comosus, Echinacea purpurea on September 27, 2021 by guest. Protected copyright. 42 43 Moench, Eucalyptus globules, Glycyrrhiza glabra L., Hedera helix L., Malva sylvestris 44 45 L., Mentha spp* (M. piperita or M. villosa), Mikania spp* (M. glomerata or M. 46 laevigata), Pelargonium sidoides D.C., Petasites hybridus L., Pimpinella anisum L., 47 48 Polygala senega L., Psychotria ipecacuanha (Brot.) Stokes, Sambucus nigra L. 49 50 51 52 OBJECTIVES 53
54 55 The primary objective is to address the safety and efficacy of Brazilian medicinal plants 56 to treat the common cold and upper respiratory tract infection and associated cough. 57 58 59 60 7 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 22 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 METHODS/DESIGN 4 5 Protocol and registration 6 7 Our protocol is registered on PROSPERO (CRD42014007057), Available from 8 http://www.crd.york.ac.uk/PROSPERO 9 10 11 Our review will conform to the PRISMA guidelines for reporting systematic reviews. 12 13 Inclusion criteria 14 15 Our eligibilityFor criteria peerare as follows: review only 16 Patients: Studies must include patients with adult or pediatric patients with upper 17 18 respiratory disease: the common cold, sinusitis, tonsillitis, otitis media, pharyngitis or 19 20 laryngitis; or symptoms arising from the upper part of the lower respiratory tract (either 21 secondary to upper respiratory tract symptoms - e.g. post-nasal drip - or to acute 22 23 bronchitis, bronchiolitis). 24 25 Interventions: study must include an arm in which patient are taking one of Brazilian 26 herbal medicine from any of the following plant preparation (whole, powder, extract, 27 28 standardized mixture) with one of select plants (Ananas comosus, Echinacea purpurea 29 30 Moench, Eucalyptus globules, Glycyrrhiza glabra L., Hedera helix L., Malva sylvestris 31 L., Mentha spp* (M. piperita or M. villosa), Mikania spp* (M. glomerata or M. 32
33 laevigata), Pelargonium sidoides D.C., Petasites hybridus L., Pimpinella anisum L., http://bmjopen.bmj.com/ 34 35 Polygala senega L., Psychotria ipecacuanha (Brot.) Stokes, Sambucus nigra L.) 36 37 38 Type of study and design: We will include (1) any comparison (randomized controlled 39 40 trials or observational study) including in an arm patients taking one of herbal medicine
41 listed above compared with an inert (placebo) or active control and open label, via any on September 27, 2021 by guest. Protected copyright. 42 43 route of administration. 44 45 46 Exclusion criteria 47 48 Trials that included more than 20% of patients with any of the following conditions will 49 50 exclude: chronic obstructive pulmonary disease (COPD), pneumonia, bronchiectasis, 51 cystic fibrosis, bronco-pulmonary dysplasia, asthma or tuberculosis; underlying 52 53 immunodeficiency or respiratory tract anatomical defect; acute respiratory distress 54 55 requiring mechanical ventilation, and in which results from the clearly eligible 56 population were not separately reported. Also if the population of study uses two of the 57 58 eligible plants we will exclude. 59 60 8 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 22 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 Search methods for primary studies: 5 6 We will search the CENTRAL MEDLINE, EMBASE, CINAHL, AMED, Web of 7 8 Science, Ovid Healthstar, Pubmed, Scielo and The Cochrane Central Register of 9 10 Controlled Trials (CENTRAL), which includes the Cochrane Airways Group 11 Specialized Trials Register. 12 13 We will restrict the search to human subjects but we will not restrict the searches or 14 15 inclusion Forcriteria to any peer specific languages. review only 16 For every eligible study we identify and for studies such as other review articles that we 17 18 identify that may have citations including eligible studies, one reviewer will examine 19 20 the reference list. We will obtain and evaluate the full text of any potentially eligible 21 studies thus identified and determine their eligibility as described below. 22 23 We will write to the principal authors of the identified trials and the pharmaceutical 24 25 companies involved in the production of medicinal herbs and inquire about additional 26 trials of which they are aware. 27 28 29 30 Outcome Measures 31 32 Outcomes 33 Outcomes of interest will include time to resolution of clinical symptoms and/or signs http://bmjopen.bmj.com/ 34 35 (where clinical symptoms and signs include cough, sputum production or activity 36 37 limitations). Also of interest will be severity of symptoms prior to resolution. Other 38 important outcomes will include hospitalization rates and duration of hospital stay, and 39 40 days receiving antibiotics. Finally, we will summarize data addressing quality of life or
41 on September 27, 2021 by guest. Protected copyright. 42 functional status (including number of days of disability that may be defined as days in 43 bed, days off work or days where patients were unable to undertake normal activities 44 45 during the illness), and adverse events. 46 47 48 49 Eligibility determination 50 51 We will also record minor and serious adverse effects of the intervention and the 52 53 proportion of patients requiring discontinuation of the herbal medicine. Four reviewers 54 authors, working in pairs (MC/MW; AM/LL) will independently screen potentially 55 56 relevant citations and if available abstracts and apply the selection criteria. We will 57 58 obtain full texts of all articles that either reviewer feels might be eligible. Two 59 60 9 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 22 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 reviewers (MC/CB) will independently assess the eligibility of each full-text article and 4 5 resolve disagreements by consensus. 6 To exclude duplicate articles, one member (MCS) will look through all eligible articles 7 8 and identify those in which one more authors are common. For such articles, detailed 9 10 review will determine if there is duplicate publication and, if there is, MCS will decide 11 which has the more complete data and. We will record the less complete as a duplicate 12 13 and abstract data only for the more complete. 14 15 For peer review only 16 Data extraction 17 18 The reviewers, working in pairs (MCS-MW and MCS-LL), will independently extract 19 the data, recording information regarding patients, methods, interventions, outcomes, 20 21 missing outcome data, and results using standardized, pretested, data extraction forms 22 23 with accompanying instructions. For articles published in abstract form only, or for 24 articles in which important information is missing, we will seek complete information 25 26 regarding methods and results from authors. Individually, reviewers will evaluate 2 27 28 articles and then check agreement with one another. This process will continue every 2 29 articles until reviewers are confident they can achieve very high rates of agreement. 30 31 Disagreements will be resolved through discussion with any unresolved issues referred 32 to another reviewer (GG). 33 http://bmjopen.bmj.com/ 34 35 Risk of bias in individual studies 36 37 For randomized trials, two reviewers independently will assess the risk of bias, 38 39 including sequence generation, allocation concealment, blinding, number of patients 40 with missing outcome data, selective outcome reporting, and other sources of bias using
41 on September 27, 2021 by guest. Protected copyright. 42 a modified version of the Cochrane collaboration risk of bias tool[35]. We will assess 43 44 the risk of bias of observational studies with a modified version of the Newcastle- 45 46 Ottawa instrument that includes confidence in assessment of exposure and outcome, 47 prognostic stratification, accuracy of outcomes assessment, and missing data.[36] 48 49 50 51 52 53 Confidence in pooled estimates of effect 54 55 We also independently rated the overall quality of evidence (confidence in effect 56 57 estimates) for each of the outcomes by using the Grading of Recommendations 58 59 60 10 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 22 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 [37-40] 3 Assessment, Development and Evaluation (GRADE) approach . We will make 4 5 separate ratings for bodies of evidence from randomized trials and bodies of evidence 6 from observational studies. In the GRADE approach, randomized trials begin as high- 7 8 quality evidence but may be rated down by 1 or more of 5 categories of limitations: risk 9 10 of bias, inconsistency, indirectness, imprecision, and reporting bias. Observational 11 studies begin as low quality evidence but can be rated up for a large effect size, 12 13 evidence of a dose-response gradient observational studies, or for consideration of all 14 plausible confounding. 15 For peer review only 16 17 Documentation of agreement 18 19 20 We will document chance-corrected agreement for i) eligibility and ii) risk of bias of 21 individual studies and iii. all GRADE rating (precision, consistency, directness, and 22 23 publication bias). To measures of agreement we will use Kappa statistical. Values of 24 25 kappa between 0.40 and 0.59 have been considered to reflect fair agreement, between 26 0.60 and 0.74 to reflect good agreement and 0.75 or more to reflect excellent agreement 27 28 [41]. 29 30 31 32 Data synthesis
33 http://bmjopen.bmj.com/ 34 Where meta-analysis is not appropriate (excessive heterogeneity of population, 35 intervention, comparator, outcome, or methodology), we will construct summary tables 36 37 and provide a narrative synthesis. When meta-analysis is appropriate, we will conduct 38 39 analyses for each herbal intervention separately. We will conduct an analysis for each 40 outcome of interest. For interventions and outcomes for which there are both
41 on September 27, 2021 by guest. Protected copyright. 42 randomized trials and observational studies available we will determine the confidence 43 44 in estimates for each body of evidence and conduct an analysis for the body of evidence 45 that warrants greater confidence. If the two bodies of evidence warrant similar 46 47 confidence, we will conduct analyses for both bodies of evidence. 48 49 Meta-analyses will be conducted using Comprehensive Meta-Analysis (Biostat, 50 51 Englewood, NJ, USA). We will use random effects meta-analyses[42], which are 52 53 conservative in that they consider both within and between studies differences in 54 calculating the error term used in the analysis For trials that report dichotomous 55 56 outcomes, we will calculate the pooled relative risk with associated 95% confidence 57 58 intervals. 59 60 11 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 22 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 When pooling across trials that report continuous outcomes using the same instrument, 4 5 we will calculate the weighted mean difference (WMD), which maintains the original 6 unit of measurement and represents the average difference between groups, with studies 7 8 weighted by the inverse of their variance. Once the WMD has been calculated, we will 9 10 contextualize this value by noting, when available, the corresponding minimally 11 important difference (MID) - the smallest change in instrument score that patients 12 13 perceive is important. 14 If studies reported the same construct using different measurement instruments, we will 15 For peer review only 16 calculate the standardized mean difference (SMD). The SMD expresses the intervention 17 18 effect in SD units, rather than the original units of measurement, with the value of a 19 20 SMD depending on both the size of the effect (the difference between means) and the 21 SD of the outcomes (the inherent variability among participants). For outcome measures 22 23 that have an established anchor-based MID, we will use this measure to convert the 24 25 SMD into an odds ratio (OR). We will complement this presentation by either 26 converting the SMD into natural units of a widely accepted instrument used to measure 27 28 changes in the domain of interest or, if such an instrument is not available, we will 29 30 substitute the MID for the SD (denominator) in the SMD equation, which will result in 31 more-readily interpretable MID units instead of SD units[43]. If an estimated of the 32
33 MID is not available we will use a statistical approach developed by Suissa [44]to http://bmjopen.bmj.com/ 34 35 provide a summary estimate of the proportion of patients who benefit from treatment 36 across all studies. The statistical approaches to enhancing the interpretability of results 37 38 of continuous outcomes outlined in this paragraph will use methods cited as well as 39 40 those described by Thorlund et. al.[45] Funnel plots will be created to explore possible
41 publication bias. on September 27, 2021 by guest. Protected copyright. 42 43 We will use recently developed approaches to address missing participant data for 44 45 dichotomous outcomes[46] and continuous outcomes[47]. We will only apply these 46 approaches to patient-important outcomes that meet the following criteria: 1) show a 47 48 significant treatment effect and 2) report sufficient missing participant data to 49 50 potentially introduce clinically important bias. Thresholds for important missing 51 participant data will be determined on an outcome-by-outcome basis. 52 53 54 55 56 57 58 59 60 12 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 22 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 Explaining heterogeneity 4 5 We hypothesize the following possible explanations for heterogeneity: (1) Doses 6 7 (higher vs lower) with expected larger effect with higher than lower doses; (2) Risk of 8 bias, with an expected larger effect in trials at high or unclear risk of bias versus trials at 9 10 low risk of bias; (3) Bacterial and viral illnesses, with larger effect in viral illnesses than 11 12 bacterial; (4) Age (adult versus pediatric) with postulated larger effect in pediatric. The 13 presence of heterogeneity will be investigated with the use of likelihood ratio test 14 15 statistic. AnyFor heterogeneity peer between reviewthe study results would only have been described and 16 17 tested to determine if it reached statistical significance using a chi-squared test. 18 19 20 21 Summarizing evidence 22 23 We will present results in Evidence Profiles (EP) as recommended by the GRADE 24 Working Group[48 49]. EPs provide succinct, easily digestible presentations of quality 25 26 of evidence and magnitude of effects. Our EPs will be constructed with the help of a 27 28 software program called GRADEpro (http://ims.cochrane.org/gradepro) to include the 29 following 7 elements: 1. A list of all important outcomes, both desirable and 30 31 undesirable; 2. a measure of the typical burden of these outcomes (e.g. control group, 32
33 estimated risk); 3. a measure of the difference between the risks with and without http://bmjopen.bmj.com/ 34 intervention; 4. the relative magnitude of effect; 5. numbers of participants and studies 35 36 addressing these outcomes, and follow-up time; 6. a rating of the overall confidence in 37 38 estimate of effect for each outcome and; 7. comments, which will include the MID if 39 available. 40
41 on September 27, 2021 by guest. Protected copyright. 42 Discussion 43 44 Our review will evaluate Brazilian herbal intervention for respiratory illness associated 45 cough, provide estimates of the effectiveness of treatments and their associated harms, 46 47 and evaluate the quality of the evidence in a thorough and consistent manner using the 48 49 GRADE approach [[50-52]. We will prioritize patient important outcomes. The results 50 of our systematic review will be of interest to of interest to public health and primary 51 52 care practitioners in Brazil. Our review will inform these practitioners about best 53 54 estimates of effect and confidence in those estimates for both effectiveness and safety of 55 herbal medicines and identify key areas for future research. 56 57 58 59 60 13 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 22 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 List of abbreviations 4 5 CAM = Complementary and alternative medicine 6 7 CCMs = Cough and cold medications 8 9 COPD = Chronic obstructive pulmonary disease 10 11 EP = Evidence Profiles 12 13 GRADE = Assessment, Development and Evaluation 14 15 IN05 = HerbalFor medicines peer simplified registrationreview only 16 17 MID = Minimally important difference 18 19 OR = Odds ratio 20 21 OTC =Over-the-counter 22 23 RENISUS = Ministry of Health of Brazil has built a list 24 25 SMD = Standardized mean difference 26 27 28 URTI = Upper respiratory tract infection 29 30 WMD = Weighted mean difference 31 32
33 http://bmjopen.bmj.com/ 34 35 Acknowledgements/ Funding: 36 This project is funded from a grant by CAPES Research Programme (grant) and is 37 Support Program Graduate Education Institutions Private - PROSUP - CAPES / UNISO 38 39 Contributorship Statement: 40 LCL is Principal Investigator and led the writing of the manuscript. GG is project manager and 41 on September 27, 2021 by guest. Protected copyright. 42 co-investigator and contributed to the writing and revision of the manuscript. MCOS, MW, 43 CBM, JCO and AMQ are co-investigators and contributed to the writing and revision of the 44 manuscript. All authors read and approved the final manuscript. 45 46 47 Competing interests 48 49 The authors declare that they have no competing interests. 50 51 52 Supplementary Material 53 Additional file 1: Search Strategy. 54 55 56 57 58 59 60 14 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 22 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 7 8 9 10 References 11 12 13 14 1. Goldman R. Canadian Paediatric Society. Treating cough and cold: Guidance for caregivers 15 of Forchildren and youth.peer Paediatr Childreview Health 2011;16(9):564-6 only 16 2. Arroll B. Common cold. Clinical evidence 2006(15):2006-14 17 3. Shields MD, Bush A, Everard ML, et al. BTS guidelines: Recommendations for the 18 assessment and management of cough in children. Thorax 2008;63 Suppl 3:iii1-iii15 19 doi: 10.1136/thx.2007.077370[published Online First: Epub Date]|. 20 4. Ramanuja S, Kelkar PS. The approach to pediatric cough. Annals of allergy, asthma & 21 immunology : official publication of the American College of Allergy, Asthma, & 22 Immunology 2010;105(1):3-8; quiz 9-11, 42 doi: 10.1016/j.anai.2009.11.011[published 23 Online First: Epub Date]|. 24 5. Cherry DK, Burt CW, Woodwell DA. National Ambulatory Medical Care Survey: 2001 25 summary. Advance data 2003(337):1-44 26 6. Gwaltney JM. Clinical significance and pathogenesis of viral respiratory infections. The 27 American journal of medicine 2002;112 Suppl 6A:13S-18S 28 7. ALA. American Lung Association. The common cold. http:// 29 www.lungusa.org/diseases/c&f02/cold.html (Accessed 10 Sep, 2013) 2013 30 8. Britt H, Miller GC, Knox S, et al. Bettering the evaluation and care of health - a study of 31 general practice activity (AIHW Cat. No. GEP-10). Australian Institute of Health and 32 Welfare. 2002 33 9. Pratter MR. Cough and the common cold: ACCP evidence-based clinical practice guidelines. http://bmjopen.bmj.com/ 34 Chest 2006;129(1 Suppl):72s-74s doi: 10.1378/chest.129.1_suppl.72S[published Online 35 First: Epub Date]|. 36 10. Chang AB, Glomb WB. Guidelines for evaluating chronic cough in pediatrics : Accp 37 evidence-based clinical practice guidelines. CHEST Journal 2006;129(1_suppl):260S- 38 83S doi: 10.1378/chest.129.1_suppl.260S[published Online First: Epub Date]|. 39 11. Hay AD, Wilson AD. The natural history of acute cough in children aged 0 to 4 years in 40 primary care: a systematic review. The British journal of general practice : the journal
41 on September 27, 2021 by guest. Protected copyright. of the Royal College of General Practitioners 2002;52(478):401-9 42 12. Mitra A, Hannay D, Kapur A, et al. The natural history of acute upper respiratory tract 43 infections in children. Primary health care research & development 2011;12(4):329-34 44 doi: 10.1017/s1463423611000193[published Online First: Epub Date]|. 45 13. Pratter MR. Cough and the common cold : Accp evidence-based clinical practice guidelines. 46 CHEST Journal 2006;129(1_suppl):72S-74S doi: 47 10.1378/chest.129.1_suppl.72S[published Online First: Epub Date]|. 48 49 14. Smith SM, Schroeder K, Fahey T. Over-the-counter (OTC) medications for acute cough in 50 children and adults in ambulatory settings. Cochrane database of systematic reviews 51 (Online) 2012;8:Cd001831 doi: 10.1002/14651858.CD001831.pub4[published Online 52 First: Epub Date]|. 53 15. Suara RO, Jr. CJ. Effect of zinc salts on respiratory syncytial virus replication. Antimicrob 54 Agents Chemother 2004;48(3):783-90 55 16. Goldman R. Canadian Paediatric Society. Drug Therapy and Hazardous Substances 56 Committee. Treating cough and cold: Guidance for caregivers of children and youth. 57 Paediatr Child Health 2011;16(9):564-6 58 59 60 15 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 22 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 17. D. H. Efficacy of zinc against common cold viruses: An overview. J Am Pharm Assoc 4 2004;44(5):594-603 5 18. Caruso TJ, Prober CG, Jr. GJ. Treatment of naturally acquired common colds with zinc: A 6 structured review. Clin Infect Dis 2007;45(5):569-74 7 19. Douglas RM, Hemila H, D'Souza R, et al. Vitamin C for preventing and treating the 8 common cold. Cochrane database of systematic reviews (Online) 2004(4):Cd000980 9 doi: 10.1002/14651858.CD000980.pub2[published Online First: Epub Date]|. 10 20. Arroll B, Kenealy T. Antibiotics for the common cold and acute purulent rhinitis. Cochrane 11 database of systematic reviews (Online) 2005(3):Cd000247 doi: 12 10.1002/14651858.CD000247.pub2[published Online First: Epub Date]|. 13 21. De Sutter AI, Lemiengre M, Campbell H. WITHDRAWN: Antihistamines for the common 14 cold. Cochrane database of systematic reviews (Online) 2009(4):Cd001267 doi: 15 10.1002/14651858.CD001267.pub2[publishedFor peer review Online First: only Epub Date]|. 16 22. De Sutter AI, van Driel ML, Kumar AA, et al. Oral antihistamine-decongestant-analgesic 17 combinations for the common cold. Cochrane database of systematic reviews (Online) 18 2012;2:Cd004976 doi: 10.1002/14651858.CD004976.pub3[published Online First: 19 Epub Date]|. 20 23. Cano Garcinuno A, Casares Alonso I, Rodriguez Barbero J, et al. [Prescription of systemic 21 cold and cough drugs to children 0-13 years old. An unresolved problem]. Anales de 22 pediatria (Barcelona, Spain : 2003) 2013;78(1):43-50 doi: 23 10.1016/j.anpedi.2012.04.003[published Online First: Epub Date]|. 24 24. Sen EF, Verhamme KM, Felisi M, et al. Effects of safety warnings on prescription rates of 25 cough and cold medicines in children below 2 years of age. British journal of clinical 26 pharmacology 2011;71(6):943-50 doi: 10.1111/j.1365-2125.2010.03860.x[published 27 Online First: Epub Date]|. 28 25. Mourdi N, Dubus JC, Bavoux F, et al. [Mucolytic drugs: towards a contraindication in 29 infants]. Archives de pediatrie : organe officiel de la Societe francaise de pediatrie 30 2010;17(6):735-6 doi: 10.1016/s0929-693x(10)70084-1[published Online First: Epub 31 Date]|. 32 26. Bricks LF, Leone C. [Use of medicines by children attending nursery schools]. Revista de 33 saude publica 1996;30(6):527-35 http://bmjopen.bmj.com/ 34 27. Eisenberg DM, Davis RB, Ettner SL, et al. Trends in alternative medicine use in the United 35 States, 1990-1997: results of a follow-up national survey. JAMA : the journal of the 36 American Medical Association 1998;280(18):1569-75 37 28. Vickers A. Recent advances: complementary medicine. BMJ (Clinical research ed) 38 2000;321:683-6 39 29. GUERRA PM, NODARI ORB, al. e. Biodiversidade: aspectos biológicos, geográficos, 40 legais e éticos. In: SIMÕES, M. O. Farmacognosia: da planta ao medicamento. 3. ed.
41 on September 27, 2021 by guest. Protected copyright. 42 Porto Alegre: UFRGS; Florianópolis: UFSC, 2001. p.15. 43 30. REUTERS. Brasil terá primeiro banco de dados de plantas medicinais. Folha Online , 44 Brasil, 2002. Disponível em:
41 on September 27, 2021 by guest. Protected copyright. 42 Online First: Epub Date]|. 43 48. Guyatt GH, Thorlund K, Oxman AD, et al. GRADE guidelines: 13. Preparing summary of 44 findings tables and evidence profiles-continuous outcomes. Journal of clinical 45 epidemiology 2013;66(2):173-83 doi: 10.1016/j.jclinepi.2012.08.001[published Online 46 First: Epub Date]|. 47 49. Guyatt GH, Oxman AD, Santesso N, et al. GRADE guidelines: 12. Preparing summary of 48 findings tables-binary outcomes. Journal of clinical epidemiology 2013;66(2):158-72 49 doi: 10.1016/j.jclinepi.2012.01.012[published Online First: Epub Date]|. 50 50. Guyatt GH, Oxman AD, Kunz R, et al. Going from evidence to recommendations. BMJ 51 (Clinical research ed) 2008;336(7652):1049-51 doi: 52 10.1136/bmj.39493.646875.AE[published Online First: Epub Date]|. 53 51. Guyatt GH, Oxman AD, Kunz R, et al. What is "quality of evidence" and why is it important 54 to clinicians? BMJ (Clinical research ed) 2008;336(7651):995-8 doi: 55 10.1136/bmj.39490.551019.BE[published Online First: Epub Date]|. 56 52. Guyatt GH, Oxman AD, Vist GE, et al. GRADE: an emerging consensus on rating quality 57 of evidence and strength of recommendations. BMJ (Clinical research ed) 58 59 60 17 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 22 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 2008;336(7650):924-6 doi: 10.1136/bmj.39489.470347.AD[published Online First: 4 Epub Date]|. 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40
41 on September 27, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 18 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 19 of 22 BMJ Open
1 2 3 Appendix A: Search Strategy
4 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 5 6 We will acquire eligible studies through a systematic search of CINAHL, EMBASE, 7 MEDLINE, AMED, Web of Science, Ovid Health Star, Cochrane Library, Pubmed and 8 Scielo and the Cochrane Central Registry of Controlled Trials. 9 10 11 Searches were conducted individually for each plant. Thus the descriptors for the plants 12 were: 13 Ananas comosus = Ananas comosus 14 Echinacea purpurea Moench = Echinacea purpurea 15 Eucalyptus globules = Eucalyptus globules 16 Glycyrrhiza glabra L. = Glycyrrhiza glabra 17 For peer review only 18 Mentha spp* (M. piperita or M. villosa) = Mentha piperita AND Mentha villosa 19 Mikania spp* (M. glomerata or M. laevigata) = Mikania. glomerata AND Mikania 20 laevigata 21 Pimpinella anisum L. = Pimpinella anisum 22 Polygala senega L. = Polygala senega 23 24 Sambucus nigra L. = Sambucus nigra 25 26 Database: CINAHL <1982 to 2014 Week > 27 1. (Name of the plant).mp 28 2. Select LIMITS “Human” 29 30 3. exp (NAME OF THE PLANT (Search as Keyword)) 31 32 Database: EMBASE <1980 to 2014 Week -> 33 1. (Name of the plant).mp 34 35 2. Select “Human” 36 3. exp (NAME OF THE PLANT (Search as Keyword)) 37 http://bmjopen.bmj.com/ 38 Database: Ovid MEDLINE(R) <1946 to 2014 Week> 39 40 1. (Name of the plant).mp 41 2. Select “Human” 42 3. exp (NAME OF THE PLANT (Search as Keyword)) 43 44 Database: AMED (Allied and Complementary Medicine) <1985 to 2013 Week 12> 45 46 1. (Name of the plant). mp on September 27, 2021 by guest. Protected copyright. 47 2. exp (NAME OF THE PLANT (Search as Keyword)) 48 3. human 49 4. exp HUMAN.mp. search as Keyword 50 5. and/ 1-2 51 52 53 Database: Web of Science < 1976 to 2014 Week> 54 1. (Name of the plant). Mp 55 2. Refine Results: HUMAN 56 57 58 Database: Ovid Health star <1966 to 2014 Week> 59 1. (Name of the plant) 60 2. Select LIMITS “Human” 3. exp (NAME OF THE PLANT (Search as Keyword))
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 20 of 22
1 2 3 Database: Cochrane Library < to 2014 Week>
4 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 5 1. (Name of the plant) 6 7 Database: Pubmed <1950 to 2014 Week> 8 1. (Name of the plant) 9 10 2. Select Species: “Humans” 11 12 Database: Scielo < to 2014 Week> 13 14 1. Ananas comosus AND humans 15 16 Database: Cochrane Central Registry of Controlled Trials 17 <1980 to 2014For Week > peer review only 18 1. (Name of the plant) 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 46 on September 27, 2021 by guest. Protected copyright. 47 48 49 50 51 52 53 54 55 56 57 58 59 60
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 8 8 9 9 9 9 8-9 8-9 AppendixA on page # # page on Reported Reported 12-13 12-13 12-13 12-13 10-11 10-11 10 10 10 10 9 8 4-7 2 1 up) report up) characteristics and (e.g., considered, years
BMJ Open http://bmjopen.bmj.com/ analysis. - analysis). - on September 27, 2021 by guest. Protected copyright.
foreach meta ) ) For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 2 registration registration informationincluding registration number. Describe the Describe methods of handling dataand combining ofresults studies, if done, includingmeasures of consistency I (e.g., State the State principal summarymeasures (e.g., risk ratio,means). difference in Describe methods Describe used for assessing risk studies of of bias individual (including specification of this whether was the study at done outcomeor andhow this level), information be usedto is in any datasynthesis. List and define List variables all for data which sought were (e.g., PICOS,fundingsources) any assumptionsand and simplificationsmade. Describe method Describe of dataextraction from reports(e.g., forms, piloted inindependently, duplicate) and any processes forand confirming obtaining datafrom investigators. State the State process for studies selecting (i.e., screening, eligibility, includedin systematic and,if review, applicable, includedin the meta Present full electronicPresent search strategyfor least one at database, including limits any used, such that it could be repeated. Describe all Describe information sources databases (e.g., of with dates coverage, contact study with authors to identify studies) additional in the search and datelast searched. Specify study Specify characteristics PICOS, (e.g., length of follow - Indicate Indicate review if a protocolexists, if andcan it where Webaccessed (e.g.,be address), and, if available, provide Provide anProvide statement explicit of questions being addressedwithreference to participants, interventions,comparisons, outcomes,study and (PICOS). design Describe the Describe rationale for review inthe the context of what already is known. Provide a structured a Provide summary including, asapplicable: background; objectives;datasources; study eligibilitycriteria, participants,and interventions; study appraisal and synthesis methods; results; limitations;conclusions and implicationsoffindings; key systematic registration review number. Identify the Identify report systematic asa meta-analysis, review, both.or Checklist item item Checklist For peer review language, publicationstatus) ascriteria used for eligibility, rationale. giving only
1 8 2 4 7 9 3 6 # 11 12 14 13
PRISMA2009 ChecklistPRISMA 2009 Checklist PRISMA2009 ChecklistPRISMA 2009 Checklist
Synthesisof results Summarymeasures Riskindividual in of bias studies Data Data items Data collection Data process 10 Study selection Study Search Search Information sourcesInformation Eligibility criteria Eligibility Protocol and registrationProtocol 5 METHODS METHODS Objectives Objectives Rationale Rationale INTRODUCTION INTRODUCTION Structuredsummary ABSTRACT ABSTRACT Title Title TITLE TITLE Section/topic Section/topic Page 21 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from
Page 22 of on page # # page on Reported Reported 14 14 NA NA 3 13 13 NA NA NA NA NA NA NA NA NA NA NA NA NA NA 13 13 13 13 . . - statement.org
www.prisma
Page 2 of 2 2 of Page Page 1 of 2 2 1 of Page BMJ Open http://bmjopen.bmj.com/ For more information, visit: For information, more on September 27, 2021 by guest. Protected copyright.
specified. -
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Describe sources Describe of funding for systematic the review support and (e.g., other supply of data); role of funders for the systematic review. Provide a a Provide general interpretationof the results thecontext in of other evidence, and implications for future research. Discussstudy limitations at and outcome(e.g., level risk review-level ofand at bias), (e.g., incomplete retrieval of identifiedresearch, reporting bias). Summarizethe main findings the strengthincluding ofevidencefor main each outcome; their consider relevance to Give results ofresults Give additional analyses, if (e.g.,done sensitivity subgroupor analyses, meta-regression Item [see 16]). Present results Present of any assessmentof risk across of (see Itembias studies 15). Present results Present of each meta-analysis done, including confidence intervalsand measures of consistency. For all outcomes Forall considered(benefits harms),or for study: each present, (a) simple summary datafor each groupintervention (b) effectestimates confidence and intervals, forest ideally witha plot. Present Present dataon risk of ofstudy and, each bias ifavailable, outcome any assessmentlevel (see item 12). For each study, Foreach present characteristics for which data extracted were (e.g., study PICOS,follow-upsize, period) and citations. provide the Give numbersGive screened,of studies assessed for eligibility, and includedin the review, reasons with for exclusions at stage, each flow ideally witha diagram. Describe methods Describe of additional analyses (e.g., sensitivity subgroupor analyses, meta-regression),if indicating done, Specify Specify assessment any riskofthat may bias cumulative affectof evidence(e.g., the publication bias,selective studies).reporting within Checklist item item Checklist key groups key (e.g., healthcare providers, users, and policymakers). For prewhich were peer review only
# 21 17 25 27 26 16 23
PRISMA2009 ChecklistPRISMA 2009 Checklist PRISMA2009 ChecklistPRISMA 2009 Checklist
Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group (2009). Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. PLoS Med 6(6): e1000097. From: Funding Funding FUNDING FUNDING Conclusions Limitations Summaryof evidence 24 DISCUSSION DISCUSSION Riskacross of bias studies 22 Synthesisof results Results of Results individual studies 20 Riskstudies within of bias 19 Study characteristics Study 18 Study selection Study RESULTS RESULTS Riskacross of bias studies 15 Section/topic Section/topic doi:10.1371/journal.pmed1000097 doi:10.1371/journal.pmed1000097 Additional Additional analysis Additional Additional analyses 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from
Brazilian medicinal plants to treat upper respiratory tract and bronchial illness: systematic review and meta-analyses- Study Protocol
For peer review only Journal: BMJ Open
Manuscript ID: bmjopen-2014-005267.R1
Article Type: Protocol
Date Submitted by the Author: 24-Jun-2014
Complete List of Authors: Lopes, Luciane; UNISO, Pharmacie Science; University of Sorocaba, Pharmaceutical Science Silva, Maria Carolina; University of Sorocaba, Pharmaceutical Science Motta, Cristiane; University of Sorocaba, Pharmaceutical Science Quirós, Antonio; Universidad de Córdoba, Biología Celular, Fisiología e Inmunología. Facultad de Medicina Biavatti, Maique; Federal University of Florianopolis, Pharmaceutical Department de Oliveira, Jardel; Secretaria Municipal de Saúde, Médico de Família e Comunidade, Especialista em Saúde da Família, Geriatria e Gerontologia Guyatt, Gordon; McMaster Unviersity, Department of Clinical Epidemiology and Biostatistics http://bmjopen.bmj.com/ Primary Subject Complementary medicine Heading:
Pharmacology and therapeutics, Respiratory medicine, Complementary Secondary Subject Heading: medicine
COMPLEMENTARY MEDICINE, RESPIRATORY MEDICINE (see Thoracic Keywords: Medicine), THERAPEUTICS, Herbal medicine < THERAPEUTICS
on September 27, 2021 by guest. Protected copyright.
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 49 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 Brazilian medicinal plants to treat upper respiratory tract and 4 5 bronchial illness: systematic review and meta-analyses- Study 6 7 Protocol 8 9 Luciane C Lopes 10 Pharmaceutical Sciences Post graduate Course, University of Sorocaba, UNISO , SP, Brazil. 11 [email protected] 12 13 Maria Carolina O. Silva 14 15 PharmaceuticalFor Sciences Postpeer graduate Course, review University of Sorocaba, only UNISO, SP, Brazil 16 [email protected] 17 18 Cristiane Bergamashi Motta 19 Pharmaceutical Sciences Post graduate Course, University of Sorocaba, UNISO, SP, Brazil 20 [email protected] 21 22 Antonio Macho Quirós 23 Biología Celular, Fisiología e Inmunología. Facultad de Medicina, Universidad de Córdoba, Spain 24 [email protected] 25 26 27 Maique Weber Biavatti 28 Pharmaceutical Department, Federal University of Florianopolis, SC, Brazil 29 [email protected] 30 31 Jardel Corrêa de Oliveira 32 Médico de Família e Comunidade, Especialista em Saúde da Família, Geriatria e Gerontologia,
33 Secretaria Municipal de Saúde, Florianópolis, SC, Brasil http://bmjopen.bmj.com/ 34 [email protected] 35 36 Gordon Guyatt 37 Department of Clinical Epidemiology and Biostatistics, McMaster University, Canada. 38 [email protected] 39
40
41 on September 27, 2021 by guest. Protected copyright. Contact information for the corresponding author: Luciane Cruz Lopes, 42 [email protected] 43 44 Rua Gomes Carneiro, 570 apto 141, Postal code: 13400-530 45 Piracicaba, SP, Brazil. 46 Telephone: 55 19 978-18441 47 Fax: 55 15 21017074 48 49 Financial support: This project is funded by governmental Program Graduate Education 50 Institutions - PROSUP - CAPES / UNISO 51 No conflict of interest 52 Word count: 2921 (without references) Figures: 0; tables: 0; references: 57 53 Additional file: 1 54 Keywords: Common cold, upper respiratory disease, Sputum, bronchial illness, Cough, 55 Systematic review, Meta-analysis, Brazilian medicinal plants 56 57 58 59 60 1 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 49 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 ABSTRACT 4 5 Introduction: Respiratory illness, often associated with cough and sputum, is 6 7 frequent. In Brazil, herbal medicines are often recommended as a first-line treatment for 8 respiratory illness. There exists uncertainty regarding the effectiveness of these 9 10 treatments. No systematic review has evaluated Brazilian medicinal plants to treat upper 11 12 respiratory tract and bronchial illness (URTI). 13 14 Methods and analysis: We will conduct a systematic review and if appropriate a 15 For peer review only 16 series of meta-analyses evaluating the safety and effectiveness of Brazilian medicinal 17 18 plants (BMP) for URTI. Eligible randomized controlled trials and observational studies 19 will enroll adult or pediatric patients presenting with upper respiratory tract and 20 21 bronchial illness treated by BMP approved in Brazilian Health Surveillance Agency 22 23 compared with placebo, no treatment, or an alternative therapy. Our search will include 24 Cochrane Central Register of Controlled Trials (CENTRAL) which contains the 25 26 Cochrane Acute Respiratory Illness Group’s Specialized Register; MEDLINE, 27 28 EMBASE; CINAHL; Web of Science; AMED; LILACS, CAB abstracts, clinical 29 trial.gov, WHO Trial Register and Brazilian thesis database (CAPES) without any 30 31 language restrictions. Outcomes of interest are time to resolution of clinical symptoms 32 and/or signs (cough, sputum production or activity limitations), severity of symptoms 33 http://bmjopen.bmj.com/ 34 prior to resolution and major/minor adverse events. Teams of reviewers will, 35 36 independently and in duplicate, screen titles and abstracts and complete full text to 37 38 determine eligibility. For eligible studies, reviewers will perform data abstraction and 39 assess risk of bias of eligible trials. When appropriate, we will conduct meta-analyses. 40
41 We also assess the quality of body of evidence (confidence in estimates of effect) for on September 27, 2021 by guest. Protected copyright. 42 43 each of the outcomes using the GRADE approach. 44 45 Ethics and Dissemination: The systematic review will be published in a peer- 46 47 reviewed journal. Brief reports of review findings will be disseminated directly to 48 49 appropriate audiences via email and other modes of communication. The review will 50 guide health care practice and policy in Brazil. 51 52 53 Register Protocol: Prospero CRD42014007057 54 55 56 57 58 59 60 2 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 49 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 STRENGTHS AND LIMITATIONS OF THIS STUDY 4 5 • This will be the first systematic review to assess Brazilian medicinal plants 6 7 approved in Brazilian Health Surveillance Agency (ANVISA) to treat upper 8 respiratory tract and bronchial illness associated with cough and sputum. 9 • The results of this systematic review will help clinicians in making decisions in 10 11 clinical practice and help patients with cough and sputum seeking effective and 12 safe treatment options. 13 • The methods of the review are state-of-art, including explicit eligibility criteria, 14 a comprehensive search, independent duplicate assessment of eligibility, and use 15 For peer review only 16 of the GRADE approach to assess confidence in estimates of effect including 17 independent assessment of risk of bias, precision, consistency, directness and 18 publication bias. We will make separate ratings for bodies of evidence from 19 20 randomized trials and bodies of evidence from observational studies. 21 • Because primary studies are likely to be limited to non-randomized studies and 22 randomized trials with a high risk of bias confidence in estimates is likely to be 23 24 low. Eligible studies will likely differ substantially in study design and outcome 25 measures. The exact constituents of the plants being tested are likely to be 26 associated with some uncertainty. 27 28 29 30 31 32
33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40
41 on September 27, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 3 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 49 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 INTRODUCTION 4 5 Use of herbal medicines is frequent, particularly in Brazil 6 7 8 In high-income countries there is increasing public interest in, and use of, 9 a wide range of therapies that lie outside the main stream of traditional Western medical 10 11 practice1 2. Complementary and alternative medicine (CAM) has grown rapidly over the 12 1 13 last two decades . In the United States, approximately 38 percent of adults and 14 approximately 12 percent of children are using some form of CAM3. In Brazil, of total 15 For peer review only 16 revenues of the pharmaceutical industry from sales of drugs in the period from 1996 to 17 4 18 the present, up to 25% came from preparations derived from plants . The government’s 19 decision to include herbal medicine in the list of publicly subsidized medicine in the 20 21 Brazilian Health System (SUS) may have contributed to an increase in expenditures on 22 23 herbal medicine in Brazil of 12% in 2012 over 2011, with total of $ 1.147 billion 24 dollars5. 25 26 27 The license approval process for herbal medicines vary across countries, 28 6 7 29 including wide variation in evidence of effectiveness required for licensing . . Some 30 countries, including Brazil, demand for licensing only evidence of long standing and 31 32 widespread use of a plant. In such countries, the extent of pharmacovigilance of
33 8 http://bmjopen.bmj.com/ 34 licensed products differs; relatively rigorous pharmacovigilance exists in Australia , 35 Canada , Germany, among others, but not in Brazil 7 9. 36 37 38 In many countries, traditional herbal medicines are available over-the- 39 3 40 counter (ie no need for a prescription for their purchase or use ). These medications are
41 typically not recommended for serious medical conditions, but rather as adjunctive on September 27, 2021 by guest. Protected copyright. 42 43 treatments and for short-term use in conditions that are not serious10 11. Aside from 44 45 Brazil, there is no country that provides public support for payment for herbal medicines 46 approved only on the basis of long standing and widespread prior use. Nowadays Brazil 47 48 has a list of 12 such herbal medicines funded by the government.12 13. 49 50 51 Primary care physicians often recommend herbal medicines to their 52 patients as first line of treatment14 This is particularly the case in Brazil, perhaps 53 54 encouraged by government funding for these drugs. Furthermore, people frequently 55 56 self-prescribe over-the-counter cough medications. One reason for concern about this 57 58 59 60 4 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 49 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 widespread use is that patients are less likely to consult their general practitioner 4 15 5 because of an adverse reaction to an herbal remedy than for a conventional medicine . 6 7 Respiratory Illness and herbal medicine 8 9 10 Respiratory illness, in particular upper respiratory tract and bronchial 11 illness (URTI), often with associated cough and sputum, are frequent16 and a major 12 13 cause of morbidity, especially in children and the elderly 17. Although in most cases 14 18 19 15 benign , Forrespiratory illnesspeer is a cause review for concern for parents only and a major cause of 16 outpatient visits in most settings 20 21 22 23. URTI can adversely impact on quality of 17 18 life24. Patients spend billions of dollars annually on OTC medications for URTI, and in 19 25 20 particular for the frequently accompanying cough symptoms . 21 Numerous OTC cough preparations are available but a Cochrane review 22 23 that did not address the plants that are the topic of the current review suggests there is 24 26 27 25 no conclusive evidence regarding their efficacy . In children, OTC medications may 26 be associated with serious adverse events such as death, altered consciousness and 27 28 arrhythmias28-32. 29 30 31 A search in the database of the Brazilian Health Surveillance Agency 32 (ANVISA) revealed that 15 species of herbal medicines are approved for treatment of 33 http://bmjopen.bmj.com/ 34 acute cough from an URTI. Of these, Public Health System (SUS) funding is available 35 36 for two. There are no systematic reviews available addressing the benefits and harms of 37 the herbal medication approved by ANVISA for URTI. Identification of ineffective 38 39 preparations could reduce costs for consumers and healthcare providers, and reduce the 40 risk of adverse events from treatments with no benefit27. This current systematic review
41 on September 27, 2021 by guest. Protected copyright. 42 therefore aims to collect the evidence to evaluate the effectiveness and safety of 15 43 44 Brazilian herbal medicines currently approved to management cough from an upper 45 46 respiratory tract and bronchial illness. 47 48 49 50 51 52 53 54 55 56 57 58 59 60 5 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 49 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 OBJECTIVES 7 8 The primary objective is to address the safety and efficacy of 15 Brazilian herbal 9 medicines approved by ANVISA for acute cough from upper respiratory tract and 10 11 bronchial illness. 12 13 14 METHODS AND ANALYSES 15 For peer review only 16 The systematic review will be performed according to the recommendations specified in 17 the Cochrane Handbook for Intervention Reviews33. The reporting of the review will 18 19 follow the Preferred Reporting Items for Systematic Reviews and Meta-Analyses 20 34 21 (PRISMA) statement . 22 Protocol and registration 23 24 Our protocol is registered on PROSPERO (CRD42014007057), and is available from 25 26 http://www.crd.york.ac.uk/PROSPERO 27 28 Eligibility criteria for considering studies for review 29 30 Inclusion criteria 31 32
33 Patients: Studies must include patients with adult (>18 years old) or pediatric (0-18 http://bmjopen.bmj.com/ 34 years old) patients with upper respiratory disease: the common cold, sinusitis, tonsillitis, 35 36 otitis media, pharyngitis or laryngitis; or symptoms arising from the upper part of the 37 38 lower respiratory tract (either secondary to upper respiratory tract symptoms - e.g. post- 39 nasal drip - or to acute bronchitis or bronchiolitis). 40
41 on September 27, 2021 by guest. Protected copyright. 42 Interventions: Studies must include an arm in which patient are taking one of Brazilian 43 44 herbal medicine from any of the following plant preparation (whole, powder, extract, 45 standardized mixture) with one of select plants: 46 47 Ananas comosus (L.) Merr., Bromeliaceae; 48 Echinacea purpurea (L.) Moench, Asteraceae; 49 50 Eucalyptus globulus Labill., Myrtaceae; 51 Glycyrrhiza glabra L., Fabaceae; 52 53 Hedera helix L., Araliaceae; 54 55 Malva sylvestris L., Malvaceae; 56 Mentha spp (Mentha x piperita L., Mentha x villosa Huds., or other hybrids), Lamiaceae; 57 58 Mikania glomerata Spreng.or Mikania laevigata Sch.Bip. ex Baker, Asteraceae; 59 60 6 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 49 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 Pelargonium sidoides DC., Geraniaceae; 4 Petasites hybridus (L.) G. Gaertn., B. Mey. & Scherb., Asteraceae; 5 6 Pimpinella anisum L., Apiaceae; 7 Polygala senega L., Polygalaceae; 8 9 Psychotria ipecacuanha (Brot.) Stokes or Cephaelis ipecacuanha (Brot.) A. Rich., Rubiaceae; 10 Sambucus nigra L., Adoxaceae. 11 12 13 14 Outcome Measures 15 For peer review only 16 We will include studies that report any of the following outcomes: 17 18 - time to resolution of clinical symptoms and/or signs (e.g. cough, sputum 19 production). 20 21 - frequency and severity of symptoms prior to resolution. 22 23 - minor and serious adverse effects of the intervention and the proportion of 24 patients requiring discontinuation of the herbal medicine 25 26 - hospitalization rates 27 28 - duration of hospital stay 29 - days receiving antibiotics 30 31 - functional status (including number of days of disability that may be defined as 32 days in bed, days off work or days when patients were unable to undertake 33 http://bmjopen.bmj.com/ 34 normal activities) 35 36 - Quality-of-life measured by a validated instrument. The score will be evaluated 37 24 38 using Cough-Specific Quality-of-Life Questionnaire , Leicester Cough 39 Questionnaire 35 or other validated questionnaires. 40
41 on September 27, 2021 by guest. Protected copyright. 42 Study Design 43 44 We will include (1) any comparison (randomized controlled trials or observational 45 46 study) including in an arm in which patients are taking one of the herbal medicines 47 listed above via any route of administration compared to an arm in which patients 48 49 receive with an inert (placebo) or no treatment or an active non-herbal medicine control, 50 51 or case-control studies comparing the frequency of use of herbal medicine in patients 52 with better versus poorer outcomes of their respiratory illness. 53 54 55 We will exclude studies in which more than 20% of participating patients suffered from 56 57 one or more of the following conditions in which results from the eligible population 58 59 60 7 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 49 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 were not separately reported: chronic obstructive pulmonary disease (COPD), 4 5 pneumonia, bronchiectasis, cystic fibrosis, bronco-pulmonary dysplasia, asthma or 6 tuberculosis; underlying immunodeficiency or respiratory tract anatomical defect; acute 7 8 respiratory distress requiring mechanical ventilation. Also, we will exclude studies that 9 10 investigate the simultaneous use of more than one the eligible plants. 11 12 13 Search methods for primary studies: 14 15 ElectronicFor searches peer review only 16 We will search the following electronic databases irrespective of language or 17 18 publication status: Cochrane Central Register of Controlled Trials (CENTRAL) which 19 20 contains the Cochrane Acute Respiratory Infections Group’s Specialized Register ; 21 MEDLINE; EMBASE; CINAHL (Cumulative Index to Nursing and Allied Health 22 23 Literature); Web of Science; Health Star (via OVID); AMED, the database of the 24 25 Cochrane Complementary Medicine Field, LILACS; CAB abstracts, clinical trial.gov, 26 WHO Trial Register36 and Brazilian thesis database (CAPES). 27 28 29 Searching other resources 30 For every eligible study we identify and for studies such as review articles that may 31 32 have citations including eligible studies, one reviewer will examine the reference list.
33 http://bmjopen.bmj.com/ 34 We will write to the principal authors of the identified trials and the pharmaceutical 35 companies involved in the production of medicinal herbs and inquire about additional 36 37 trials of which they are aware. 38 39 Unpublished studies will be identified by searching in the books including List of 40 references for evaluation of safety and efficacy of herbal medicines described in the
41 on September 27, 2021 by guest. Protected copyright. 42 Brazilian legislation for herbal medicines in Brazil and conference proceedings 43 44 (Medicinal Symposium of Brazilian medicinal plants; International Congress of 45 Ethnopharmacology). 46 47 48 The following Brazilian scientific journals will also be scanned manually for eligible 49 50 studies: Journal of Basic and Applied Pharmaceutical Sciences; Brazilian Journal of 51 Pharmacy; Brazilian Journal of Pharmacognosy; Brazilian Journal of Medicinal Plants; 52 53 Brazilian Journal of Pharmaceutical Sciences. 54 55 56 Search strategy 57 58 59 60 8 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 49 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 We will restrict the search to human subjects but we will not restrict the searches or 4 5 inclusion criteria to any specific languages. We stated the search strategy in in 6 Appendix section to search MEDLINE and CENTRAL. We will not combine the 7 8 MEDLINE search with the Cochrane Highly Sensitive Search Strategy for identifying 9 37 10 randomized trials in MEDLINE because we think will find only few results. We will 11 adapt the search string to search EMBASE, CINAHL, LILACS and Web of Science. 12 13 The search will be conducted using individually for each plant. The following terms will 14 be used to: 1) intervention: medicine, herbal; plants, plant; extracts; medicinal; 15 For peer review only 16 medicine, traditional; herb$; phytomedicine; phytotherapy; complementary therap*; 17 18 complementary Medicine*; alternative therap*; traditional medicine*; ethnomedicine*; 19 20 ethnobotany; ethnopharmacology; oriental traditional medicine*; scientific name of 21 plant, synonymies of each medicinal plants; popular name of each medicinal plant 22 23 selected; 2)Condition: respiratory tract diseases, respirat*, cough*, sputum; bronchial 24 25 illness. The complete search strategy is available in Appendix 1. 26 27 Eligibility determination 28 29 30 Four reviewers, working in pairs, will independently screen potentially relevant 31 citations and if available abstracts and apply the selection criteria. We will obtain full 32
33 texts of all articles that either reviewer feels might be eligible. Two reviewers will http://bmjopen.bmj.com/ 34 35 independently assess the eligibility of each full-text article and resolve disagreements by 36 consensus. 37 38 To exclude duplicate articles, one reviewer will examine all eligible articles and identify 39 40 those that have one or more authors in common. For such articles, detailed review will
41 on September 27, 2021 by guest. Protected copyright. determine if there is duplicate publication and, if there is, we will use the article with the 42 43 more complete data. 44 45 46 Data extraction 47 48 The reviewers, working in pairs, will independently extract the data, recording 49 information regarding patients, methods, interventions, outcomes, missing outcome 50 51 data, and results using standardized, pretested, data extraction forms with accompanying 52 53 instructions. For articles published in abstract form only, or for articles in which 54 55 important information is missing, we will seek complete information regarding methods 56 and results from authors. Individually, reviewers will evaluate 2 articles and then check 57 58 agreement with one another. This process will continue every 2 articles until reviewers 59 60 9 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 49 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 are confident they can achieve very high rates of agreement. Disagreements will be 4 5 resolved through discussion with any unresolved issues referred to another reviewer. 6 7 Risk of bias in individual studies 8 9 For randomized trials, two reviewers will independently assess the risk of bias, 10 11 including sequence generation, allocation concealment, blinding, number of patients 12 13 with missing outcome data, selective outcome reporting, and other sources of bias using 14 a modified version of the Cochrane collaboration risk of bias tool38. We will assess the 15 For peer review only 16 risk of bias of observational studies with a modified version of the Newcastle-Ottawa 17 18 instrument that includes confidence in assessment of exposure and outcome, adjusted 19 analysis for differences between groups in prognostic characteristics, accuracy of 20 39 21 outcomes assessment, and missing data. 22 23 24 Confidence in pooled estimates of effect 25 26 We will also independently rate the quality of evidence (confidence in effect estimates) 27 for each of the outcomes by using the Grading of Recommendations Assessment, 28 29 Development and Evaluation (GRADE) approach 40-43. We will make separate ratings 30 31 for bodies of evidence from randomized trials and bodies of evidence from 32 observational studies. In the GRADE approach, randomized trials begin as high quality
33 http://bmjopen.bmj.com/ 34 evidence but may be rated down by 1 or more of 5 categories of limitations: risk of bias, 35 36 inconsistency, indirectness, imprecision, and reporting bias. Observational studies 37 begin as low quality evidence but can be rated up for a large effect size, evidence of 38 39 dose-response gradient observational studies, or for consideration of all plausible 40 confounding.
41 on September 27, 2021 by guest. Protected copyright. 42 43 Documentation of agreement 44 45 We will document chance-corrected agreement for i) eligibility and ii) risk of bias of 46 47 individual studies. To measures of agreement we will use Kappa statistical. Values of 48 49 kappa between 0.40 and 0.59 have been considered to reflect fair agreement, between 50 0.60 and 0.8 to reflect good agreement and 0.75 or more to reflect excellent 51 52 agreement44. 53 54
55 Data synthesis 56 Where meta-analysis is not appropriate (excessive heterogeneity of population, 57 58 intervention, comparator, outcome, or methodology), we will construct summary tables 59 60 10 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 49 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 and provide a narrative synthesis. When meta-analysis is appropriate, we will conduct 4 5 analyses for each herbal intervention separately for each outcome of interest. For 6 interventions and outcomes for which there are both randomized trials and observational 7 8 studies available we will determine the confidence in estimates for each body of 9 10 evidence and conduct an analysis for the body of evidence that warrants greater 11 confidence. If the two bodies of evidence warrant similar confidence, we will conduct 12 13 analyses for both bodies of evidence. 14 15 Meta-analysesFor will bepeer conducted review using Comprehensive only Meta-Analysis (Biostat, 16 17 Englewood, NJ, USA). We will use random effects meta-analyses45, which are 18 19 conservative in that they consider both within and between studies differences in 20 calculating the error term used in the analysis For trials that report dichotomous 21 22 outcomes, we will calculate the pooled relative risk with associated 95% confidence 23 24 intervals. In the cross-sectional and case-control studies we will document the 25 proportion of patients in the intervention and control groups who experience each of the 26 27 outcomes of interest. For case-control studies, we will use odds ratios rather than 28 29 relative risks. 30 When pooling across trials that report continuous outcomes using the same instrument, 31 32 we will calculate the weighted mean difference (WMD), which maintains the original
33 http://bmjopen.bmj.com/ 34 unit of measurement, with studies weighted by the inverse of their variance. Once the 35 WMD has been calculated, we will contextualize this value by noting, when available, 36 37 the corresponding minimally important difference (MID) - the smallest change in 38 39 instrument score that patients perceive is important. 40 If studies reported the same construct using different measurement instruments, we will
41 on September 27, 2021 by guest. Protected copyright. 42 calculate the standardized mean difference (SMD). The SMD expresses the intervention 43 44 effect in SD units, rather than the original units of measurement, with the value of a 45 SMD depending on both the size of the effect (the difference between means) and the 46 47 SD of the outcomes (the inherent variability among participants). For outcome measures 48 49 that have an established anchor-based MID, we will use this measure to convert the 50 SMD into an odds ratio (OR) and risk difference. We will complement this presentation 51 52 by either converting the SMD into natural units of a widely accepted instrument used to 53 54 measure changes in the domain of interest or, if such an instrument is not available, we 55 will substitute the MID for the SD (denominator) in the SMD equation, which will 56 57 result in more-readily interpretable MID units instead of SD units46. If an estimate of the 58 59 60 11 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 49 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 47 3 MID is not available we will use a statistical approach developed by Suissa to provide 4 5 a summary estimate of the proportion of patients who benefit from treatment across all 6 studies. The statistical approaches to enhancing the interpretability of results of 7 8 continuous outcomes outlined in this paragraph will use methods cited as well as those 9 48 10 described by Thorlund et. al. 11 Funnel plots will be created to explore possible publication bias. 12 13 We will use recently developed approaches to address missing participant data for 14 dichotomous outcomes49 and continuous outcomes50. We will only apply these 15 For peer review only 16 approaches to patient-important outcomes that meet the following criteria: 1) show a 17 18 significant treatment effect and 2) report sufficient missing participant data to 19 20 potentially introduce clinically important bias. Thresholds for important missing 21 participant data will be determined on an outcome-by-outcome basis. 22 23 24 25 Explaining heterogeneity 26 27 We hypothesize the following possible explanations for heterogeneity: (1) Doses 28 (higher vs lower) with expected larger effect with higher doses; (2) Risk of bias, with an 29 30 expected larger effect in trials at high or unclear risk of bias versus trials at low risk of 31 32 bias; (3) Bacterial and viral illnesses, with larger effect in viral illnesses than bacterial;
33 (4) Age (adult versus pediatric) with postulated larger effect in pediatric patients. The http://bmjopen.bmj.com/ 34 35 presence of heterogeneity will be investigated with the use of chi-squared test statistic 36 2 37 and the I statistic – the percentage of variability that is due to true differences between 38 studies (heterogeneity) rather than sampling error (chance).51 52 39 40
41 on September 27, 2021 by guest. Protected copyright. 42 Summarizing evidence 43 44 We will present results in Evidence Profiles (EP) as recommended by the GRADE 45 53 54 46 Working Group . EPs provide succinct, easily digestible presentations of quality of 47 evidence and magnitude of effects. Our EPs will be constructed with the help of a 48 49 software program, GRADEpro (http://ims.cochrane.org/gradepro) to include the 50 51 following 7 elements: 1. A list of all important outcomes, both desirable and 52 undesirable; 2. a measure of the typical burden of these outcomes (e.g. control group, 53 54 estimated risk); 3. a measure of the difference between the risks with and without 55 56 intervention; 4. the relative magnitude of effect; 5. numbers of participants and studies 57 addressing these outcomes, and follow-up time; 6. a rating of the overall confidence in 58 59 60 12 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 49 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 estimate of effect for each outcome and; 7. comments, which will include the MID if 4 5 available. 6 7 DISCUSSION 8 9 Our review will evaluate Brazilian herbal intervention for respiratory illness associated 10 11 with cough, provide estimates of the effectiveness of treatments and their associated 12 13 harms, and evaluate the quality of the evidence in a thorough and consistent manner 14 using the GRADE approach55-57. We will prioritize patient important outcomes. The 15 For peer review only 16 results of our systematic review will be of interest to of interest to public health and 17 18 primary care practitioners in Brazil. Our review will inform these practitioners about 19 best estimates of effect and confidence in those estimates for both effectiveness and 20 21 safety of herbal medicines for URTI and identify key areas for future research. 22 23 24 List of abbreviations 25 CAM = Complementary and alternative medicine 26 27 COPD = Chronic obstructive pulmonary disease 28 29 30 EP = Evidence Profiles 31 32 GRADE = Assessment, Development and Evaluation
33 http://bmjopen.bmj.com/ 34 MID = Minimally important difference 35 36 OR = Odds ratio 37 38 OTC =Over-the-counter 39 40 SMD = Standardized mean difference
41 on September 27, 2021 by guest. Protected copyright. 42 URTI = Upper respiratory tract and bronchial illness 43 44 WMD = Weighted mean difference 45 46 47 Authors’ contributions 48 49 LCL is Principal Investigator and led the writing of the manuscript. GG is project 50 manager and co-investigator and contributed to the writing and revision of the 51 manuscript. MCOS, MW, CBM and AMQ are co-investigators and contributed to the 52 writing and revision of the manuscript. All authors read and approved the final 53 manuscript. 54 55 56 57 58 59 60 13 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 49 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 Competing interests 4 5 The authors declare that they have no competing interests. 6 7 8 Acknowledgements/ Funding: 9 This project is funded from a governmental Program Graduate Education Institutions - 10 11 PROSUP - CAPES / UNISO (Process Number: 349.984.188-66) 12 13 References 14 1. Eisenberg DM, Davis RB, Ettner SL, et al. Trends in alternative medicine use in the United 15 States,For 1990-1997: peer results of a follow-up review national survey. onlyJAMA : the journal of the 16 American Medical Association 1998;280(18):1569-75. 17 18 2. Vickers A. Recent advances: complementary medicine. BMJ (Clinical research ed) 19 2000;321:683-6. 20 3. Ekor M. The growing use of herbal medicines: issues relating to adverse reactions and 21 challenges in monitoring safety. Front Pharmacol 2014;4(177):177. 22 4. GUERRA PM, NODARI ORB, al. e. Biodiversidade: aspectos biológicos, geográficos, legais e 23 éticos. In: SIMÕES, M. O. Farmacognosia: da planta ao medicamento. 3. ed. Porto 24 Alegre: UFRGS; Florianópolis: UFSC, 2001. p.15. 25 5. Vale ASY. Growing consumption of herbal medicines. 26 http://impressaodigital126combr/2013/09/11/incentivos-e-crescimento-do-consumo- 27 marcam-segmento-de-fitoterapicos/ ( Acessed in June 7) 2012. 28 6. BRASIL. Provides for registration of herbal medicines and registration and notification of 29 traditional herbal products. RDC No. 26. 9 Maio 2014. DIARIO OFICIAL DA UNIÃO - 30 31 SEÇÃO 1 13 DE Brasília, DF 2014. 32 7. WHO. National policy on traditional medicine and regulation of herbal medicines: Report of
33 a WHO global survey. WHO Library Cataloguing-in-Publication Data http://bmjopen.bmj.com/ 34 http://appswhoint/medicinedocs/pdf/s7916e/s7916epdf (Accessed 7 June 2014) 2005. 35 8. DR B. The regulation of herbal medicines in Australia. Toxicology 2002;181-182:565-70. 36 9. WHO. R. Regulatory Situation of Herbal Medicines. A Worldwide Review (WHO/trm/98.1) 37 Geneva: 1998. World Health Organization. 38 http://appswhoint/medicinedocs/pdf/whozip57e/whozip57epdf (Accessed 7 June 39 2014) 1998. 40 10. Harrington P, Shepherd MD. Analysis of the movement of prescription drugs to over-the-
41 counter status. Journal of managed care pharmacy : JMCP 2002;8(6):499-508; quiz 09- on September 27, 2021 by guest. Protected copyright. 42 11. 43 44 11. Cohen JP, Paquette C, Cairns CP. Switching prescription drugs to over the counter. BMJ 45 (Clinical research ed) 2005;330(7481):39-41. 46 12. BRASIL. Fitoterápicos são alternativa de tratamento no SUS. Portal do Ministério da Saúde. 47 Disponível em: http://portalsaudesaudegovbr/portalsaude/noticia/8061/162/sus- 48 oferece-fitoterapicos-como-alternativa-de-tratamentohtml 2012. 49 13. BRASIL. Agência Nacional de Vigilância Sanitária. Instrução Normativa n° 05, de 11 de 50 Dezembro de 2008, IN05/2008. Lista de Medicamentos Fitoterápicos de Registro 51 Simplificado. Brasília, DF. 2008b. 52 14. Medicine AACP. Over-the-counter medications: use in general and special populations, 53 therapeutic errors, misuse, storage and disposal. 54 http://cymcdncom/sites/wwwacpmorg/resource/resmgr/timetools- 55 files/otcmedsclinicalreferencepdf (acessed 7 of June 2014) 2011. 56 57 58 59 60 14 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 49 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 15. Barnes J, Mills SY, Abbot NC, et al. Different standards for reporting ADRs to herbal 4 remedies and conventional OTC medicines: face-to-face interviews with 515 users of 5 herbal remedies. British journal of clinical pharmacology 1998;45(5):496-500. 6 16. Goldman R. Canadian Paediatric Society. Treating cough and cold: Guidance for caregivers 7 of children and youth. Paediatr Child Health 2011;16(9):564-6. 8 17. Arroll B. Common cold. Clinical evidence 2006(15):2006-14. 9 18. Butler CC, Hood K, Kinnersley P, et al. Predicting the clinical course of suspected acute viral 10 upper respiratory tract infection in children. Family practice 2005;22(2):92-5. 11 19. Hay AD, Wilson A, Fahey T, et al. The duration of acute cough in pre-school children 12 13 presenting to primary care: a prospective cohort study. Family practice 14 2003;20(6):696-705. 15 20. GonzalesFor R, Sande MA.peer Uncomplicated review acute bronchitis. Ann Internonly Med 2000;133(12):981- 16 91. 17 21. Kusel MM, de Klerk N, Holt PG, et al. Occurrence and management of acute respiratory 18 illnesses in early childhood. Journal of paediatrics and child health 2007;43(3):139-46. 19 22. Britt H, Miller GC, Knox S, et al. Bettering the evaluation and care of health. A study of 20 general practice activity (AIHW Cat. No. GEP-10). Australian Institute of Health and 21 Welfare. http://wwwaihwgovau/WorkArea/DownloadAssetaspx?id=6442455894 22 (Accessed 4 June 2014) 2002. 23 23. Cherry DK, Burt CW, Woodwell DA. National Ambulatory Medical Care Survey: 2001 24 summary. Advance data 2003(337):1-44. 25 26 24. French CT, Irwin RS, Fletcher KE, et al. Evaluation of a cough-specific quality-of-life 27 questionnaire. Chest 2002;121(4):1123-31. 28 25. Dicpinigaitis PV, Colice GL, Goolsby MJ, et al. Acute cough: a diagnostic and therapeutic 29 challenge. Cough (London, England) 2009;5(11):11. 30 26. Chang CC, Cheng AC, Chang AB. Over-the-counter (OTC) medications to reduce cough as an 31 adjunct to antibiotics for acute pneumonia in children and adults. Cochrane database 32 of systematic reviews (Online) 2012;2:CD006088.
33 27. Smith SM, Schroeder K, Fahey T. Over-the-counter (OTC) medications for acute cough in http://bmjopen.bmj.com/ 34 children and adults in ambulatory settings. Cochrane database of systematic reviews 35 (Online) 2012;8:CD001831. 36 28. Gunn VL, Taha SH, Liebelt EL, et al. Toxicity of over-the-counter cough and cold 37 medications. Pediatrics 2001;108(3):E52. 38 29. Centers for Disease C, Prevention. Infant deaths associated with cough and cold 39 40 medications--two states, 2005. MMWR Morbidity and mortality weekly report 2007;56(1):1-4.
41 on September 27, 2021 by guest. Protected copyright. 42 30. Centers for Disease C, Prevention. Revised product labels for pediatric over-the-counter 43 cough and cold medicines. MMWR Morbidity and mortality weekly report 44 2008;57(43):1180. 45 31. Kuehn BM. Debate continues over the safety of cold and cough medicines for children. 46 JAMA : the journal of the American Medical Association 2008;300(20):2354-6. 47 32. Mallet P, Mourdi N, Dubus JC, et al. Respiratory paradoxical adverse drug reactions 48 associated with acetylcysteine and carbocysteine systemic use in paediatric patients: a 49 national survey. PloS one 2011;6(7):e22792. 50 33. http://www.cochrane.org/training/ CHfSRoITCC, 2014) c-hAJ. 2014. 51 34. Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items for systematic reviews and 52 53 meta-analyses: the PRISMA statement. Ann Intern Med 2009;151(264). 54 35. Birring SS, Prudon B, Carr AJ, et al. Development of a symptom specific health status 55 measure for patients with chronic cough: Leicester Cough Questionnaire (LCQ). Thorax 56 2003;58(4):339-43. 57 36. WHO. International Clinical Trials Registry Platform (ICTRP) 58 (http://apps.who.int/trialsearch/) 59 60 15 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 49 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 37. Lefebvre C, Manheimer E, J. G. Chapter 6: Searching for studies. In: Higgins JPT, Green S 4 editor(s). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 5 [updated March 2011]. The Cochrane Collaboration. Available from www.cochrane- 6 handbook.org. Chichester, UK: Wiley-Blackwell. 2011. 7 38. Therapeutic Goods Administration-Australian Government’s Department of Health and 8 Ageing. Guideline on clinical investigation of medicinal products indicated for the 9 treatment of psoriasis [online]. Available from: 10 http://wwwtgagovau/docs/pdf/euguide/ewp/245402enpdf 2010. 11 39. Wells GA, Shea B, O'Connell D, et al. The Newcastle-Ottawa Scale (NOS) for assessing the 12 13 quality of nonrandomised studies in meta-analyses. Secondary The Newcastle-Ottawa 14 Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses 2013. 15 http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp.For peer review only 16 40. Guyatt GH, Oxman AD, Kunz R, et al. GRADE guidelines: 7. Rating the quality of evidence-- 17 inconsistency. Journal of clinical epidemiology 2011;64(12):1294-302. 18 41. Guyatt GH, Oxman AD, Kunz R, et al. GRADE guidelines: 8. Rating the quality of evidence-- 19 indirectness. Journal of clinical epidemiology 2011;64(12):1303-10. 20 42. Guyatt GH, Oxman AD, Kunz R, et al. GRADE guidelines 6. Rating the quality of evidence-- 21 imprecision. Journal of clinical epidemiology 2011;64(12):1283-93. 22 43. Guyatt GH, Oxman AD, Montori V, et al. GRADE guidelines: 5. Rating the quality of 23 evidence--publication bias. Journal of clinical epidemiology 2011;64(12):1277-82. 24 44. RG O. Evaluating coding decisions. In: Cooper H, Hedges LV (editors). The Handbook of 25 26 Research Synthesis. New York (NY): Russell Sage Foundation, 1994. 27 45. Montori V, Ioannidis J, Cook DJ, et al. Advanced topics in Systematic Reviews. Fixed-effects 28 and random-Effects models. In: Guyatt G, Rennie D, Meade M, Cook D Users' Guides to 29 the Medical Literature: A Manual for Evidence-Based Clinical Practice McGraw-Hill 30 2008. 31 46. Johnston BC, Thorlund K, Schunemann HJ, et al. Improving the interpretation of quality of 32 life evidence in meta-analyses: the application of minimal important difference units.
33 Health and quality of life outcomes 2010;8:116. http://bmjopen.bmj.com/ 34 47. Suissa S. Binary methods for continuous outcomes: a parametric alternative. Journal of 35 clinical epidemiology 1991;44(3):241-8. 36 48. Thorlund K, Walter S, Johnston B, et al. Pooling health-related quality of life outcomes in 37 meta-analysis—a tutorial and review of methods for enhancing interpretability. Res 38 Syn Meth 2011;2:188-203. 39 40 49. Akl EA, Johnston BC, Alonso-Coello P, et al. Addressing dichotomous data for participants excluded from trial analysis: a guide for systematic reviewers. PloS one
41 on September 27, 2021 by guest. Protected copyright. 42 2013;8(2):e57132. 43 50. Ebrahim S, Akl EA, Mustafa RA, et al. Addressing continuous data for participants excluded 44 from trial analysis: a guide for systematic reviewers. Journal of clinical epidemiology 45 2013;66(9):1014-21 e1. 46 51. Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Statistics in 47 medicine 2002;21(11):1539-58. 48 52. Higgins JP, Thompson SG, Deeks JJ, et al. Measuring inconsistency in meta-analyses. BMJ 49 (Clinical research ed) 2003;327(7414):557-60. 50 53. Guyatt GH, Thorlund K, Oxman AD, et al. GRADE guidelines: 13. Preparing summary of 51 findings tables and evidence profiles-continuous outcomes. Journal of clinical 52 53 epidemiology 2013;66(2):173-83. 54 54. Guyatt GH, Oxman AD, Santesso N, et al. GRADE guidelines: 12. Preparing summary of 55 findings tables-binary outcomes. Journal of clinical epidemiology 2013;66(2):158-72. 56 55. Guyatt GH, Oxman AD, Kunz R, et al. Going from evidence to recommendations. BMJ 57 (Clinical research ed) 2008;336(7652):1049-51. 58 59 60 16 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 49 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 56. Guyatt GH, Oxman AD, Kunz R, et al. What is "quality of evidence" and why is it important 4 to clinicians? BMJ (Clinical research ed) 2008;336(7651):995-8. 5 57. Guyatt GH, Oxman AD, Vist GE, et al. GRADE: an emerging consensus on rating quality of 6 evidence and strength of recommendations. BMJ (Clinical research ed) 7 2008;336(7650):924-6. 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40
41 on September 27, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 17 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 49 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 Brazilian medicinal plants to treat upper respiratory 7 8 diseasetract and bronchial illness: systematic review and meta- 9 10 analyses- Study Protocol 11 12 Luciane C Lopes 13 Pharmaceutical Sciences Post graduate Course, University of Sorocaba, UNISO , SP, Brazil. [email protected] 14 15 Maria Carolina O. SilvaFor peer review only 16 Pharmaceutical Sciences Post graduate Course, University of Sorocaba, UNISO, SP, Brazil 17 [email protected] Formatted: Font color: Auto 18 Formatted: Font color: Auto 19 Cristiane Bergamashi Motta 20 Pharmaceutical Sciences Post graduate Course, University of Sorocaba, UNISO, SP, Brazil 21 [email protected] Formatted: Font color: Auto 22 Formatted: Font color: Auto 23 Antonio Macho Quirós 24 Biología Celular, Fisiología e Inmunología. Facultad de Medicina, Universidad de Córdoba, Spain 25 [email protected] Formatted: Font color: Auto 26 Formatted: Font color: Auto 27 Maique Weber Biavatti 28 Pharmaceutical Department, Federal University of Florianopolis, SC, Brazil 29 [email protected] Formatted: Font color: Auto 30 Formatted: Font color: Auto 31 Jardel Corrêa de Oliveira Formatted: Font color: Auto 32 Médico de Família e Comunidade, Especialista em Saúde da Família, Geriatria e Gerontologia,
33 Secretaria Municipal de Saúde, Florianópolis, SC, Brasil http://bmjopen.bmj.com/ 34 [email protected] Formatted: Font color: Auto 35 Formatted: Font color: Auto Gordon Guyatt 36 Formatted: Font color: Auto 37 Department of Clinical Epidemiology and Biostatistics, McMaster University, Canada. [email protected] 38
39 40 Contact information for the corresponding author: Luciane Cruz Lopes,
41 [email protected] Formatted: Font color: Auto on September 27, 2021 by guest. Protected copyright. 42 Rua Gomes Carneiro, 570 apto 141, Postal code: 13400-530 Formatted: Font color: Auto 43 Piracicaba, SP, Brazil. Formatted: Font color: Auto 44 Telephone: 55 19 978-18441 45 Fax: 55 15 21017074 46 Formatted: Font: 10.5 pt, Bold, Not Italic 47 Financial support: This project is funded from a grant by CAPES Research Programme Formatted: Default, Justified (grant) and is Supportgovernmental Program Graduate Education Institutions Private - 48 Formatted: Font: (Default) Arial 49 PROSUP - CAPES / UNISO No conflict of interest Formatted: Font: 12 pt 50 Word count: 33812921 (without references) Figures: 0; tables: 0; references: Formatted: Font: 12 pt 51 5257 Formatted: Font: 12 pt 52 Additional file: 1 Formatted: Font: 12 pt 53 54 Formatted: Font color: Auto 55 56 1 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 19 of 49 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40
41 on September 27, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 2 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 20 of 49 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 Keywords: Common cold, upper respiratory disease, Sputum, bronchial illness, Cough, 7 Systematic review, Meta-analysis, Brazilian medicinal plants 8 9 10 11 ABSTRACT Formatted: Line spacing: single, Tab 12 stops: Not at 0.5" 13 BackgroundIntroduction: Respiratory illness, often associated with cough and 14 sputum, is frequent. In Brazil, herbal medicines are often recommended as a first-line 15 For peer review only 16 treatment for respiratory illness. There exists uncertainty regarding the effectiveness of 17 these treatments. No systematic review has evaluated Brazilian medicinal plants to treat 18 19 the common cold, lower and upper respiratory tract infection and associated 20 cough.bronchial illness (URTI). 21 22 Methods/design and analysis: We will conduct a systematic review and if Formatted: Normal 23 Formatted: Font: Times New Roman, 12 24 appropriate a series of meta-analyses evaluating the safety and effectiveness of pt 25 Brazilian medicinal plants (BMP) for the common cold and upper respiratory tract Formatted: Font: Times New Roman, 12 26 pt 27 infection and associated cough. We will acquire eligibleURTI. Eligible randomized Formatted: Font: Times New Roman, 12 28 pt controlled trials and observational studies inwill enroll adult or pediatric patients 29 Formatted: Font: Times New Roman, 12 30 presenting with suchupper respiratory tract and bronchial illness treated by any form of pt 31 Formatted: Font: Times New Roman, 12 BMP approved in Brazilian herbalHealth Surveillance Agency compared with placebo, 32 pt Formatted: Font: Times New Roman, 12 33 no treatment, or an alternative therapy through a systematic. Our search of CINAHL, http://bmjopen.bmj.com/ 34 pt EMBASE, MEDLINE, AMED, Web of Science, Ovid Health star, Cochrane Library, Pubmed and Formatted: Font: Times New Roman, 12 35 pt 36 Scielo and the Cochrane will include Cochrane Central RegistryRegister of Controlled Formatted: Font: Times New Roman, 12 37 Trials (CENTRAL) which contains the Cochrane Acute Respiratory Illness Group’s pt 38 Formatted: Font: Times New Roman, 12 39 Specialized Register; MEDLINE, EMBASE; CINAHL; Web of Science; AMED; pt 40 LILACS, CAB abstracts, clinical trial.gov, WHO Trial Register and Brazilian thesis Formatted: Font: Times New Roman, 12 pt
41 on September 27, 2021 by guest. Protected copyright. 42 database (CAPES) without any language restrictions. Outcomes of interest are time to Formatted: Font: Times New Roman, 12 43 resolution of clinical symptoms and/or signs (cough, sputum production or activity pt 44 45 limitations), severity of symptoms prior to resolution and major/minor adverse events. Formatted: Font: Times New Roman, 12 pt 46 Teams of reviewers will, independently and in duplicate, screen titles and abstracts and 47 Formatted: Font: Times New Roman, 12 48 complete full text reviews to determine eligibility, and subsequently. For eligible studies, pt 49 reviewers will perform data abstraction and assess risk of bias of eligible trials. When Formatted: Font: Times New Roman, 12 50 pt 51 appropriate, we will conduct meta-analyses to establish. We also assess the quality of Formatted: Font: Times New Roman, 12 pt 52 body of evidence (confidence in estimates of effect of all reported therapies on patient- Formatted: Font: Times New Roman, 12 53 important) for each of the outcomes. using the GRADE approach. pt 54 Formatted: Font: Times New Roman, 12 55 pt 56 3 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 21 of 49 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 Discussion: Our review will be the first to evaluate all Brazilian medicinal plants to 7 8 treat the common cold, upper respiratory tract infection and associated cough and 9 establish best estimates of the safety and effectiveness of treatments. Our review will 10 11 facilitate evidence-based management of patients in primary care and identify key areas 12 for future research. 13 14 Ethics and Dissemination: The systematic review will be published in a peer- Formatted: Space After: 10 pt, Adjust 15 For peer review only space between Latin and Asian text, Adjust 16 reviewed journal. Brief reports of review findings will be disseminated directly to space between Asian text and numbers 17 appropriate audiences via email and other modes of communication. The review will 18 19 guide healthcarehealth care practice and policy in Brazil. 20 21 Register Protocol: Prospero CRD42014007057 Formatted: Line spacing: 1.5 lines 22 Formatted: Font: Bold 23 24 25 26 27 28 29 30 31 32
33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40
41 on September 27, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 4 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 22 of 49 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 Keywords: Common cold, upper respiratory disease, Cough, Systematic review, Meta- 7 8 analysis, Brazilian medicinal plants 9 10 11 ARTICLE FOCUS: 12 Will a systematic review of Brazilian medicinal plants to treat the common cold, lower 13 and upper respiratory tract infection and associated cough reveal effectiveness of these 14 15 agents? For peer review only 16 Do Brazilian medicinal plants lead to symptom improvement and cough control in 17 18 common cold, lower and upper respiratory tract infection in patients? 19 20 21 KEY MESSAGE: 22 23 This study will investigate the effects of all marketed Brazilian medicinal plants 24 indicated to use in respiratory disease: Ananas comosus, Echinacea purpurea Moench, 25 Eucalyptus globules, Glycyrrhiza glabra L., Hedera helix L., Malva sylvestris L., Mentha spp* Formatted: Emphasis, Font: Calibri, 11 pt, 26 Not Italic, English (U.S.) 27 (M. piperita or M. villosa), Mikania spp* (M. glomerata or M. laevigata), Pelargonium 28 sidoides D.C., Petasites hybridus L., Pimpinella anisum L., Polygala senega L., Psychotria Formatted: Emphasis, Font: Calibri, 11 pt, 29 Not Italic, English (U.S.) 30 ipecacuanha (Brot.) Stokes, Sambucus nigra L. Formatted: Emphasis, Font: Calibri, 11 pt, 31 Not Italic, English (U.S.) 32 Outcomes of interest will include time to resolution of clinical symptoms and/or signs
33 http://bmjopen.bmj.com/ 34 (where clinical symptoms and signs include cough, sputum production or activity 35 limitations). Also of interest will be severity of symptoms prior to resolution. Other 36 37 important outcomes will include hospitalization rates and duration of hospital stay, and 38 days receiving antibiotics 39 40
41 on September 27, 2021 by guest. Protected copyright. STRENGTHS AND LIMITATIONS OF THIS STUDY Formatted: Font color: Auto 42 Formatted: Font: 43 Formatted: Font: 44 • This will be the first systematic review to assess specifically the Brazilian Formatted: Font: 45 medicinal plants approved in Brazilian Health Surveillance Agency (ANVISA) 46 to treat the common cold, lower and upper respiratory tract infection and Formatted: Font: 47 bronchial illness associated with cough. and sputum. Formatted: Font: 48 • The results of this systematic review will help clinicians in making decisions in Formatted: Font: 49 clinical practice and help patients with cough and sputum seeking effective and Formatted: Font: 50 safe treatment options. Formatted: Font: 51 Formatted: English (U.S.) 52 • The methods of the review are state-of-art, including explicit eligibility criteria, Formatted: List Paragraph, Bulleted + 53 a comprehensive search, independent duplicate assessment of eligibility, and use Level: 1 + Aligned at: 0.25" + Indent at: 54 of the GRADE approach to assessingassess confidence in estimates of effect 0.5" 55 Formatted: English (U.S.) 56 5 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 23 of 49 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 including independent duplicate assessment of risk of bias, precision, Formatted: English (U.S.) 7 consistency, directness and publication bias. We will make separate ratings for 8 bodies of evidence from randomized trials and bodies of evidence from 9 observational studies. 10 11 12 13 Our results Because primary studies are likely to be limited by limitations in the primary 14 studies includeto non-randomized studies and randomized trials with a high risk of bias. 15 confidence in estimatesFor is likely peer to be low. review Eligible studies will likelyonly differ 16 17 substantially in study design and outcome measures. 18
19 20 BACKGROUND: 21 22 Respiratory illness is common and important 23 24 25 Respiratory illness, including the common cold and upper respiratory tract infection 26 (URTI), and The exact constituents of the plants being tested are likely to be associated 27 28 cough, are frequent[1] and a major cause of morbidity, especially in children and the 29 elderly [2]. Although in most cases benign, respiratory illness results in many 30 31 consultations to primary care [3 4]. 32 Formatted: List Paragraph, Line spacing: 33 • Acute URTIs are the most common reason for people to seek medical care in the http://bmjopen.bmj.com/ single, Bulleted + Level: 1 + Aligned at: 34 United States [5] and at least one billion colds occur there per year, with a 0.25" + Indent at: 0.5", Adjust space 35 frequency of two to six colds per person [6]. Symptoms of the common cold between Latin and Asian text, Adjust space between Asian text and numbers 36 typically include a runny nose, congestion, sneezing, weakened sense of taste Formatted: English (U.S.) 37 and smell, scratchy throat and cough. These start developing in the first three 38 days following infection. Infants and young children are more likely than adults 39 40 and teens to also develop fever. Symptoms usually abate within seven to 10 days but some colds last longer, especially in children, the elderly and those with Formatted: English (U.S.)
41 on September 27, 2021 by guest. Protected copyright. 42 generally poor health [7]. The most common symptom of respiratory tract 43 infection, cough, is also the most common symptom presenting to general 44 practitioners [5 8]. uncertainty. 45 46 47 48 49 50 51 52 53 54 55 56 6 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 24 of 49 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 INTRODUCTION 7 8 Use of herbal medicines is frequent, particularly in Brazil 9 Formatted: Font: Calibri, 11 pt, Bold, Italic, 10 In high-income countries there is increasing public interest in, and use of, a wide range Font color: Auto 11 Formatted: Space After: 0 pt, Line 12 of therapies that lie outside the main stream of traditional Western medical spacing: Multiple 1.15 li, Adjust space between Latin and Asian text, Adjust space 13 practiceCough may arise from at least three mechanisms. One is virus-induced between Asian text and numbers, Tab 14 stops: 0.5", Left 15 postnasal drip. Alternatively,For it haspeer been proposed review that a viral upper respiratory only tract 16 infection produces inflammatory mediators that result in an increase in the sensitivity of 17 18 the afferent sensory nerves in the upper airway[9]. Third, the infection may spread from 19 the upper to the lower respiratory tract with a resulting bronchitis. This third 20 21 mechanism is often associated with sputum production. 22 Cough can be characterized based on time frame (ie, duration of cough), quality (eg, dry 23 24 or wet, brassy, or staccato), or suggested etiology (ie, specific and nonspecific)[10]. 25 Cough can be designated as acute (<3 weeks in duration), prolonged acute cough (3 to 8 26 27 weeks in duration) or chronic (> 8 weeks in duration) [11 12]. 28 In adults, although there is no prospective study of the causes of acute cough, it has long 29 been considered that the common cold is the single most common cause of acute cough 30 31 (ie, cough < 3 weeks in duration)[9]. 32
33 In most children acute coughing is usually due to a viral upper respiratory tract infection http://bmjopen.bmj.com/ 34 35 such as a simple head cold with bronchitis or croup. Less often, but still common, 36 pathogens can involve the lower respiratory tract system causing bronchiolitis, 37 38 whooping cough, or pneumonia. Symptomatic URTI with cough in school children 39 typically occurs around 7–10 times per year[13]. 40
41 Non-herbal medication use to treat respiratory infection is common on September 27, 2021 by guest. Protected copyright. 42 43 Worldwide, the desire to reduce the symptom of cough is reflected in the billions of 44 45 dollars spent on over-the-counter (OTC) products, mostly cough and cold medications 46 (CCMs) [1]. The preparations are usually a combination of several medications 47 48 including antitussives, expectorants, antibiotic, antihistamines, decongestants and 49 antipyretics [14]. 50 51 Current systematic reviews addressing the use of CCMs show insufficient evidence to 52 53 decide whether OTC medications for cough beneficial[14]. It has been suggested that 54 zinc can inhibit viral growth[15]. As such, the treatment of cough and cold with zinc 55 56 7 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 25 of 49 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 was tested in several studies[16]. While some of them showed benefits, especially if 7 8 used within 24 h of the onset of common cold symptoms [17][18], others failed to show 9 the same effect[18]. At the present time, the use of zinc in children with cough and cold 10 is not recommended[1]. The available evidence does not support the use of high doses 11 12 of vitamin C for treating the common cold [19]. Likewise, there is insufficient evidence 13 to support the treatment of upper respiratory tract infections with antibiotics, and an 14 15 increased adverse effectsFor associated peer with antibiotic usereview in adult patients [20]. only 16 17 Antihistamines in monotherapy - in children as well as in adults - do not alleviate nasal 18 19 congestion, rhinorrhea and sneezing, or produce subjective improvement of the common 20 cold[21]. First generation antihistamines also cause more side-effects than placebo, in 21 22 particular they increase sedation in cold sufferers. Combinations of antihistamines with 23 decongestives are not effective in small children. In older children and adults most trials 24 show a beneficial effect on general recovery as well as on nasal symptoms. However, 25 26 the extent to which improvement is important to patients remains unclear [22]. 27 28 There are no effective licensed antiviral drugs for the common cold. Of 29 30 the mucolytic drugs available to treat acute upper and lower URTI, the cysteine 31 derivatives (that is, acetylcysteine and carbocysteine) are the most commonly prescribed 32 in many European [23 24] and African countries[25] and in Brazil 1 2. Complementary 33 http://bmjopen.bmj.com/ 34 and alternative medicine (CAM) has grown rapidly over the last two decades1. In the 35 United States, approximately 38 percent of adults and approximately 12 percent of 36 37 children are using some form of CAM3. In Brazil, of total revenues of the 38 pharmaceutical industry from sales of drugs in the period from 1996 to the present, up 39 40 to 25% came from preparations derived from plants4. The government’s decision to
41 on September 27, 2021 by guest. Protected copyright. include herbal medicine in the list of publicly subsidized medicine in the Brazilian 42 43 Health System (SUS) may have contributed to an increase in expenditures on herbal 44 medicine in Brazil of 12% in 2012 over 2011, with total of $ 1.147 billion dollars5. 45 46 47 The license approval process for herbal medicines vary across countries, 48 including wide variation in evidence of effectiveness required for licensing .6 7. Some 49 50 countries, including Brazil, demand for licensing only evidence of long standing and 51 widespread use of a plant. In such countries, the extent of pharmacovigilance of 52 8 53 licensed products differs; relatively rigorous pharmacovigilance exists in Australia , 54 Canada , Germany, among others, but not in Brazil 7 9. 55 56 8 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 26 of 49 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 In many countries, traditional herbal medicines are available over-the- 7 3 8 counter (ie no need for a prescription for their purchase or use ). These medications are 9 typically not recommended for serious medical conditions, but rather as adjunctive 10 treatments and for short-term use in conditions that are not serious10 11. Aside from 11 12 Brazil, there is no country that provides public support for payment for herbal medicines 13 approved only on the basis of long standing and widespread prior use. Nowadays Brazil 14 15 has a list of 12 suchFor herbal medicines peer funded by the review government.12 13. only 16 17 Primary care physicians often recommend herbal medicines to their 18 14 19 patients as first line of treatment This is particularly the case in Brazil, perhaps 20 encouraged by government funding for these drugs. Furthermore, people frequently 21 22 self-prescribe over-the-counter cough medications. One reason for concern about this 23 widespread use is that patients are less likely to consult their general practitioner 24 15 25 because of an adverse reaction to an herbal remedy than for a conventional medicine . 26 27 Respiratory Illness and herbal medicine 28 29 Respiratory illness, in particular upper respiratory tract and bronchial illness (URTI), 30 16 31 often with associated cough and sputum, are frequent and a major cause of morbidity, 32 especially in children and the elderly 17. Although in most cases benign18, respiratory
33 http://bmjopen.bmj.com/ 19 34 illness is a cause for concern for parents and a major cause of outpatient visits in most 35 settings [8 26]20 21 and seem to have a limited efficacy and also appear to be safe in 36 37 children older than two years. 38 39 40 Herbal medicines represent an alternative for treatment of respiratory illness
41 on September 27, 2021 by guest. Protected copyright. 42 The synonyms of herbal medicines include herbal remedies, herbal 43 44 medications, herbal products, herbal preparations, medicinal herbs and 45 phytopharmaceuticals. Patients and physicians may be unaware that products with the 46 47 same label differ appreciably in their composition, mainly due to the use of variable 48 plant material, extraction methods and the addition of other components. 22 23. URTI 49 50 can adversely impact on quality of life24. Patients spend billions of dollars annually on 51 OTC medications for URTI, and in particular for the frequently accompanying cough 52 25 53 symptoms . 54 55 56 9 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 27 of 49 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 Numerous OTC cough preparations are available but a Cochrane review 7 8 that did not address the plants that are the topic of the current review suggests there is 9 no conclusive evidence regarding their efficacy26 27. In children, OTC medications may 10 be associated with serious adverse events such as death, altered consciousness and 11 12 arrhythmias28-32. 13 14 A search in the database of the Brazilian Health Surveillance Agency 15 For peer review only 16 (ANVISA) revealed that 15 species of herbal medicines are approved for treatment of 17 acute cough from an URTI. Of these, Public Health System (SUS) funding is available 18 19 for two. There are no systematic reviews available addressing the benefits and harms of 20 the herbal medication approved by ANVISA for URTI. Identification of ineffective 21 22 preparations could reduce costs for consumers and healthcare providers, and reduce the 23 risk of adverse events from treatments with no benefit27. This current systematic review 24 25 therefore aims to collect the evidence to evaluate the effectiveness and safety of 15 26 Brazilian herbal medicines currently approved to management cough from an upper 27 28 respiratory tract and bronchial illness. 29 30 31 32
33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40
41 on September 27, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 10 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 28 of 49 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 Formatted: Font: Calibri, 11 pt, Bold, Italic, 7 In high-income countries there is increasing public interest in, and use of, a wide range Font color: Auto 8 Formatted: Space After: 0 pt, Line 9 of therapies that lie outside the main stream of traditional Western medical practice[27 spacing: Multiple 1.15 li, Adjust space between Latin and Asian text, Adjust space 10 28]. Complementary and alternative medicine (CAM) has grown rapidly over the last between Asian text and numbers, Tab 11 stops: 0.5", Left 12 two decades[27]. The Cochrane Collaboration defines complementary and alternative 13 medicine as a broad domain of healing resources that encompasses all health systems, 14 15 modalities, and practicesFor and their peer accompanying theoriesreview and beliefs, other only than those 16 intrinsic to the politically dominant health systems of a particular society or culture in a 17 given historical period. In the United States, approximately 38 percent of adults (about 4 18 19 in 10) and approximately 12 percent of children (about 1 in 9) are using some form of 20 CAM. 21 22 23 In Brazil, up to 25% of the $ 8 billion of revenues of the pharmaceutical industry, in 24 1996, may come from sales of drugs derived from plants [29]. United States and 25 26 Germany are among the largest consumers of Brazilian natural products. Between 1994 27 and 1998, imported, respectively, 1,521 and 1,466 tons plants that follow for these 28 29 countries under the generic label "Plant material of Brazil," according to Brazilian 30 Institute of Environment and Renewable Natural Resources[30]. 31 32 Herbal medicines have been widely used in cough[31]. Antitussives act either centrally
33 http://bmjopen.bmj.com/ 34 on the cough center of the brain or peripherally on the cough receptors in the respiratory 35 passages. The putative antitussive effect of many herbs may result from the content of 36 37 mucilage, which exerts protective and demulcent activity[31]. 38 39 Expectorant herbs containing saponins may reduce the surface tension of the secretions, 40 facilitating their separation from the mucous membranes. This induces reflex
41 on September 27, 2021 by guest. Protected copyright. 42 stimulation which leads to an increase in the secretion of bronchial glands. Volatile-oil 43 type expectorant herbs exert a direct stimulatory effect on the bronchial glands by 44 45 means of local irritation. They may also have antibacterial activity. In colds and 46 influenza, herbs containing volatile oil can be used; also, volatile oils are ingredients of 47 48 syrups and liquids as well as external phytomedicines in the form of liniments, 49 ointments, and inhalations[32]. 50 51 Both limited research into possible mechanisms of action, and widespread use with 52 53 many testimonials to success, suggest that treatment with herbal medicines may have 54 great potential to treat respiratory diseases. The effectiveness of such treatment, 55 56 11 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 29 of 49 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 however, needs to be reviewed systematically and appraised critically to inform current 7 8 practice and direct future research. 9 10 In 2008, the Ministry of Health of Brazil has built a list (Renisus), with 71 species, pre- 11 selected by regions that alluded to its use for indications of use and according to the 12 13 categories of the International Classification of Diseases (ICD-10 ), with the potential to 14 advance the productive chain and generate products of interest to the public health 15 For peer review only 16 system in Brazil[33]. Moreover, in this same year, 36 species were recorded as herbal 17 medicines simplified registration (IN05) ie not need to prove through clinical trials the 18 19 effectiveness or safety record of its clinical indications only considered data from 20 tradition of popular use[34]. 21 22 A search in the database of the Brazilian Health Surveillance Agency, for herbal 23 24 medicines registered for commercialization indicated for the treatment of respiratory 25 diseases resulted in 15 species currently registered for this indication. Of these, six 26 27 belong to Renisus and nine to IN05. 28 29 Therefore, this study will investigate the effects of all marketed Brazilian medicinal 30 plants indicated to use in respiratory disease: Ananas comosus, Echinacea purpurea 31 32 Moench, Eucalyptus globules, Glycyrrhiza glabra L., Hedera helix L., Malva sylvestris
33 http://bmjopen.bmj.com/ L., Mentha spp* (M. piperita or M. villosa), Mikania spp* (M. glomerata or M. 34 35 laevigata), Pelargonium sidoides D.C., Petasites hybridus L., Pimpinella anisum L., 36 Polygala senega L., Psychotria ipecacuanha (Brot.) Stokes, Sambucus nigra L. 37 38 Formatted: Font: +Body (Calibri), 11 pt, 39 Not Italic, Not Highlight 40 Formatted: Justified, Indent: First line: 1", OBJECTIVES Space After: 10 pt, Line spacing: 1.5 lines
41 on September 27, 2021 by guest. Protected copyright. 42 Formatted: Font: 14 pt, Bold 43 The primary objective is to address the safety and efficacy of 15 Brazilian medicinal Formatted: Normal 44 45 plants to treat the common cold and herbal medicines approved by ANVISA for acute Formatted: Font: Italic 46 cough from upper respiratory tract infection and associated coughbronchial illness. 47 48 METHODS/DESIGN AND ANALYSES Formatted: Font: 14 pt, Bold 49 Formatted: Font: Italic 50 The systematic review will be performed according to the recommendations specified in 51 Formatted: Normal 33 52 the Cochrane Handbook for Intervention Reviews . The reporting of the review will 53 follow the Preferred Reporting Items for Systematic Reviews and Meta-Analyses 54 34 55 (PRISMA) statement . 56 12 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 30 of 49 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 Protocol and registration Formatted: Font: 12 pt 7 Formatted: Space Before: 0 pt, Line 8 Our protocol is registered on PROSPERO (CRD42014007057), Availableand is available spacing: 1.5 lines 9 from http://www.crd.york.ac.uk/PROSPERO Formatted: Font: (Default) Times New 10 Roman, 12 pt, Not Bold 11 Formatted: Font: (Default) Times New 12 Our review will conform to the PRISMA guidelines for reporting systematic reviews. Roman, 12 pt, Not Bold 13 Formatted: Font: (Default) Times New 14 Eligibility criteria for considering studies for review Roman, 12 pt, Not Bold 15 Formatted: Default Paragraph Font, Font: Inclusion criteria For peer review only (Default) Times New Roman, 12 pt, Not 16 Bold 17 Formatted: Font: (Default) Times New 18 Our eligibility criteria are as follows: Roman, 12 pt, Not Bold 19 Patients: Studies must include patients with adult (>18 years old) or pediatric (0-18 Formatted: Font: Times New Roman, 12 20 pt years old) patients with upper respiratory disease: the common cold, sinusitis, tonsillitis, 21 Formatted: Font: Times New Roman, 12 22 otitis media, pharyngitis or laryngitis; or symptoms arising from the upper part of the pt, Not Italic 23 Formatted: Normal, Justified, Line lower respiratory tract (either secondary to upper respiratory tract symptoms - e.g. post- spacing: 1.5 lines, Don't adjust space 24 between Latin and Asian text, Don't adjust 25 nasal drip - or to acute bronchitis, or bronchiolitis). space between Asian text and numbers, 26 Tab stops: Not at 0.5" 27 Interventions: studyStudies must include an arm in which patient are taking one of Formatted: Underline 28 Formatted: Space After: 10 pt Brazilian herbal medicine from any of the following plant preparation (whole, powder, 29 Formatted: Underline, English (U.S.) 30 extract, standardized mixture) with one of select plants (Ananas comosus, Echinacea Formatted: Normal (Web), Left, Adjust 31 space between Latin and Asian text, Adjust 32 purpurea Moench, Eucalyptus globules, Glycyrrhiza glabra L., Hedera helix L., Malva space between Asian text and numbers
33 sylvestris L., Mentha spp* (M. piperita or M. villosa), Mikania spp* (M. glomerata or http://bmjopen.bmj.com/ Formatted: English (U.S.) 34 M. laevigata), Pelargonium sidoides D.C., Petasites hybridus L., Pimpinella anisum L., Formatted: English (U.S.) 35 36 Polygala senega L., Psychotria ipecacuanha (Brot.) Stokes, Sambucus nigra L.) : 37 Ananas comosus (L.) Merr., Bromeliaceae; 38 39 Echinacea purpurea (L.) Moench, Asteraceae; 40 Eucalyptus globulus Labill., Myrtaceae;
41 on September 27, 2021 by guest. Protected copyright. Glycyrrhiza glabra L., Fabaceae; Formatted: Emphasis, Font: Times New 42 Roman, 12 pt, Not Italic, English (U.S.) 43 Araliaceae; Hedera helix L., Formatted: Emphasis, Font: Times New 44 Malva sylvestris L., Malvaceae; Roman, 12 pt, Not Italic, English (U.S.) 45 Formatted: Emphasis, Font: Times New 46 Mentha spp (Mentha x piperita L., Mentha x villosa Huds., or other hybrids), Lamiaceae; Roman, 12 pt, Not Italic, English (U.S.) 47 Mikania glomerata Spreng.or Mikania laevigata Sch.Bip. ex Baker, Asteraceae; 48 49 Pelargonium sidoides DC., Geraniaceae; Formatted: Emphasis, Font: Times New 50 Petasites hybridus (L.) G. Gaertn., B. Mey. & Scherb., Asteraceae; Roman, 12 pt, Not Italic, English (U.S.) 51 Formatted: Font: (Default) Times New Pimpinella anisum L., Apiaceae; Roman, Not Italic, English (U.S.) 52 Polygala senega L., Polygalaceae; Formatted: Emphasis, Font: Times New 53 Roman, 12 pt, Not Italic, English (U.S.) 54 Psychotria ipecacuanha (Brot.) Formatted: Emphasis, Font: Calibri, 11 pt, 55 Not Italic, English (U.S.) 56 13 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 31 of 49 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 Type of study and design: Stokes or Cephaelis ipecacuanha (Brot.) A. Rich., Rubiaceae; 7 8 Sambucus nigra L., Adoxaceae. 9 Formatted: Font: 12 pt, Italic 10 Formatted: Heading 3, Left, Add space 11 Outcome Measures between paragraphs of the same style, Line spacing: 1.5 lines, Tab stops: Not at 12 We will include studies that report any of the following outcomes: 0.5" 13 Formatted: Font: 12 pt, Not Italic, 14 - time to resolution of clinical symptoms and/or signs (e.g. cough, sputum Underline 15 production).For peer review only Formatted: Heading 3, Left, Line spacing: 16 1.5 lines, Tab stops: Not at 0" 17 - frequency and severity of symptoms prior to resolution. Formatted: Font: 12 pt, Not Bold, Not 18 - minor and serious adverse effects of the intervention and the proportion of Italic, Underline 19 Formatted 20 patients requiring discontinuation of the herbal medicine 21 - hospitalization rates 22 23 - duration of hospital stay 24 - days receiving antibiotics 25 26 - functional status (including number of days of disability that may be defined as 27 days in bed, days off work or days when patients were unable to undertake 28 normal activities) 29 30 - Quality-of-life measured by a validated instrument. The score will be evaluated 31 using Cough-Specific Quality-of-Life Questionnaire 24 , Leicester Cough 32 35 33 Questionnaire or other validated questionnaires. http://bmjopen.bmj.com/ 34 35 Study Design 36 37 We will include (1) any comparison (randomized controlled trials or observational Formatted: Space After: 10 pt 38 39 study) including in an arm in which patients are taking one of the herbal 40 medicinemedicines listed above via any route of administration compared to an arm in
41 on September 27, 2021 by guest. Protected copyright. 42 which patients receive with an inert (placebo) or no treatment or an active non-herbal 43 medicine control and open label, via any route of administration, or case-control studies 44 45 comparing the frequency of use of herbal medicine in patients with better versus poorer 46 outcomes of their respiratory illness. 47 48 49 50 51 Exclusion criteria 52 53 Trials that includedWe will exclude studies in which more than 20% of participating 54 patients with anysuffered from one or more of the following conditions will excludein 55 56 14 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 32 of 49 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 which results from the eligible population were not separately reported: chronic 7 8 obstructive pulmonary disease (COPD), pneumonia, bronchiectasis, cystic fibrosis, 9 bronco-pulmonary dysplasia, asthma or tuberculosis; underlying immunodeficiency or 10 respiratory tract anatomical defect; acute respiratory distress requiring mechanical 11 12 ventilation, and in which results from the clearly eligible population were not separately 13 reported.. Also if the population, we will exclude studies that investigate the 14 15 simultaneous use ofFor study uses two peer ofmore than one review the eligible plants we will only exclude. 16 17 18 Search methods for primary studies: 19 We will search the CENTRAL MEDLINE, EMBASE, CINAHL, AMED, Web of 20 21 Science, Ovid Healthstar, Pubmed, Scielo and The Cochrane Central Register of 22 Controlled Trials (CENTRAL), which includes the Cochrane Airways Group 23 24 Specialized Trials Register. 25 We will restrict the search to human subjects but we will not restrict the searches or 26 27 inclusion criteria to any specific languages. 28 Electronic searches 29 30 We will search the following electronic databases irrespective of language or 31 publication status: Cochrane Central Register of Controlled Trials (CENTRAL) which 32
33 contains the Cochrane Acute Respiratory Infections Group’s Specialized Register ; http://bmjopen.bmj.com/ 34 MEDLINE; EMBASE; CINAHL (Cumulative Index to Nursing and Allied Health 35 36 Literature); Web of Science; Health Star (via OVID); AMED, the database of the 37 Cochrane Complementary Medicine Field, LILACS; CAB abstracts, clinical trial.gov, 38 36 39 WHO Trial Register and Brazilian thesis database (CAPES). 40 Searching other resources
41 on September 27, 2021 by guest. Protected copyright. 42 For every eligible study we identify and for studies such as other review articles that we 43 44 identify that may have citations including eligible studies, one reviewer will examine 45 the reference list. We will obtain and evaluate the full text of any potentially eligible 46 47 studies thus identified and determine their eligibility as described below. 48 We will write to the principal authors of the identified trials and the pharmaceutical 49 50 companies involved in the production of medicinal herbs and inquire about additional 51 trials of which they are aware. 52
53 54 55 56 15 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 33 of 49 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 Unpublished studies will be identified by searching in the books including List of 7 8 references for evaluation of safety and efficacy of herbal medicines described in the 9 Brazilian legislation for herbal medicines in Brazil and conference proceedings 10 (Medicinal Symposium of Brazilian medicinal plants; International Congress of 11 12 Ethnopharmacology). 13 14 The following Brazilian scientific journals will also be scanned manually for eligible 15 For peer review only 16 studies: Journal of Basic and Applied Pharmaceutical Sciences; Brazilian Journal of 17 Pharmacy; Brazilian Journal of Pharmacognosy; Brazilian Journal of Medicinal Plants; 18 19 Brazilian Journal of Pharmaceutical Sciences. 20 21 Search strategy 22 We will restrict the search to human subjects but we will not restrict the searches or 23 24 inclusion criteria to any specific languages. We stated the search strategy in in 25 Appendix section to search MEDLINE and CENTRAL. We will not combine the 26 27 MEDLINE search with the Cochrane Highly Sensitive Search Strategy for identifying 28 randomized trials in MEDLINE37 because we think will find only few results. We will 29 30 adapt the search string to search EMBASE, CINAHL, LILACS and Web of Science. 31 The search will be conducted using individually for each plant. The following terms will 32
33 be used to: 1) intervention: medicine, herbal; plants, plant; extracts; medicinal; http://bmjopen.bmj.com/ 34 medicine, traditional; herb$; phytomedicine; phytotherapy; complementary therap*; 35 36 complementary Medicine*; alternative therap*; traditional medicine*; ethnomedicine*; 37 ethnobotany; ethnopharmacology; oriental traditional medicine*; scientific name of 38 39 plant, synonymies of each medicinal plants; popular name of each medicinal plant 40 selected; 2)Condition: respiratory tract diseases, respirat*, cough*, sputum; bronchial
41 on September 27, 2021 by guest. Protected copyright. 42 illness. The complete search strategy is available in Appendix 1. 43 Formatted: Font: 12 pt, Italic 44 Formatted: Heading 3, Left, Add space 45 Outcome Measures between paragraphs of the same style, 46 Line spacing: 1.5 lines, Tab stops: Not at Outcomes 0.5" 47 Outcomes of interest will include time to resolution of clinical symptoms and/or signs Formatted: Font: 12 pt, Not Italic, 48 Underline 49 (where clinical symptoms and signs include cough, sputum production or activity Formatted: Heading 3, Left, Line spacing: 50 limitations). Also of interest will be severity of symptoms prior to resolution. Other 1.5 lines, Tab stops: Not at 0" 51 Formatted: Font: 12 pt, Not Bold, Not 52 important outcomes will include hospitalization rates and duration of hospital stay, and Italic, Underline 53 days receiving antibiotics. Finally, we will summarize data addressing quality of life or 54 55 functional status (including number of days of disability that may be defined as days in 56 16 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 34 of 49 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 bed, days off work or days where patients were unable to undertake normal activities 7 8 during the illness), and adverse events. 9 10 Formatted: Font: Not Bold 11 Eligibility determination Formatted: Normal, Left, Add space between paragraphs of the same style, 12 Line spacing: single, Don't adjust space 13 We will also record minor and serious adverse effects of the intervention and the between Latin and Asian text, Don't adjust 14 space between Asian text and numbers, proportion of patients requiring discontinuation of the herbal medicine. Four reviewers Tab stops: Not at 0.5" 15 For peer review only 16 authors, working in pairs (MC/MW; AM/LL), will independently screen potentially 17 relevant citations and if available abstracts and apply the selection criteria. We will 18 19 obtain full texts of all articles that either reviewer feels might be eligible. Two 20 reviewers (MC/CB) will independently assess the eligibility of each full-text article and 21 22 resolve disagreements by consensus. 23 To exclude duplicate articles, one member (MCS)reviewer will look throughexamine all 24 25 eligible articles and identify those in whichthat have one or more authors arein common. 26 For such articles, detailed review will determine if there is duplicate publication and, if 27 28 there is, MCSwe will decide which hasuse the article with the more complete data and. 29 We will record the less complete as a duplicate and abstract data only for the more 30 31 complete. . 32
33 Data extraction http://bmjopen.bmj.com/ 34 35 The reviewers, working in pairs (MCS-MW and MCS-LL),, will independently extract 36 the data, recording information regarding patients, methods, interventions, outcomes, 37 38 missing outcome data, and results using standardized, pretested, data extraction forms 39 with accompanying instructions. For articles published in abstract form only, or for 40 articles in which important information is missing, we will seek complete information
41 on September 27, 2021 by guest. Protected copyright. 42 regarding methods and results from authors. Individually, reviewers will evaluate 2 43 44 articles and then check agreement with one another. This process will continue every 2 45 articles until reviewers are confident they can achieve very high rates of agreement. 46 Disagreements will be resolved through discussion with any unresolved issues referred 47 48 to another reviewer (GG).. 49 50 Risk of bias in individual studies 51 Formatted: Font: 12 pt 52 For randomized trials, two reviewers will independently will assess the risk of bias, 53 Formatted: Font: 12 pt 54 including sequence generation, allocation concealment, blinding, number of patients Formatted: Font: 12 pt 55 Formatted: Font: 12 pt 56 17 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 35 of 49 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 with missing outcome data, selective outcome reporting, and other sources of bias using 7 38 8 a modified version of the Cochrane collaboration risk of bias tool[35]. . We will assess Formatted: Font: 12 pt 9 the risk of bias of observational studies with a modified version of the Newcastle- 10 Ottawa instrument that includes confidence in assessment of exposure and outcome, 11 12 adjusted analysis for differences between groups in prognostic 13 stratificationcharacteristics, accuracy of outcomes assessment, and missing data.[36].39 14 15 For peer review only 16 Confidence in pooled estimates of effect Formatted: Heading 3, Space After: 0 pt, Adjust space between Latin and Asian text, 17 Adjust space between Asian text and 18 We will also independently ratedrate the overall quality of evidence (confidence in numbers 19 effect estimates) for each of the outcomes by using the Grading of Recommendations Formatted: Font: 12 pt 20 Formatted: Font: 12 pt Assessment, Development and Evaluation (GRADE) approach [37-40].40-43. We will 21 Formatted: Font: 12 pt 22 make separate ratings for bodies of evidence from randomized trials and bodies of Formatted: Font: 12 pt 23 evidence from observational studies. In the GRADE approach, randomized trials begin Formatted: Font: 12 pt 24 25 as high- quality evidence but may be rated down by 1 or more of 5 categories of Formatted: Font: 12 pt 26 limitations: risk of bias, inconsistency, indirectness, imprecision, and reporting bias. 27 28 Observational studies begin as low quality evidence but can be rated up for a large 29 effect size, evidence of a dose-response gradient observational studies, or for Formatted: Font: 12 pt 30 31 consideration of all plausible confounding. Formatted: English (U.S.) 32
33 Documentation of agreement Formatted: Font: Arial http://bmjopen.bmj.com/ 34 Formatted: Heading 3, Left 35 We will document chance-corrected agreement for i) eligibility and ii) risk of bias of Formatted: Font: Arial, Not Bold 36 individual studies and iii. all GRADE rating (precision, consistency, directness, and 37 38 publication bias).. To measures of agreement we will use Kappa statistical. Values of 39 kappa between 0.40 and 0.59 have been considered to reflect fair agreement, between 40 0.60 and 0.74 to reflect good agreement and 0.75 or more to reflect excellent agreement
41 on September 27, 2021 by guest. Protected copyright. 42 [41].8 to reflect good agreement and 0.75 or more to reflect excellent agreement44. 43 44
45 46 Data synthesis 47 Where meta-analysis is not appropriate (excessive heterogeneity of population, 48 49 intervention, comparator, outcome, or methodology), we will construct summary tables 50 and provide a narrative synthesis. When meta-analysis is appropriate, we will conduct 51 52 analyses for each herbal intervention separately. We will conduct an analysis for each 53 outcome of interest. For interventions and outcomes for which there are both 54 55 randomized trials and observational studies available we will determine the confidence 56 18 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 36 of 49 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 in estimates for each body of evidence and conduct an analysis for the body of evidence 7 8 that warrants greater confidence. If the two bodies of evidence warrant similar 9 confidence, we will conduct analyses for both bodies of evidence. 10 11 Meta-analyses will be conducted using Comprehensive Meta-Analysis (Biostat, 12 Englewood, NJ, USA). We will use random effects meta-analyses[42],We will use 13 random effects meta-analyses45, which are conservative in that they consider both 14 within and between studies differences in calculating the error term used in the analysis 15 For trials that reportFor dichotomous peeroutcomes, we will review calculate the pooled relative only risk 16 with associated 95% confidence intervals. In the cross-sectional and case-control studies 17 we will document the proportion of patients in the intervention and control groups who 18 experience each of the outcomes of interest .For case-control studies, we will use odds 19 ratios rather than relative risks. 20 When pooling across trials that report continuous outcomes using the same instrument, Formatted: Font color: Auto, English 21 (U.S.) 22 we will calculate the weighted mean difference (WMD), which maintains the original 23 unit of measurement and represents the average difference between groups, with studies Formatted: Font color: Auto, English 24 (U.S.) weighted by the inverse of their variance. Once the WMD has been calculated, we will 25 26 contextualize this value by noting, when available, the corresponding minimally 27 important difference (MID) - the smallest change in instrument score that patients 28 29 perceive is important. Formatted: Font color: Auto 30 If studies reported the same construct using different measurement instruments, we will Formatted: Font color: Auto, English 31 (U.S.) 32 calculate the standardized mean difference (SMD). The SMD expresses the intervention
33 effect in SD units, rather than the original units of measurement, with the value of a http://bmjopen.bmj.com/ 34 35 SMD depending on both the size of the effect (the difference between means) and the 36 SD of the outcomes (the inherent variability among participants). For outcome measures Formatted: Font color: Auto, English 37 (U.S.) 38 that have an established anchor-based MID, we will use this measure to convert the 39 SMD into an odds ratio (OR). We will complement this presentation by either 40 converting the SMD into natural units of a widely accepted instrument used to measure
41 on September 27, 2021 by guest. Protected copyright. 42 changes in the domain of interest or, if such an instrument is not available, we will 43 44 substitute the MID for the SD (denominator) in the SMD equation, which will result in 45 more-readily interpretable MID units instead of SD units[43]. If an estimated of the 46 47 MID is not available we will use a statistical approach developed by Suissa [44]to 48 provide a summary estimate of the proportion of patients who benefit from treatment 49 50 across all studies. The statistical approaches to enhancing the interpretability of results 51 of continuous outcomes outlined in this paragraph will use methods cited as well as 52 those described by Thorlund et. al.[45] ) and risk difference. We will complement this 53 54 presentation by either converting the SMD into natural units of a widely accepted 55 56 19 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 37 of 49 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 instrument used to measure changes in the domain of interest or, if such an instrument is 7 8 not available, we will substitute the MID for the SD (denominator) in the SMD 9 equation, which will result in more-readily interpretable MID units instead of SD 10 units46. If an estimate of the MID is not available we will use a statistical approach 11 12 developed by Suissa 47 to provide a summary estimate of the proportion of patients who 13 benefit from treatment across all studies. The statistical approaches to enhancing the 14 15 interpretability of resultsFor of continuous peer outcomes review outlined in this paragraph only will use 16 48 methods cited as well as those described by Thorlund et. al. Funnel plots will be 17 18 created to explore possible publication bias. Formatted: Font color: Auto 19 Funnel plots will be created to explore possible publication bias.We will use recently 20 21 developed approaches to address missing participant data for dichotomous 22 outcomes[46] and continuous outcomes[47]. We will only apply these approaches to 23 24 patient-important outcomes that meet the following criteria: 1) show a significant 25 treatment effect and 2) report sufficient missing participant data to potentially introduce 26 27 clinically important bias. Thresholds for important missing participant data will be 28 determined on an outcome-by-outcome basis. 29 30 31 Explaining heterogeneity 32
33 http://bmjopen.bmj.com/ 34 We will use recently developed approaches to address missing participant data for 35 49 50 36 dichotomous outcomes and continuous outcomes . We will only apply these 37 approaches to patient-important outcomes that meet the following criteria: 1) show a 38 significant treatment effect and 2) report sufficient missing participant data to 39 40 potentially introduce clinically important bias. Thresholds for important missing
41 on September 27, 2021 by guest. Protected copyright. participant data will be determined on an outcome-by-outcome basis. 42 43 44 Explaining heterogeneity 45 46 We hypothesize the following possible explanations for heterogeneity: (1) Doses Formatted: Body Text, Left, Adjust space 47 between Latin and Asian text, Adjust space 48 (higher vs lower) with expected larger effect with higher than lower doses; (2) Risk of between Asian text and numbers 49 bias, with an expected larger effect in trials at high or unclear risk of bias versus trials at 50 51 low risk of bias; (3) Bacterial and viral illnesses, with larger effect in viral illnesses than 52 bacterial; (4) Age (adult versus pediatric) with postulated larger effect in pediatric 53 54 patients. The presence of heterogeneity will be investigated with the use of likelihood 55 56 20 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 38 of 49 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 ratiochi-squared test statistic. Any and the I2 statistic – the percentage of variability that 7 8 is due to true differences between studies (heterogeneity between the study results Formatted: Font color: Auto 9 would have been described and tested to determine if it reached statistical significance 10 51 52 using a chi-squared test. ) rather than sampling error (chance). Formatted: Font color: Auto 11 12 Formatted: Heading 3, Left, Tab stops: 13 Not at -0.56" 14 Summarizing evidence Formatted: Font: Arial, 13 pt 15 Formatted: Font: Arial, 13 pt, Not Bold We will present resultsFor in Evidence peer Profiles (EP) review as recommended by theonly GRADE 16 17 Working Group[48 49]53 54. EPs provide succinct, easily digestible presentations of 18 quality of evidence and magnitude of effects. Our EPs will be constructed with the help 19 20 of a software program called, GRADEpro (http://ims.cochrane.org/gradepro) to include 21 the following 7 elements: 1. A list of all important outcomes, both desirable and 22 23 undesirable; 2. a measure of the typical burden of these outcomes (e.g. control group, 24 estimated risk); 3. a measure of the difference between the risks with and without 25 26 intervention; 4. the relative magnitude of effect; 5. numbers of participants and studies 27 addressing these outcomes, and follow-up time; 6. a rating of the overall confidence in 28 29 estimate of effect for each outcome and; 7. comments, which will include the MID if 30 available. 31 32 Discussion 33 http://bmjopen.bmj.com/ 34 DISCUSSION 35 36 Our review will evaluate Brazilian herbal intervention for respiratory illness associated 37 38 with cough, provide estimates of the effectiveness of treatments and their associated 39 harms, and evaluate the quality of the evidence in a thorough and consistent manner 40 using the GRADE approach [[50-52].55-57. We will prioritize patient important
41 on September 27, 2021 by guest. Protected copyright. 42 outcomes. The results of our systematic review will be of interest to of interest to public 43 44 health and primary care practitioners in Brazil. Our review will inform these 45 practitioners about best estimates of effect and confidence in those estimates for both 46 47 effectiveness and safety of herbal medicines for URTI and identify key areas for future 48 research. 49 50 51 List of abbreviations 52 CAM = Complementary and alternative medicine 53 54 CCMs = Cough and cold medications 55 56 21 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 39 of 49 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 COPD = Chronic obstructive pulmonary disease 7 8 EP = Evidence Profiles 9 10 GRADE = Assessment, Development and Evaluation 11 12 IN05 = Herbal medicines simplified registration 13 MID = Minimally important difference Formatted: Font: Font color: Auto 14 15 OR = Odds ratio For peer review only 16 17 OTC =Over-the-counter 18 19 RENISUS = Ministry of Health of Brazil has built a list 20 21 SMD = Standardized mean difference Formatted: Font: Font color: Auto 22 23 URTI = Upper respiratory tract infectionand bronchial illness 24 WMD = Weighted mean difference Formatted: Font: Font color: Auto 25 26 27 28 29 Competing interests 30 31 The authors declare that they have no competing interests. 32
33 Authors’ contributions http://bmjopen.bmj.com/ 34 35 LCL is Principal Investigator and led the writing of the manuscript. GG is project 36 manager and co-investigator and contributed to the writing and revision of the 37 manuscript. MCOS, MW, CBM, JCO and AMQ are co-investigators and contributed to 38 the writing and revision of the manuscript. All authors read and approved the final 39 manuscript. 40
41 Competing interests on September 27, 2021 by guest. Protected copyright. 42 43 The authors declare that they have no competing interests. 44 45 46 Supplementary Material 47 Additional file 1: Search Strategy. 48 49 Acknowledgements/ Funding: 50 This project is funded from a grant by CAPES Research Programme (grant) and is Formatted: Font: 12 pt 51 Supportgovernmental Program Graduate Education Institutions Private - PROSUP - 52 Formatted: Font: 12 pt CAPES / UNISO (Process Number: 349.984.188-66) 53 Formatted: Font: 12 pt 54 Formatted: Font color: Auto 55 56 22 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 40 of 49 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 Formatted: Font: 14 pt, Bold, French 6 References (France) 7 Formatted: French (France) 8 9 1.1. Goldman R. Canadian Paediatric Society. Treating cough and cold: Guidance for caregivers 10 of children and youth. Paediatr Child Health 2011;16(9):564-6 11 2. Arroll B. Common cold. Clinical evidence 2006(15):2006-14 12 3. Shields MD, Bush A, Everard ML, et al. BTS guidelines: Recommendations for the assessment 13 and management of cough in children. Thorax 2008;63 Suppl 3:iii1-iii15 doi: 14 10.1136/thx.2007.077370[published Online First: Epub Date]|. 15 4. Ramanuja S, KelkarFor PS. The approach peer to pediatric cough. review Annals of allergy, asthma only & 16 immunology : official publication of the American College of Allergy, Asthma, & 17 Immunology 2010;105(1):3-8; quiz 9-11, 42 doi: 10.1016/j.anai.2009.11.011[published 18 Online First: Epub Date]|. 19 5. Cherry DK, Burt CW, Woodwell DA. National Ambulatory Medical Care Survey: 2001 20 summary. Advance data 2003(337):1-44 21 6. Gwaltney JM. Clinical significance and pathogenesis of viral respiratory infections. The 22 American journal of medicine 2002;112 Suppl 6A:13S-18S 23 7. ALA. American Lung Association. The common cold. http:// 24 www.lungusa.org/diseases/c&f02/cold.html (Accessed 10 Sep, 2013) 2013 25 8. Britt H, Miller GC, Knox S, et al. Bettering the evaluation and care of health - a study of 26 general practice activity (AIHW Cat. No. GEP-10). Australian Institute of Health and 27 Welfare. 2002 28 9. Pratter MR. Cough and the common cold: ACCP evidence-based clinical practice guidelines. Chest 2006;129(1 Suppl):72s-74s doi: 10.1378/chest.129.1_suppl.72S[published Online 29 First: Epub Date]|. 30 10. Chang AB, Glomb WB. Guidelines for evaluating chronic cough in pediatrics : Accp 31 evidence-based clinical practice guidelines. CHEST Journal 2006;129(1_suppl):260S-83S 32 doi: 10.1378/chest.129.1_suppl.260S[published Online First: Epub Date]|. 33 11. Hay AD, Wilson AD. The natural history of acute cough in children aged 0 to 4 years in http://bmjopen.bmj.com/ 34 primary care: a systematic review. The British journal of general practice : the journal 35 of the Royal College of General Practitioners 2002;52(478):401-9 36 12. Mitra A, Hannay D, Kapur A, et al. The natural history of acute upper respiratory tract 37 infections in children. Primary health care research & development 2011;12(4):329-34 38 doi: 10.1017/s1463423611000193[published Online First: Epub Date]|. 39 13. Pratter MR. Cough and the common cold : Accp evidence-based clinical practice guidelines. 40 CHEST Journal 2006;129(1_suppl):72S-74S doi: 10.1378/chest.129.1_suppl.72S[published Online First: Epub Date]|.
41 on September 27, 2021 by guest. Protected copyright. 42 14. Smith SM, Schroeder K, Fahey T. Over-the-counter (OTC) medications for acute cough in 43 children and adults in ambulatory settings. Cochrane database of systematic reviews 44 (Online) 2012;8:Cd001831 doi: 10.1002/14651858.CD001831.pub4[published Online 45 First: Epub Date]|. 46 15. Suara RO, Jr. CJ. Effect of zinc salts on respiratory syncytial virus replication. Antimicrob 47 Agents Chemother 2004;48(3):783-90 48 16. Goldman R. Canadian Paediatric Society. Drug Therapy and Hazardous Substances 49 Committee. Treating cough and cold: Guidance for caregivers of children and youth. Paediatr Child Health 2011;16(9):564-6 50 17. D. H. Efficacy of zinc against common cold viruses: An overview. J Am Pharm Assoc 51 2004;44(5):594-603 52 18. Caruso TJ, Prober CG, Jr. GJ. Treatment of naturally acquired common colds with zinc: A 53 structured review. Clin Infect Dis 2007;45(5):569-74 54 55 56 23 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 41 of 49 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 19. Douglas RM, Hemila H, D'Souza R, et al. Vitamin C for preventing and treating the common 7 cold. Cochrane database of systematic reviews (Online) 2004(4):Cd000980 doi: 8 10.1002/14651858.CD000980.pub2[published Online First: Epub Date]|. 9 20. Arroll B, Kenealy T. Antibiotics for the common cold and acute purulent rhinitis. Cochrane 10 database of systematic reviews (Online) 2005(3):Cd000247 doi: 11 10.1002/14651858.CD000247.pub2[published Online First: Epub Date]|. 12 21. De Sutter AI, Lemiengre M, Campbell H. WITHDRAWN: Antihistamines for the common 13 cold. Cochrane database of systematic reviews (Online) 2009(4):Cd001267 doi: 14 10.1002/14651858.CD001267.pub2[published Online First: Epub Date]|. 15 22. De Sutter AI, van DrielFor ML, Kumar peer AA, et al. Oral antihistamine-decongestant-analgesic review only 16 combinations for the common cold. Cochrane database of systematic reviews (Online) 17 2012;2:Cd004976 doi: 10.1002/14651858.CD004976.pub3[published Online First: Epub Date]|. 18 23. Cano Garcinuno A, Casares Alonso I, Rodriguez Barbero J, et al. [Prescription of systemic 19 cold and cough drugs to children 0-13 years old. An unresolved problem]. Anales de 20 pediatria (Barcelona, Spain : 2003) 2013;78(1):43-50 doi: 21 10.1016/j.anpedi.2012.04.003[published Online First: Epub Date]|. 22 24. Sen EF, Verhamme KM, Felisi M, et al. Effects of safety warnings on prescription rates of 23 cough and cold medicines in children below 2 years of age. British journal of clinical 24 pharmacology 2011;71(6):943-50 doi: 10.1111/j.1365-2125.2010.03860.x[published 25 Online First: Epub Date]|. 26 25. Mourdi N, Dubus JC, Bavoux F, et al. [Mucolytic drugs: towards a contraindication in 27 infants]. Archives de pediatrie : organe officiel de la Societe francaise de pediatrie 28 2010;17(6):735-6 doi: 10.1016/s0929-693x(10)70084-1[published Online First: Epub 29 Date]|. 30 26. Bricks LF, Leone C. [Use of medicines by children attending nursery schools]. Revista de 31 saude publica 1996;30(6):527-35 32 27. Eisenberg DM, Davis RB, Ettner SL, et al. Trends in alternative medicine use in the United Formatted: Font: French (France), Do not check spelling or grammar 33 States, 1990-1997: results of a follow-up national survey. JAMA : the journal of the http://bmjopen.bmj.com/ 34 American Medical Association 1998;280(18):1569-75 . 35 282. Vickers A. Recent advances: complementary medicine. BMJ (Clinical research ed) Formatted: Font: English (U.S.), Do not check spelling or grammar 36 2000;321:683-6 . 37 293. Ekor M. The growing use of herbal medicines: issues relating to adverse reactions and Formatted: English (U.S.) 38 challenges in monitoring safety. Front Pharmacol 2014;4(177):177. 39 4. GUERRA PM, NODARI ORB, al. e. Biodiversidade: aspectos biológicos, geográficos, legais e éticos. In: SIMÕES, M. O. Farmacognosia: da planta ao medicamento. 3. ed. Porto Font: English (U.S.), Do not 40 Formatted: Alegre: UFRGS; Florianópolis: UFSC, 2001. p.15. check spelling or grammar
41 on September 27, 2021 by guest. Protected copyright. 30. REUTERS. Brasil terá primeiro banco de dados de plantas medicinais. Folha Online , Brasil, Formatted: English (U.S.) 42 2002. Disponível em:
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 42 of 49 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 6. BRASIL. Provides for registration of herbal medicines and registration and notification of 7 traditional herbal products. RDC No. 26. 9 Maio 2014. DIARIO OFICIAL DA UNIÃO - 8 SEÇÃO 1 13 DE Brasília, DF 2014. 9 7. WHO. National policy on traditional medicine and regulation of herbal medicines: Report of 10 a WHO global survey. WHO Library Cataloguing-in-Publication Data 11 http://appswhoint/medicinedocs/pdf/s7916e/s7916epdf (Accessed 7 June 2014) 2005. 12 8. DR B. The regulation of herbal medicines in Australia. Toxicology 2002;181-182:565-70. 13 9. WHO. R. Regulatory Situation of Herbal Medicines. A Worldwide Review (WHO/trm/98.1) 14 Geneva: 1998. World Health Organization. 15 http://appswhoint/medicinedocs/pdf/whozip57e/whozip57epdfFor peer review (Accessed only 7 June 16 2014) 1998. 17 10. Harrington P, Shepherd MD. Analysis of the movement of prescription drugs to over-the- counter status. Journal of managed care pharmacy : JMCP 2002;8(6):499-508; quiz 09- 18 11. 19 11. Cohen JP, Paquette C, Cairns CP. Switching prescription drugs to over the counter. BMJ 20 (Clinical research ed) 2005;330(7481):39-41. 21 12. BRASIL. Fitoterápicos são alternativa de tratamento no SUS. Portal do Ministério da Saúde. 22 Disponível em: http://portalsaudesaudegovbr/portalsaude/noticia/8061/162/sus- 23 oferece-fitoterapicos-como-alternativa-de-tratamentohtml 2012. 24 13. BRASIL. Agência Nacional de Vigilância Sanitária. Instrução Normativa n° 05, de 11 de 25 Dezembro de 2008, IN05/2008. Lista de Medicamentos Fitoterápicos de Registro 26 Simplificado. Brasília, DF,. 2008b. 27 35. [Effect of early high-loading-dose tirofiban on platelet activity in patients with acute ST- 28 segment elevation myocardial infarction undergoing primary percutaneous coronary 29 intervention]. Zhonghua xin xue guan bing za zhi;40(2):131-5 30 36.14. Medicine AACP. Over-the-counter medications: use in general and special populations, 31 therapeutic errors, misuse, storage and disposal. 32 http://cymcdncom/sites/wwwacpmorg/resource/resmgr/timetools-
33 files/otcmedsclinicalreferencepdf (acessed 7 of June 2014) 2011. http://bmjopen.bmj.com/ 34 15. Barnes J, Mills SY, Abbot NC, et al. Different standards for reporting ADRs to herbal 35 remedies and conventional OTC medicines: face-to-face interviews with 515 users of 36 herbal remedies. British journal of clinical pharmacology 1998;45(5):496-500. 37 16. Goldman R. Canadian Paediatric Society. Treating cough and cold: Guidance for caregivers 38 of children and youth. Paediatr Child Health 2011;16(9):564-6. 39 17. Arroll B. Common cold. Clinical evidence 2006(15):2006-14. 18. Butler CC, Hood K, Kinnersley P, et al. Predicting the clinical course of suspected acute viral 40 upper respiratory tract infection in children. Family practice 2005;22(2):92-5.
41 on September 27, 2021 by guest. Protected copyright. 19. Hay AD, Wilson A, Fahey T, et al. The duration of acute cough in pre-school children 42 presenting to primary care: a prospective cohort study. Family practice 43 2003;20(6):696-705. 44 20. Gonzales R, Sande MA. Uncomplicated acute bronchitis. Ann Intern Med 2000;133(12):981- 45 91. 46 21. Kusel MM, de Klerk N, Holt PG, et al. Occurrence and management of acute respiratory 47 illnesses in early childhood. Journal of paediatrics and child health 2007;43(3):139-46. 48 22. Britt H, Miller GC, Knox S, et al. Bettering the evaluation and care of health. A study of 49 general practice activity (AIHW Cat. No. GEP-10). Australian Institute of Health and 50 Welfare. http://wwwaihwgovau/WorkArea/DownloadAssetaspx?id=6442455894 51 (Accessed 4 June 2014) 2002. 52 23. Cherry DK, Burt CW, Woodwell DA. National Ambulatory Medical Care Survey: 2001 53 summary. Advance data 2003(337):1-44. 54 24. French CT, Irwin RS, Fletcher KE, et al. Evaluation of a cough-specific quality-of-life 55 questionnaire. Chest 2002;121(4):1123-31. 56 25 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 43 of 49 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 25. Dicpinigaitis PV, Colice GL, Goolsby MJ, et al. Acute cough: a diagnostic and therapeutic 7 challenge. Cough (London, England) 2009;5(11):11. 8 26. Chang CC, Cheng AC, Chang AB. Over-the-counter (OTC) medications to reduce cough as an 9 adjunct to antibiotics for acute pneumonia in children and adults. Cochrane database 10 of systematic reviews (Online) 2012;2:CD006088. 11 27. Smith SM, Schroeder K, Fahey T. Over-the-counter (OTC) medications for acute cough in 12 children and adults in ambulatory settings. Cochrane database of systematic reviews 13 (Online) 2012;8:CD001831. 14 28. Gunn VL, Taha SH, Liebelt EL, et al. Toxicity of over-the-counter cough and cold 15 medications. ForPediatrics 2001; peer108(3):E52. review only 16 29. Centers for Disease C, Prevention. Infant deaths associated with cough and cold 17 medications--two states, 2005. MMWR Morbidity and mortality weekly report 2007;56(1):1-4. 18 30. Centers for Disease C, Prevention. Revised product labels for pediatric over-the-counter 19 cough and cold medicines. MMWR Morbidity and mortality weekly report 20 2008;57(43):1180. 21 31. Kuehn BM. Debate continues over the safety of cold and cough medicines for children. 22 JAMA : the journal of the American Medical Association 2008;300(20):2354-6. 23 32. Mallet P, Mourdi N, Dubus JC, et al. Respiratory paradoxical adverse drug reactions 24 associated with acetylcysteine and carbocysteine systemic use in paediatric patients: a 25 national survey. PloS one 2011;6(7):e22792. 26 33. http://www.cochrane.org/training/ CHfSRoITCC, 2014) c-hAJ. 2014. 27 34. Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items for systematic reviews and 28 meta-analyses: the PRISMA statement. Ann Intern Med 2009;151(264). 29 35. Birring SS, Prudon B, Carr AJ, et al. Development of a symptom specific health status 30 measure for patients with chronic cough: Leicester Cough Questionnaire (LCQ). Thorax 31 2003;58(4):339-43. 32 36. WHO. International Clinical Trials Registry Platform (ICTRP)
33 (http://apps.who.int/trialsearch/) http://bmjopen.bmj.com/ 34 37. Lefebvre C, Manheimer E, J. G. Chapter 6: Searching for studies. In: Higgins JPT, Green S 35 editor(s). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 36 [updated March 2011]. The Cochrane Collaboration. Available from www.cochrane- 37 handbook.org. Chichester, UK: Wiley-Blackwell. 2011. 38 38. Therapeutic Goods Administration-Australian Government’s Department of Health and 39 Ageing. Guideline on clinical investigation of medicinal products indicated for the treatment of psoriasis [online]. Available from: 40 http://wwwtgagovau/docs/pdf/euguide/ewp/245402enpdf 2010.
41 on September 27, 2021 by guest. Protected copyright. 39. Wells GA, Shea B, O'Connell D, et al. The Newcastle-Ottawa Scale (NOS) for assessing the 42 quality of nonrandomised studies in meta-analyses. Secondary The Newcastle-Ottawa 43 Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses 2013. 44 http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp.http://www.ohri.ca/p 45 rograms/clinical_epidemiology/oxford.asp. 46 3740. Guyatt GH, Oxman AD, Kunz R, et al. GRADE guidelines: 7. Rating the quality of evidence- 47 -inconsistency. Journal of clinical epidemiology 2011;64(12):1294-302 doi: 48 10.1016/j.jclinepi.2011.03.017[published Online First: Epub Date]|.. 49 3841. Guyatt GH, Oxman AD, Kunz R, et al. GRADE guidelines: 8. Rating the quality of evidence- 50 -indirectness. Journal of clinical epidemiology 2011;64(12):1303-10 doi: 51 10.1016/j.jclinepi.2011.04.014[published Online First: Epub Date]|.. 52 3942. Guyatt GH, Oxman AD, Kunz R, et al. GRADE guidelines 6. Rating the quality of evidence-- 53 imprecision. Journal of clinical epidemiology 2011;64(12):1283-93 doi: 54 10.1016/j.jclinepi.2011.01.012[published Online First: Epub Date]|.. 55 56 26 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 44 of 49 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 4043. Guyatt GH, Oxman AD, Montori V, et al. GRADE guidelines: 5. Rating the quality of 7 evidence--publication bias. Journal of clinical epidemiology 2011;64(12):1277-82 doi: 8 10.1016/j.jclinepi.2011.01.011[published Online First: Epub Date]|.. 9 4144. RG O. Evaluating coding decisions. In: Cooper H, Hedges LV (editors). The Handbook of 10 Research Synthesis. New York (NY): Russell Sage Foundation, 1994. 11 4245. Montori V, Ioannidis J, Cook DJ, et al. Advanced topics in Systematic Reviews. Fixed- 12 effects and random-Effects models. In: Guyatt G, Rennie D, Meade M, Cook D Users' 13 Guides to the Medical Literature: A Manual for Evidence-Based Clinical Practice 14 McGraw-Hill 2008 . 15 4346. Johnston BC, ThorlundFor K, Schunemann peer HJ, et al. Improvingreview the interpretation only of quality 16 of life evidence in meta-analyses: the application of minimal important difference 17 units. Health and quality of life outcomes 2010;8:116 doi: 10.1186/1477-7525-8- 116[published Online First: Epub Date]|.. 18 4447. Suissa S. Binary methods for continuous outcomes: a parametric alternative. Journal of 19 clinical epidemiology 1991;44(3):241-8 . 20 4548. Thorlund K, Walter S, Johnston B, et al. Pooling health-related quality of life outcomes in 21 meta-analysis—a tutorial and review of methods for enhancing interpretability. Res 22 Syn Meth 2011;2:188-203 . 23 4649. Akl EA, Johnston BC, Alonso-Coello P, et al. Addressing dichotomous data for participants 24 excluded from trial analysis: a guide for systematic reviewers. PloS one 25 2013;8(2):e57132 doi: 10.1371/journal.pone.0057132[published Online First: Epub 26 Date]|.. 27 4750. Ebrahim S, Akl EA, Mustafa RA, et al. Addressing continuous data for participants 28 excluded from trial analysis: a guide for systematic reviewers. Journal of clinical 29 epidemiology 2013;66(9):1014-21.e1 doi: 10.1016/j.jclinepi.2013.03.014[published 30 Online First: Epub Date]|. e1. 31 4851. Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Statistics in 32 medicine 2002;21(11):1539-58.
33 52. Higgins JP, Thompson SG, Deeks JJ, et al. Measuring inconsistency in meta-analyses. BMJ http://bmjopen.bmj.com/ 34 (Clinical research ed) 2003;327(7414):557-60. 35 53. Guyatt GH, Thorlund K, Oxman AD, et al. GRADE guidelines: 13. Preparing summary of 36 findings tables and evidence profiles-continuous outcomes. Journal of clinical 37 epidemiology 2013;66(2):173-83 doi: 10.1016/j.jclinepi.2012.08.001[published Online 38 First: Epub Date]|.. 39 4954. Guyatt GH, Oxman AD, Santesso N, et al. GRADE guidelines: 12. Preparing summary of findings tables-binary outcomes. Journal of clinical epidemiology 2013;66(2):158-72 40 doi: 10.1016/j.jclinepi.2012.01.012[published Online First: Epub Date]|..
41 on September 27, 2021 by guest. Protected copyright. 5055. Guyatt GH, Oxman AD, Kunz R, et al. Going from evidence to recommendations. BMJ 42 (Clinical research ed) 2008;336(7652):1049-51 doi: 43 10.1136/bmj.39493.646875.AE[published Online First: Epub Date]|.. 44 5156. Guyatt GH, Oxman AD, Kunz R, et al. What is "quality of evidence" and why is it 45 important to clinicians? BMJ (Clinical research ed) 2008;336(7651):995-8 doi: 46 10.1136/bmj.39490.551019.BE[published Online First: Epub Date]|.. 47 5257. Guyatt GH, Oxman AD, Vist GE, et al. GRADE: an emerging consensus on rating quality of 48 evidence and strength of recommendations. BMJ (Clinical research ed) 49 2008;336(7650):924-6 doi: 10.1136/bmj.39489.470347.AD[published Online First: 50 Epub Date]|.. 51 52 53 54 55 56 27 57
58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 45 of 49 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 Appendix A: Search Strategy 5 6 7 We will acquire eligible studies through a systematic search of Cochrane Central Register of 8 Controlled Trials (CENTRAL) which contains the Cochrane Acute Respiratory Illness Group’s 9 10 Specialized Register; MEDLINE, EMBASE; CINAHL; Web of Science; AMED; LILACS, CAB 11 abstracts, clinical trial.gov, WHO Trial Register and Brazilian thesis database (CAPES) 12 13 Searches were conducted individually for each plant. Following the logic used to perform the 14 search in databases. 15 For peer review only 16 17 Database: Ovid MEDLINE(R) In-Process & Other Non-Indexed Citations, Ovid MEDLINE(R) Daily 18 19 and Ovid MEDLINE(R) <1946 to Present> Search Strategy: 20 ------21 22 1 common cold.mp. or exp Common Cold/ (4703) 23 2 exp Respiratory Tract Infections/ (289046) 24 25 3 upper respiratory infection*.mp. (1993) 26 4 upper respiratory tract infection*.mp. (4109) 27 28 5 tonsillitis.mp. or exp Tonsillitis/ (8491) 29 6 exp Sinusitis/ or sinusitis.mp. (20370) 30 31 7 otitis media.mp. or exp Otitis Media/ (25752) 32 8 pharyngitis.mp. or exp Pharyngitis/ (15017)
33 http://bmjopen.bmj.com/ 34 9 laryngitis.mp. or exp Laryngitis/ (4279) 35 10 acute bronchitis.mp. (1138) 36 37 11 acute bronchiolitis.mp. (582) 38 12 exp Bronchiolitis/ and exp Acute Disease/ (706) 39 13 exp Acute Disease/ and exp Bronchitis/ (2044) 40
41 14 1 or 2 or 3 or 4 or 5 or 6 or 7 or 8 or 9 or 10 or 11 or 12 or 13 (321961) on September 27, 2021 by guest. Protected copyright. 42 43 This part of the research is performed individually by type of plants (the terms used are described 44 45 in Table 1), as the following example: 46 47 15. ("Echinacea"[Mesh]) OR "Rudbeckia"[Mesh] 48 49 16. Echinaceas OR Echinacea purpúrea OR Echinacea purpúreas OR purpurea, Echinacea 50 OR purpureas, Echinacea OR Coneflower, Purple OR Coneflowers, Purple OR Purple 51 52 Coneflower OR Purple Coneflowers OR Rudbeckia purpúrea OR Rudbeckia purpúreas OR 53 purpurea, Rudbeckia OR purpureas, Rudbeckia OR Echinacea angustifólia OR Echinacea 54 55 angustifólias OR angustifolia, Echinacea OR angustifolias, Echinacea 56 17. ("Eucalyptus"[Mesh]) AND "Eucalyptus terpene oil" [Supplementary Concept] 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 46 of 49 BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 18. Eucalyptus globules OR Eucalyptus OR Eucalipto 5 19. 15 or 16 or 17 or 18 6 7 8 From here the same for all plants, terms to search for the types of study design to be included 9 (typically a ‘filter’ for randomized trials). 10 11 20 randomized controlled trial.pt. (376556) 12 13 21 controlled clinical trial.pt. (88568) 14 22 randomized.mp. (568931) 15 For peer review only 16 23 20 or 21 or 22 (651127) 17 24 14 and 19 and 23 (8) 18 19 *************************** 20 21 Table 1.1.Table Descriptors by plants 22 23 Plants Mesh terms Others terms 24 A. comosus "Ananas"[Mesh] Ananas comosus OR Anana OR Pineapple OR Pineapples 25 E. purpurea ("Echinacea"[Mesh]) OR "Rudbeckia"[Mesh] Echinaceas OR Echinacea purpúrea OR Echinacea purpúreas OR 26 purpurea, Echinacea OR purpureas, Echinacea OR Coneflower, Purple OR 27 Coneflowers, Purple OR Purple Coneflower OR Purple Coneflowers OR Rudbeckia purpúrea OR Rudbeckia purpúreas OR purpurea, Rudbeckia OR 28 purpureas, Rudbeckia OR Echinacea angustifólia OR Echinacea 29 angustifólias OR angustifolia, Echinacea OR angustifolias, Echinacea 30 E. globules ("Eucalyptus"[Mesh]) OR "Eucalyptus terpene oil" Eucalyptus globules OR Eucalyptus OR Eucalipto [Supplementary Concept] 31 G. glabra "Glycyrrhiza"[Mesh] OR "Glycyrrhiza Glycyrrhizas OR Licorice OR Licorices OR Liquorice OR Liquorices OR 32 uralensis"[Mesh] OR "licorice-saponin L3" Glycyrrhiza uralenses OR uralenses, Glycyrrhiza OR uralensis, Glycyrrhiza
33 [Supplementary Concept] OR Gan zao OR Gan zaos OR zao, Gan OR zaos, Gan http://bmjopen.bmj.com/ 34 H. helix ("Hedera"[Mesh]) OR ("hederacoside C" Hederas OR Hedera helix OR Hedera hélices OR helices, Hedera OR helix, 35 [Supplementary Concept] OR Hedera OR Ivy, English OR English Ivy 36 "didehydrofalcarinol" [Supplementary Concept] ) 37 M. sylvestris "Malva"[Mesh] Malva sylvestris OR Malva OR Malvas OR Malva silvestre 38 M. piperita and M. "Mentha"[Mesh] OR "Mentha piperita"[Mesh] Mentha piperita OR Mentha villosa OR Menthas OR Mint OR Mints OR Mentha piperitas OR piperita, Mentha OR piperitas, Mentha OR Peppermint 39 villosa OR Peppermints OR Menta 40 M. glomerata and "Mikania"[Mesh] Mikania glomerata OR Mikania laevigata OR Cipó-caatinga OR Cipó-
41 catinga OR Guaco OR Guacos OR Mikanias on September 27, 2021 by guest. Protected copyright. 42 M. laevigata 43 P. sidoides "Pelargonium"[Mesh] Pelargonium sidoides OR Pelargoniums OR Umckaloabo 44 P. hybridus "Petasites"[Mesh] Petasite OR Butterbur OR Butterburs OR Coltsfoot, Sweet OR Coltsfoots, 45 Sweet OR Sweet Coltsfoot OR Sweet Coltsfoots P. anisum "Pimpinella"[Mesh] OR "Foeniculum"[Mesh] Pimpinellas OR Pimpinella anisum OR 46 Pimpinella anisums OR anisum, Pimpinella OR anisums, Pimpinella OR 47 Anise OR Anises OR Erva-doce OR Anis OR Fennel OR Fennels 48 P. senega "Polygala"[Mesh] Polygalas OR Snakeroot, Seneca OR Seneca Snakeroot OR Seneca 49 Snakeroots OR Snakeroots, Seneca OR Senegaroot OR rattlesnake root OR mountain flax 50 P. ipecacuanha "Cephaelis"[Mesh] Cephaeli OR Uragoga OR Uragogas OR Cephaelis ipecacuanha OR 51 Cephaelis ipecacuanhas OR ipecacuanha, Cephaelis OR 52 ipecacuanhas, Cephaelis OR Psychotria ipecacuanha OR Psychotria ipecacuanhas OR ipecacuanha, Psychotria OR 53 ipecacuanhas, Psychotria OR Cagosanga OR Raiz-do-Brasil 54 S.nigra "Sambucus nigra"[Mesh] OR "Sambucus nigra Sambucus nigras OR nigra, Sambucus OR nigras, Sambucus OR Elder, 55 lectins" [Supplementary Concept] Black OR Elder, European OR Elders, European OR European Elder OR 56 European Elders OR Black Elder OR Black Elders OR Elders, Black 57 58 59 60 Page 2 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 47 of 49 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 For peer review only 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40
41 on September 27, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 3 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml
BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from 8 8 9 9 9 9 8-9 8-9 Page 48 of 49 AppendixA on page # # page on Reported Reported 12-13 12-13 12-13 12-13 10-11 10-11 10 10 10 10 9 8 4-7 2 1 up) report up) characteristics and (e.g., considered, years
BMJ Open http://bmjopen.bmj.com/ analysis. - analysis). - on September 27, 2021 by guest. Protected copyright.
foreach meta ) ) For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml 2 registration registration informationincluding registration number. Describe the Describe methods of handling dataand combining ofresults studies, if done, includingmeasures of consistency I (e.g., State the State principal summarymeasures (e.g., risk ratio,means). difference in Describe methods Describe used for assessing risk studies of of bias individual (including specification of this whether was the study at done outcomeor andhow this level), information be usedto is in any datasynthesis. List and define List variables all for data which sought were (e.g., PICOS,fundingsources) any assumptionsand and simplificationsmade. Describe method Describe of dataextraction from reports(e.g., forms, piloted inindependently, duplicate) and any processes forand confirming obtaining datafrom investigators. State the State process for studies selecting (i.e., screening, eligibility, includedin systematic and,if review, applicable, includedin the meta Present full electronicPresent search strategyfor least one at database, including limits any used, such that it could be repeated. Describe all Describe information sources databases (e.g., of with dates coverage, contact study with authors to identify studies) additional in the search and datelast searched. Specify study Specify characteristics PICOS, (e.g., length of follow - Indicate Indicate review if a protocolexists, if andcan it where Webaccessed (e.g.,be address), and, if available, provide Provide anProvide statement explicit of questions being addressedwithreference to participants, interventions,comparisons, outcomes,study and (PICOS). design Describe the Describe rationale for review inthe the context of what already is known. Provide a structured a Provide summary including, asapplicable: background; objectives;datasources; study eligibilitycriteria, participants,and interventions; study appraisal and synthesis methods; results; limitations;conclusions and implicationsoffindings; key systematic registration review number. Identify the Identify report systematic asa meta-analysis, review, both.or Checklist item item Checklist For peer review language, publicationstatus) ascriteria used for eligibility, rationale. giving only
1 8 2 4 7 9 3 6 # 11 12 14 13
PRISMA2009 ChecklistPRISMA 2009 Checklist PRISMA2009 ChecklistPRISMA 2009 Checklist
Synthesisof results Summarymeasures Riskindividual in of bias studies Data Data items Data collection Data process 10 Study selection Study Search Search Information sourcesInformation Eligibility criteria Eligibility Protocol and registrationProtocol 5 METHODS METHODS Objectives Objectives Rationale Rationale INTRODUCTION INTRODUCTION Structuredsummary ABSTRACT ABSTRACT Title Title TITLE TITLE Section/topic Section/topic 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
BMJ Open: first published as 10.1136/bmjopen-2014-005267 on 23 July 2014. Downloaded from
on page # # page on Reported Reported 14 14 NA NA 3 13 13 NA NA NA NA NA NA NA NA NA NA NA NA NA NA 13 13 13 13 . . - statement.org
www.prisma
Page 2 of 2 2 of Page Page 1 of 2 2 1 of Page BMJ Open http://bmjopen.bmj.com/ For more information, visit: For information, more on September 27, 2021 by guest. Protected copyright.
specified. -
For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Describe sources Describe of funding for systematic the review support and (e.g., other supply of data); role of funders for the systematic review. Provide a a Provide general interpretationof the results thecontext in of other evidence, and implications for future research. Discussstudy limitations at and outcome(e.g., level risk review-level ofand at bias), (e.g., incomplete retrieval of identifiedresearch, reporting bias). Summarizethe main findings the strengthincluding ofevidencefor main each outcome; their consider relevance to Give results ofresults Give additional analyses, if (e.g.,done sensitivity subgroupor analyses, meta-regression Item [see 16]). Present results Present of any assessmentof risk across of (see Itembias studies 15). Present results Present of each meta-analysis done, including confidence intervalsand measures of consistency. For all outcomes Forall considered(benefits harms),or for study: each present, (a) simple summary datafor each groupintervention (b) effectestimates confidence and intervals, forest ideally witha plot. Present Present dataon risk of ofstudy and, each bias ifavailable, outcome any assessmentlevel (see item 12). For each study, Foreach present characteristics for which data extracted were (e.g., study PICOS,follow-upsize, period) and citations. provide the Give numbersGive screened,of studies assessed for eligibility, and includedin the review, reasons with for exclusions at stage, each flow ideally witha diagram. Describe methods Describe of additional analyses (e.g., sensitivity subgroupor analyses, meta-regression),if indicating done, Specify Specify assessment any riskofthat may bias cumulative affectof evidence(e.g., the publication bias,selective studies).reporting within Checklist item item Checklist key groups key (e.g., healthcare providers, users, and policymakers). For prewhich were peer review only
# 21 17 25 27 26 16 23
PRISMA2009 ChecklistPRISMA 2009 Checklist PRISMA2009 ChecklistPRISMA 2009 Checklist
Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group (2009). Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. PLoS Med 6(6): e1000097. From: Funding Funding FUNDING FUNDING Conclusions Limitations Summaryof evidence 24 DISCUSSION DISCUSSION Riskacross of bias studies 22 Synthesisof results Results of Results individual studies 20 Riskstudies within of bias 19 Study characteristics Study 18 Study selection Study RESULTS RESULTS Riskacross of bias studies 15 Section/topic Section/topic doi:10.1371/journal.pmed1000097 doi:10.1371/journal.pmed1000097 Additional Additional analysis Additional Additional analyses Page 49 of 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60