Upload
paolo
View
212
Download
0
Embed Size (px)
Citation preview
Pediatric Pulmonology 49:938–940 (2014)
Letters to the Editor
Cystic Fibrosis Database (CFDB):A new web-based tool for cystic fibrosis specialists
Roberto Buzzetti, MD,1* Natalia Cirilli, BSc (Hons),2 Laura Minicucci, MD,3 Valeria Raia, MD,4
Donatello Salvatore, MD,5 and Paolo Maffeis, MD6
Summary. In order to help specialists involved in CF care and clinical research to know the
current best evidence about clinical effectiveness of interventions in CF, we designed and
developed a web-based, free access tool called “CFDB”—Cystic Fibrosis DataBase (www.
inetflow.it/CFDB). The databasewas built by searching inMedline, Embase, CochraneLibrary and
worldwide trials registries all studies involving clinical interventions in CF. The tool lets the user
define queries starting from one or more types of pathological conditions and one or more
interventions. The output of the queries is structured in three levels: (1) how many and which
studies deal with the conditions formulated in the query; (2) which are the main results of these
studies; (3) a critical summary of the literature related to the query. This tool, providing a quick
overview of the available evidence in clinical research in CF, may help clinical decision making,
designing of new trials and building guidelines. Pediatr Pulmonol. 2014; 49:938–940.
� 2014 Wiley Periodicals, Inc.
Key words: database; clinical trials; systematic reviews; cystic fibrosis.
Funding source: Italian Cystic Fibrosis Foundation.
INTRODUCTION AND OBJECTIVES
Time limitation is one of the biggest obstacles toevidence-based practice but clinicians should seekevidence from as high in the appropriate hierarchy ofevidence as possible. Moreover, clinicians need to be ableto distinguish high from low quality in primary studies,systematic reviews, practice guidelines, and otherintegrative research focused on management recommen-dations.1 Therefore, physicians and other health careprofessionals, involved either in cystic fibrosis care or incystic fibrosis research, would greatly benefit from aspecific Information-Technology tool, especially de-signed to give them, in a short timeframe, the state ofthe art of a clinical topic. All health care professionals andresearchers need to know what studies can give them theproper answer to their questions, no matter if the questionis which antibiotic to use for Pseudomonas infection, ifglutathione can be useful for CF patients or what kind oftherapy is best for a patient. Sometimes they can find aclear answer (intervention works—or the interventionwill not work). It often happens that they remain uncertainthough, as the studies conducted so far do not give a clear
1Italian CF Research Foundation, Verona, Italy.
2CF Centre, United Hospitals, Ancona, Italy.
3CF Centre, Gaslini Hospital, Genova, Italy.
4CF Centre, Federico II University, Naples, Italy.
5CF Centre, Centro Pediatrico Bambino Gesu Basilicata, San Carlo
Hospital, Potenza, Italy.
6Inetflow Service SRL, Bergamo, Italy.
Conflict of interest: None.
�Correspondence to: Roberto Buzzetti, MD, Fondazione Ricerca Fibrosi
Cistica, Ospedale Maggiore, P.le Stefani 1, 37126 Verona, Italy.
E-mail [email protected]
Revised 30 September 2013; Accepted 7 October 2013.
DOI 10.1002/ppul.22954
Published online 16 February 2014 in Wiley Online Library
(wileyonlinelibrary.com).
� 2014 Wiley Periodicals, Inc.
answer, and further studies would be necessary. Mean-while, different entities have devoted themselves todeveloping Web tools for helping users quickly andefficiently search and retrieve relevant publications.These practices, have led to an increase in the numberand quality of various Web tools that provide literaturesearch service.2
The primary objective of this project was to helpspecialists involved in CF care and clinical research toknow the current best evidence about clinical effective-ness of interventions in CF. We have designed anddeveloped a web-based, free access tool called“CFDB”—Cystic Fibrosis DataBase (www.inetflow.it/CFDB) that could satisfy the clinical questions for CFhealth care professionals. In this short communication wedescribe this tool.
METHODS
Search of Studies
A project team of clinicians together with a leadingexpert in epidemiological and clinical research made asystematic search in Medline (http://www.ncbi.nlm.nih.gov/pubmed/), Embase (https://www.embase.com/), Co-chrane Library (http://www.thecochranelibrary.com/)[Cochrane Database Of Systematic Reviews (CDRS);Database Of Abstracts Of Reviews Of Effects (DARE);Cochrane Controlled Trials Register (CCTR/CEN-TRAL); Health Technology Assessment Database(HTA); NHS Economic Evaluation Database (NHSEED)], and in worldwide trials registries (http://www.clinicaltrials.gov/, https://www.clinicaltrialsregister.eu,http://apps.who.int/trialsearch/) using “cystic fibrosis”as a keyword or a title word, with no limit of time orlanguage.
Selection of Studies
Two independent reviewers (R.B. and N.C.) assessedstudies and trials for inclusion.The inclusion criteria were the following: primary
studies (interventional or observational, ended or ongo-ing) and secondary studies (systematic reviews, healthtechnology assessment reports) focusing on the effective-ness of interventions in CF. Ongoing studies from trialsregistries (only phase II, III RCTs, and phase IV studies)were also included.For the non-pharmacological interventions, also con-
trolled and observational studies (cohort studies, case–control studies) were included; for pharmacologicalinterventions and for ongoing trials only randomizedcontrolled studies were selected.Studies with patients other than CF, or having only few
CF patients, or without a control group, or focusing on
“basic science,” etiology, epidemiology, prognosis,guidelines were excluded.
Classification of Studies
Studies were classified according to an “ad hoc”dictionary of keywords. For systematic reviews thefollowing variables were analyzed: code, reference(year, title, authors, journal, and registry), interventions,
TABLE 1—The List of Topics (Level 3)
Abnormal glucose metabolism
ABPA
Active cycle of breathing technique
Agents which reduce gastric acidity
Antibiotic treatment of early Pseudomonas aeruginosa
Antibiotics for pulmonary exacerbations
Antioxidant supplementation
Azithromycin
Bisphosphonates
Bronchodilators
Calorie supplements
CFTR correctors—Potentiators
Chest physical therapy
Chronic rhinosinusitis
Dornase alpha
Exercise and physical training
Fibrosing colonopathy
Fungi
Gene therapy
General nutrition recommendations
Growth hormone
Hemoptysis
Hypertonic saline
Inhaled antibiotics
Inhaled corticosteroids
Liver diseases
Lung transplantation
Mycobacteria
MRSA Eradication
Nasal polyposis
Non-invasive ventilation
Non steroid anti-inflammatory drugs
O2 therapy
Omega-3—zinc probiotics
Oscillating devices
Osteoarthritis
Palivizumab for prophylaxis against respiratory syncytial virus
Pancreatic enzyme
Pancreatitis
Pneumothorax
Positive expiratory pressure
Prophylactic use of oral antistaphylococcal antibiotic
Psychological interventions
Scheduled IV antibiotics every 3–4 months—symptom-based
treatment
Screening
Systemic corticosteroids
Vaccination program
Vitamins A, E, K
Vitamin D
CFDB: A Web-Based Tool for CF 939
Pediatric Pulmonology
number of included studies, outcome measures, mainresults and author’s conclusions.
Database Refinement and Update
If the results of a study had multiple publications onlyone version was retained and the same criteria wereadopted for studies submitted to more conferences. Whena conference abstract or an “ongoing trial” is subsequentlypublished as an article, it is deleted from the database andreplaced by the published paper. The database is regularlyupdated and revised to improve the quality of data.
Topics Summaries
Three CF specialists (L.M., V.R., and D.S.) wrote acritical updated summary of the results from the literature,for some relevant issues (see Table 1), including a briefbackground, what is known and what is needed to study.
RESULTS
To March 1, 2013 the database included 1,784 recordsas follows: 81 Cochrane reviews, of which 15 protocols,124 other reviews (including Health Technology Assess-ment reports and Economic Evaluations), 1,376 primarystudies (randomized and non-randomized, includingcongress abstracts), and 203 ongoing trials.A user-friendly web-based tool helps the user to define
queries starting from one or more types of pathologicalconditions (concerning CF) and one or more interven-tions. The output of the queries is generally on threelevels: how many and which studies (primary andsecondary, ended or in progress) deal with the conditionssubmitted in the query (level 1); which are the mainresults of clinical trials (level 2) and, for some relevanttopics, a brief critical updated summary of the resultsfrom the literature (level 3).To build a query, the user can select one or more terms
from five search menus:
1. Condition/Disease/Complication (e.g.: diabetes, preg-nancy, etc. CF is always implied).
2. General class of Intervention (pharmacological or non-pharmacological [e.g.: diagnosis, surgery, comple-mentary medicine, etc.])
3. Category of intervention (e.g.: antibacterial agents,artificial ventilation, etc.)
4. Specific Intervention (e.g.: Amikacin, Calcium, etc.)5. Mode of Administration (how the intervention is
administered).
Within each menu, the terms are in “OR”, and the usercan choose so many items as he wants; among menus theterms are in “AND.” Example: (Pseudomonas aeruginosaOR Burkholderia cepacia) AND (Tobramycin ANDInhalation/Nebulized).When the query is complete, by clicking on a “Search”
button the user can read the results and find the link to theoriginal publication. The user can also read, for all theselected studies, several details: date, criteria for inclusionin the review (or study design), participants, interven-tions, outcome measures, main results and authors’conclusions. An option “see also” allows to readsummaries about the main related topics.
DISCUSSION
This tool may help clinicians, researchers and studentsto have a fast updated view of clinical research in CF byperforming queries on the main topics in CF care. It couldalso be helpful to anyone who is going to design newstudies, as it provides a concise description of what iscurrently known and what issues, on the contrast, needadditional research.It may be of help to those who must make
recommendations (guidelines), providing a first quickoverview of the available evidence.We hope that the database, which was built with so
much effort (includes about 1,800 citations accuratelyclassified) and which is constantly kept up to date, willserve the largest possible number of users by providinga quick and comprehensive guide ripples that can helpprofessionals in the correct decisions.It would be desirable to enlarge the team members
with international experts to improve the quality of thecontents and the methodological analysis.
ACKNOWLEDGMENTS
This project was supported by an operating grant fromthe Italian Cystic Fibrosis Foundation.
REFERENCES
1. Guyatt GH, Haynes RB, Jaeschke RZ, Cook DJ, Green L, Naylor
CD, Wilson MC, Richardson WS. Users’ Guides to the Medical
Literature: XXV. Evidence-based medicine: principles for apply-
ing the Users’ Guides to patient care. Evidence-Based Medicine
Working Group. JAMA 2000;284:1290–1296.
2. Lu Z. PubMed and beyond: a survey of web tools for searching
biomedical literature. Database 2011; baq036.
940 Buzzetti et al.
Pediatric Pulmonology