The second and third stages of DM. Finding information on the Internet and electronic database. The system of bibliographic search of medical information - Medline. Analysis of the articles and their critical evaluation

Department of Clinical Pharmacology and Evidence-Based Medicine.Discipline is "evidence-based medicine."Course 3Speciality 051301 "General Medicine".NEWS AND DIDACTIC UNITPO "evidence-based Medicine"To a practical training №.

Topic: "The second and third stages of DM. Finding information on the Internet and electronic database. The system of bibliographic search of medical information - Medline. Analysis of the articles and their critical evaluation. "

Information Systems      Database - a major component of modern information systems. Information system - an interconnected set of tools, methods and personnel used for storage, processing and issuance information.     The purpose of any information system - information processing a particular subject area. Example - clinical decision support systems. Usually they are databases that integrate electronic medical records with special tools, such as automatic reminders or drug dosage calculators for use by clinicians in decision-making. Based support systems should be based on evidence-based clinical practice guidelines or other reliable scientific evidence.     Currently, medical information systems - one of the priority areas of software development throughout the world and are time-consuming task due to several factors:• a variety of factors and complexity of their interaction in the process of making medical decisions;• lack of standardization in terminology, format, scales of measurement;• high demands on computer settings for the access speed, memory, graphics processing.     Database. Database can be defined as an organized set of interrelated data stored together represented in electronic media, designed and suitable for special tasks using computer aids.Features a collection of data:• large volumes of information;• the maximum possible compactness of data storage;• the possibility of extracting from the database information into a specific subject area;• user-friendly form and shape of the extracted information;• High speed data access;• reliability of data storage and the ability to provide authorized access to data for individual users;• usability design user queries, forms and reports for data retrieval.     Creation of the database, its support and providing the user access thereto is carried out using special software tools - database management systems. Database management system (DBMS) - software for creating and editing databases, view and search information in them.    Using the technology of processing and storage database is divided into centralized and distributed.Centralized DB access methods are divided into:• Database of local access (data and procedures for processing are stored on one machine);• database with remote (network) access. Database management systems with remote access can be built using architectures file server and client-server.Architecture of a file serverThe organizing principle: one machine is selected as a central (server files), it is stored on a centralized database. Other machines network serve as workstation. Database files in accordance with user requests workstations, these stations are transmitted to and processed there. Performance of the system falls, if you want an intense simultaneous access to the same data. Currently, the architecture considered obsolete.     Client-server architecture     The organizing principle: a central machine (database server) keeps a centralized database and processing procedures. The client sends a request, it is processed by the server and the data obtained on request, transmitted to the client.A distributed database consists of several parts that are stored on different computers of the network. This method of treatment implies the presence of multiple servers (computer technician), which can be stored intersecting or even duplicating information. To work with a database using a system of distributed databases.According to the model database is divided into a hierarchical, networked relational, object-oriented and hybrid.      Hierarchical model database consists of objects with pointers from parent to offspring, linking together related information.The basic concepts of the network model database include: the level element (node) and communication. Node - a set of data attributes that describe an object. The network structure of each element is associated with any other element. Network databases are similar to the hierarchical, except that they have signs in both directions, which connect related information.       Relational model (English relation - relation) database focused on data organization in the form of two-dimensional tables. For these models is characterized by ease of data structures, user-friendly tabular representation. In relational databases using the language Structured Query Language - Structured Query Language - the universal language used to create, modify and manage data.    Inobject-oriented databases are stored as objects. With the object-oriented database is convenient to work, applying object-oriented programming. Today, however, such databases have not yet achieved such popularity as relational, as far inferior to them in performance.     Hybrid DBMS combines the functionality of relational and object-oriented databases.   Today the database - a powerful tool for improving processes of care. The emergence in the late XX century. medical database accessible to a wide range of users, allowing for fast searching, analysis and synthesis of information on the results of epidemiological and other research and contributed to the development approach to health care that has received the name of "medicine based on evidence" (evidence-based medical practice).         Choosing a strategy for information retrieval    Common types of issues in medical practice. Of the many issues that may be interested in the health care workers can identify the most common types. These are issues relating to clinical data, symptoms, prognosis, prevention, etiology and causes of disease, injury and risk of diagnostic tests, differential diagnosis, quality of life, therapy, cost effectiveness, etc.For the purposes of developing strategies for information retrieval can find 5 common types of clinical questions, which include all other types.   •Treatment (therapy) - questions about what treatment (if possible) should be appointed and what could be the results of different treatment methods. The group also includes questions regarding prevention, organization of screening programs, improving quality of care.   • Diagnosis - questions about the reliability and clinical utility of specific tests that are usually put to determine whether this test will bring enough benefit a particular patient. In most articles on the diagnosis results of investigation of the diagnostic test are compared with results of another standard test, which is considered as a reference or authoritative (the "gold standard"). They also include questions on the differential diagnosis.    •Forecast - questions about the future health of the patient, duration and quality of life when choosing a particular treatment option.    •Etiology / side effects- questions about the relationship between the disease and its possible causes, including side effects associated with various treatment options (drugs, therapeutic and diagnostic interventions).    •Cost-effectiveness - questions about cost-effectiveness of various therapeutic, preventive and diagnostic procedures.Using the technology of evidence based medicine involves searching for information in the literature and databases. The primary step of EBP (EBM) - the exact wording of "focused" issue.

Table 1 shows the pros and cons of the available sources of information.





Can be taken in the library

It takes a lot of time for searching and browsing


More recent informationEasy access Compact

Do not always contain reviewsRoadsAccess to libraries islimitedIt takes a lot of time toSearch and browseThere are too manyPublication bias

Journals secondary information


The full study

Abstracts article

In general, good quality

Often written by the authors themselves

Bibliography / reference lists

Useful links to related literature on a topicIsolated articles of high quality

The subjectivity of the authorchoiceUnclear criteria


Help is readily available

The Partiality.

Priorities may differ from yourLack of time

World Wide Web

The most recent dataValuable / unique informationProfessional networkingFull textAccess to information from around the worldHour access

Lack of a central directory - it is difficult to find informationThe lack of censorship and editingNo guarantee of quality andreliabilityRequires an Internet connection and minimal skills in the fieldInformation Technology



The threat of viruses

Mailing Lists

A one-to-many specific themeYou can only receive weekly reviewsA wide range of viewsVery cheap means of distributionCan be opened or closed to outsidersYou can search or browse the archives

The number of messages may be different erroneous information distributed to allDifficult to maintain the discussion

Bibliographic Database

A wide range of indexed journalsRegular updatingSaving of timeEase of UseFree text search developed search tools

Lack of full textReflects the bias of publicationsNot always availableSometimes inconvenient to useIt is difficult to obtain foreign article


Search by type of quality publications available filters

Bias in favor of U.S. operationsA selective approach to the selection of journalsPoor indexing

The Cochrane Library


Some subject areas are underrepresented

In terms of representation and integration of data, there are different kinds of resources (both primary and secondary): magazines, books, reports, manuals, DB. Information on clinical effectiveness is now available in a variety of medical journals and health - both national and international.      Journals secondary informationThere are magazines that publish it secondary information - overviews and summaries of evidence.       ACP Journal ClubURL: http://www. acpic. org / shared / menuabout. htmlThe primary goal of American College of Physicians Journal Club is to choose from the available biomedical literature articles describing original research and systematic reviews, which are necessary for physicians seeking to use modern advances of medicine. These articles are summarized in the form of abstracts and commented on by clinical experts. Materials for publication carefully selected from more than 100 clinical journals, using strict criteria of scientific quality and subsequent evaluation of fitness for a medical practice carried out by clinical experts.URL: http://ebm. bmiiournals. com /This is a joint publication of the British Medical Journal (BMJ) Publishing Group and American College of Physicians published bimonthly and is using less stringent criteria than the ACP Journal Club. EBM publishes essays of special interest in internal medicine, surgery, pediatrics, obstetrics and gynecology. This magazine has a strong European bias. Free access for developing countries.       Evidence-Based Mental HealthURL: http://ebmh. bmiiournals. com /Purpose of the publication Evidence-Based Mental Health is to inform psychiatrists, clinicians about important and applicable in clinical advances in treatment (including specific interventions and systems of care), diagnosis, etiology, prognosis and research results, quality improvement, training and economic evaluation. To this end, staff Magazine selected original and review articles whose results seem to be most accurate and clinically useful. Then, these articles are summarized in the form of essays and commentaries complement of expert clinicians. Free access for developing countries.       Evidence-Based NursingURL: http://ebn. bmiiournals. comA quarterly magazine published by the Royal College of Nursing and BMJ Publishing. To help the nurses, he finds and evaluates the quality and clinically important research, publishing a short informative summaries of critical articles with commentaries nurse practitioners, can determine the location of a new study in the field of medicine. Free access for developing countries.International journal of medical practice (in Russian) URL: http://www. mediasphera. ru / mimp / mimp-mn. htm # journal publishes structured abstracts of key, important for practitioners of clinical and epidemiological studies with commentaries, clinical guidelines, articles on methodology for clinical and analytical studies of biomedical statistics. The site contains the full text of all articles. Publication resumed.BMJ UpdatesURL: http://bmiupdates. mcmaster. ca / index. asp? choice = 13 This free service - a joint project of BMJ Publishing Group and McMaster University (Canada). It is possible to subscribe (via email) about new and better evidence-based publications on the clinical specialty. The service also includes access to a database of summaries ofarticles from 2002.

Database of primary informationMEDLINEURL: http://www.pubmed. comDB U.S. National Library of Medicine that contains bibliographic descriptions of articles and essays (70% of the articles have abstracts). Currently the database covers nearly 4,500 journals in the U.S. and more than 70 other countries. MEDLINE includes descriptions of 12 million articles published from 1966 to present. New records are added weekly, 400 000 records per year. For each record in MEDLINE are given medical subject headings (MeSH Terms) and the types of publications (Publication Types) in the dictionary, controlled the U.S. National Library of Medicine. Bibliographic descriptions and abstracts of MEDLINE forms the basis and are available in the database the U.S. National Library of Medicine PubMed, which can be searched via the Internet. MEDLINE contains over 13 million records. Although the database does not include the full texts of articles, but they can find a link to another site (where they are free or paid).EMBASEURL: http://www. /The database contains information on biomedicine and pharmaceuticals, including biological sciences, biochemistry, clinical medicine, forensic medicine, pediatrics, pharmacy, pharmacology and drug therapy, pharmacoeconomics, psychiatry, public health, biomedical engineering and instrumentation, the environment. Sources EMBASE - more than 3,800 journals from 70 countries, monographs, conference proceedings, dissertations and reports. Search can be carried out by bibliographic information indexed terms, abstracts, chemical name, trade name drugs, the name of the producer LS, trade names, medical devices, the names of their producers, the molecular sequences       Database of secondary informationCochrane Library - The Cochrane Library URL: http.V / www.theCochraneLibrarv. com The library contains four databases:• Database of Systematic Reviews;• base abstracts the effectiveness of medical interventions;• Register of Controlled KI;• base the work on the methodology of the reviews.The Cochrane Library is available on the Internet, or it can be bought on CD.Articles published in The Cochrane Database introduce participants Cochrane collaboration. Each party is to manually search for a specific clinical journal until his first issue.Using strict methodological criteria, these people classify each article according to the type of publication (RCTs, other controlled, ie, an epidemiological survey, etc.) and preparing structured abstracts. Quantitative data in the surveys are presented in a standard graphical format so that the doctor was able to quickly and objectively assess them.Clinical EvidenceURL: http://www. clinicalevidence.comEvidence-based medicine. Free access for developing economiesThis resource has a number of features.• Its content is formed on the basis of issues rather than the availability of evidence-based research.Here identify important clinical questions, and then to seek out and summarize the best available evidence to answer them.• Continuously updated (complete literature search for each topic, spend every eight months).Clinical Evidence specifically does not endorse, and provides material for the development of clinical guidelines of the local applications for clinicians and patients seeking to make up their own idea about the optimal course of treatment.Currently, this resource is published in Russian - http://www. mediasphera. ru / clinicalevidence/.Up To DateURL: http://www. uptodate. comUpToDate - clinical information resource, which is designed community of Ztys.experienced clinicians and is designed so that doctors could directly in the workplace to get a brief and specific answers to clinical questions. Sections UpToDate summarize the published evidence (including Cochrane reviews) and offer practical recommendations for the treatment of patients that allows physicians to:• Obtain the most up to date information on their specialty;• recognize the clinical signs of many diseases and to determine the real possibilities of diagnosis and treatment, including efficacy, dosage and compatibility of medicines;• develop optimal strategies for screening and prevention. UpToDate - formal educational program, which is recommended or are preparing, in collaboration with eight major professional medical associations, the United States.

International cooperationBEME Collaboration - Best Evidence Medical EducationURL: http://www. bemecollaboration. org /Development cooperation in evidence-based approaches in medical education BEME Collaboration is an association of individuals and institutions which have as their objective the development of evidence-based approaches in medical education. Ways to solve this problem: information dissemination, allowing teachers, institutions and all those involved in medical education, make decisions based on the best scientific evidence available, preparation of systematic reviews of medical education that reflect the best available data and meet the needs of the user; cultivation of evidence-based approach in medical education among teachers, departments and agencies. A number of thematic groups Ouémé in the final stages of preparing their reports.Evidence for Policy and Practice Information and Coordinating Centre - EPPI-CentreURL: http://eppi. ioe. ac. uk / EPPIWeb / home. aspxClearinghouse of evidence in policy and practice - EPPI-Centre. EPPl-Centre is part of the Department of Sociological Studies, Institute of Education, University of London. EPPI-Centre was established in 1993 to implement a systematic approach to the organization and analysis of social interventions based on evidence. Activities and publications of the Centre are aimed at attracting decision-makers, practitioners and consumers of health services and education to discuss ways to improve the relevance of research and applications of their results. EPPI-Centre - the official partner of the Campbell Collaboration (Kempbellovskoe cooperation), Al in 1996 is involved in shared leadership Cochran-ray group to promote healthy lifestyles and public health (Cochrane Health Promotion & Public Health Field).      Campbell Collaboration (C2)URL: http://www. campbellcollaboration. ore /The Campbell Collaboration- an international nonprofit organization which aims to help people in making informed decisions regarding interventions in the social, behavioral and educational fields. C2 tasks - training, maintenance and dissemination of systematic reviews of studies on various interventions. Based on research C2 prepares a summary and produces electronic brochures with surveys and reports for policy-makers, practitioners, researchers and the public. The website provides free access to two databases: C2 SPECTR (register of social, psychological, educational and criminological research) and C2 RIPE (Register evaluation of interventions and policies). C2-RIPE database contains reviews and related documents of interest to researchers, decision makers, practitioners and the public.To find information on evidence-based medicine, there are special search tools:        SUMSearchURL: http://sumsearch. uthscsa. edu /Metasearch service, which is seeking the following resources:• textbooks and manuals; the default management Merck Manual;• MEDLINE - find reviews and editorials from the general high quality journals that provide full-text access;• National Guideline Clearinghouse - practical guidance on the Agency's research and the quality of health care dollars;• an abstract database of reviews on the effectiveness of medical interventions;• MEDLINE - Search for original research.To search the SUMSearch can use the following filters: Intervention, Prognosis, Diagnosis, Etiology / causation, Physical findings, Screening / Prevention, Adverse treatment effects TRIPURL: http://www. tripdatabase. com /TRIP provides a database search to 75 sites containing medical information of high quality. TRIP provides direct hypertext access to the largest collection of "scientific" material on the Internet, as well as to online articles in leading journals suc

Table 2. Basic information on sources of health information




Journals secondary information

АСРJournal Club

Summaries of articles with commentsfrom experts.

http://www. acpjc. org/shared/ menu_about. html

Evidence-Based Medicine

Abstracts of papers.


Evidence-Based Mental Health

Original and review articles on Psychiatry.


Evidence-Based Nursing

Critical summaries of articles with comments nurse practitioners.


International Journal of Medical Practice

Structured abstracts of key, important for practitioners of clinical and epidemiological studies with commentary. mjmp/mjmp-mn.htm

BMJ Updates

Dissemination of information on new publications of interest to the specialty.Access to summaries of articles from 2002.


На латинице


Словарь -Открыть словарную статью

  1. существительное
    1. content
    2. contents
    3. maintenance
    4. substance
    5. keep
    6. keeping
    7. upkeep
    8. scope
    9. subject
    10. matter
    11. allowance
    12. tenor
    13. subject-matter
    14. alimony
    15. purport
    16. meat
    17. aliment
    18. alimentation
    19. effect
    20. bullion value
    21. estovers
    22. sound
    23. supply
    24. gold content of the ruble index. asp?choice=13

Theof primaryinformation


12000000 articles published since1966


10000000 articles published since 1974

Thedatabaseof secondary information

The Cochrane Library

Contains a database of systematic reviews, the effectiveness of medical interventions,controlled trials register of methodology reviews

http://www. theCochraneLibrary. com


Official educational program with eight leading medical associations of the USA


Clinical Evidence

Material for the development of clinical guidelines


International cooperation

BEME CollaborationBest Evidence Medical Education

Systematic reviews in medical education


Evidence for Policy and Practice Information and Coordinating CentreEPPI-Centre

Sociological studies, aspx

Campbell Collaboration (C2)

Systematic reviews of research on various meshatelstvam

Metasearch Service


Search the Merck Manual MEDLINE practical guides and abstract database of reviews on the effectiveness of medical interventions

http://sumsearch. uthscsa. edu/


Search from 75 sites with medical information

Basics of Search evidential information in databases

(for example, searching MEDLINE on the site


Read phonetically

For a successful search for necessary information on evidence-based medicine have great value selection of relevant databases and the application of optimal methodology. Even with a thorough search is not always possible to find information about clinical trials (for example, because of poor indexing or unwillingness sponsors to publish negative results of clinical trials). For this reason, when searching for evidence-informed computer selection to complement other methods of search: "manual" search; study of reference lists of articles found; demands of researchers and drug manufacturers.

Boolean logic

To perform a search in electronic databases is necessary to use Boolean logic operators (George Boole - English mathematician, 1815-1864): AND, OR, NOT.

AND operator means that they will find articles containing both terms. For example, «laparoscopic AND cholecystectomy».

The OR operator is used when it is necessary to select articles that contain both a and the second is logically related terms. For example, you need to find relevant articles on the use of cholecystectomy and cholecystostomy, that is, all articles that contain either the first or second term. The query will look as follows: «cholecystecomy OR cholecystostomy».

The NOT operator is used when you want to delete unwanted themes. For example, «laparoscopic AND cholecystectomy NOT laparotomy».

Database fields

Search in databases tdannyh can be done in a particular field. Often to find just enough to make the term and provide a search box with the appropriate suffix: year of publication, place of publication, search by author, etc. To indicate the fields used suffixes that are used in constructing search queries.суффиксымогутразличатьсявразличныхпоисковыхсистемах. Thus, the site PubMed suffixes should be given in square brackets.Обычно  на сайте, на котором предоставляется MEDLINE-ссылка указана на полный перечень полей.

Golf writers - [au]. For example, you need all the articles of the author David S. A. The query will look like this: David SA [au].

Field [so]. Restricting the search area at the place of publication. For example, if you want to find all articles about laparoscopic cholecystectomy in the journal Clinical Medicine, you must create the following query: laparoscopic cholecystectomy [mh] AND (Klin-MedMosk) [so]. Where so - the suffix field short name of the magazine.

Field [sb] (subset) - filters subsets of journals, subject areas, types of publications. The three main types of logs: AIM - the subject area of approximately 120 leading magazines of the world, listed in Abridged Index Medicus; Nursing - Region bibliographies International Nursing Index; Dental - Index is Dental Litherature.

Subject areas are presented in Table 3.

Table 3. Subject areas in MEDLINE, for which integratedSpecial search filters

Field [ti] (title) and [jn] (magazine). If you know the title of the article (or an approximate title), and possibly a journal in which it was published, you can use the keys to search by name or journal.

Other fields of MEDLINE database are presented in Table. 4.

Таблица 4. Основные поля БДMEDLINE








Authors' Addresses

St. petersburg[ad]


Signs of the essay (it contains the string ab, if the records have abstracts) never displayed



Publisher ID number. There are two options - Pll (controlled publisher identifier) and DOI (Digital Object Identifier)



Original author






Full name of author

wong Julia s[fau]


Last author. Project managers mentioned at the end of the list.



Date of publication in format: YYYY/MM/DD



Date of receipt of the publication in PubMed



Number of contract or grant



Number ISSN



Journal title



Publication language. Specify first three letters of the name of language. The exception (Japanese - jpn [la]) Contains phrase NON ENGLISH if publication language differs from English except the name of language



Language. Indicate the first three letters of the language. Exclusion (Japanese - jpn [la]) contains not only names the language phrase NON ENGLISH, if the language publication differs from the English.



Date of assignment indices MeSH



Another index term, if the article is indexed in the system differs from the MeSH

demographic factors[ot]


Type of publication, for example, journal article



CAS registry number or the EU



Filters selected areas



Subheadings MeSH, their use is described more low.


dh [sh] = diet therapy [sh]


Secondary source of the data — the link to one or several international databases, for example, HDB — Hybridoma Data Bank






Additional features articles, for example, Case Report

case report[tg]



common cold[ti]


The word in the title and / or abstract

common cold[tiab]


The original title (if it contains an English translation). Title of Russian articles in translation.



The word in the title or abstract.



Update code PMID. If it is known, this is the fastest way to find the link.

10703801 [pmid]

Use the subject heading indexing system

mesh when searching for articles in the database medline.

MeSH — medical subject heading — list of terms (keywords), by which articles are indexed in a database MEDLINE. MeSH-term selection can be done in two stages. In the first stage empirically find the term, and then see the results in those articles which are suitable, choose keywords and use them in future searches. In the second stage on the site PubMed, you can select the link at the base of keywords and MeSH to view a hierarchical arrangement of the term and its subcategories.

In order to use key words (MeSH), you must create a query on type "keyword from the list of MeSH». For example, if you are looking for information about laparoscopic cholecystectomy, the key word is - laparoscopic cholecystectomy, and the request will look as follows: laparoscopic cholecystectomy [mh].

Symbol «*»в MEDLINE (PubMed)

Symbol «*» used in two ways.

Using suffixes subheading MeSH

One way to refine your results - use of a subheading of MeSH. Subcategories - add-on system of indexing in MEDLINE. With the help of subheading articles in MEDLINE, depending on the MeSH-term can be classified on the etiology, prevention, treatment, side effects, etc. The most useful subheadings are listed in Table. 5.

Table 5. The most frequently used subheadings MeSH

Suffix Example




side effects












drug therapy













cerebral palsy/nu


organization / management

health service/og


prevention and control









therapeutic use of (drug)


Types of questions and search strategies

Efficient search in MEDLINE search involves a combination of medical subject headings (keywords) to search on words in the text (the words contained in the title and abstract of the article) - the only way to select all the publications on the topic.

The following simple query suggestions in conjunction with the keywords (eg MeSH-terms) can be used for these types of epidemiological studies in the field of study:

Below are a more effective search strategy in PubMed MEDLINE, depending on the type of question. To find articles on therapy:

• Enter the name of the disease (a term MeSH);

• Select the subheading / th (therapy), / dt (drug therapy), / pc (prevention and control);

• Enter the name of the action (name of drug) (the term MeSH);

• choose a subheading, and / tu (therapeutic use);

• combine the results of these two searches with AND. You can also attach (AND) to the results section MeSH.

Diagnostic studies

The fastest and most effective way to find reliable articles on the diagnosis lies in the fact that the base MeSH:

introduce a diagnosis and use the appropriate term MeSH;

Enter the name of the diagnostic test;

combine the results of the first and second list, AND;

then combine them with MeSH-heading Sensitivity and Specificity. (AND).

Prognostic studies

Effective search engine prediction research - appropriate heading Prognosis in MeSH.


The best study design to answer questions about the forecasts - a cohort study. The best way to search for such research - the use of rubrics MeSH Cohort Studies.

Research on risk factors

The most closely related to the research questions of harm, and etiology of MeSH-heading Risk. In these studies almost always use the method of cohort studies. This means that the best simple search strategy is to conduct a search on your topic by headings and Risk Cohort Study, and then combine the results using the operator OR.

Clinical Queries

Clinical queries - filters methodology of epidemiological studies.

The site is located PubMed built-in filters, Clinical Queries (Figure 1) designed based on specific strategies and allows you to search studies on the etiology, treatment, diagnosis, prognosis, and clinical guidelines.

Figure 1. The system interface filters Clinical Queries on the PubMed website

In order to use a filter system, you must specify one or more terms of the operators of Boolean logic, and then choose a category, and noted an extended or a more narrow search.


Types of information can be classified according to different criteria (by source, target audience, the media). When choosing the direction of search for an answer to a clinical question must take into account many different factors. For example, in terms of reliability of the information, the most reliable source is pechtnaya medical literature, including published simultaneously on the websites of magazines. In recent years, many of the best magazines of their online version has become a major and more comprehensive than printed on paper. But, nevertheless, that the scientific medical journals are the most reliable source of evidence-based information. In medical journals, usually describes the results of recent studies and, as a rule, they do not contain basic training or procedural information.

"Grey" literature

The term "gray" literature refers to many varieties of informational materials that are not necessarily in the public domain and available public or private organizations for which the publication of literature is not the main activity. This can be reports, brochures, manuals, dissertations, information on products, memos, budget data, research materials and other documents are sometimes not intended for wide distribution.

Access to the "gray" literature through the usual channels of literature is not easily available, so that such literature is difficult to locate and obtain. Typically, this type of literature is not included in conventional commercial databases such as Medline. Despite the fact that such literature is published and is reviewed infrequently, it may contain valuable information. "Grey" literature - an example of a valuable type of information available on Web-sites.

Peer-reviewed information

The review process is a procedure for quality control and simultaneously assess the manuscripts for publication, grant applications for funding, review of clinical efficacy, etc. Review aims at obtaining an unbiased opinion of experts in the analysis of characteristics such as presence or absence of errors, the validity, accuracy , the organization of research, the scientific contribution. Between submission of manuscripts for publication and actual publication is some time needed to review. The review process takes place in different editions in various ways, including those with varying degrees of severity. Such an objective expert analysis of information is essential to ensure its accuracy. If the information is to be used in a clinical setting, it is very important that it was refereed. The main source of peer-reviewed scientific information are the medical journals. They even specifically referred to as peer-reviewed (peer reviewed), to emphasize that the print selected only those that have endorsed fellow reviewers. Refereed literature opposes "yellow" literature, journals, published in the advertising money is usually distributed free of charge or at very low price. These logs usually contain information distorted in favor of the advertiser. Doctors - supporters of evidence-based practice is not recommended to read these magazines at all, because the separate true information from the advertising claims extremely difficult.

Differences between electronic information from the printed

Development of information technology a decisive impact on the speed and efficiency of information dissemination. Electronic publishing has many advantages over traditional paper publications. However, because they appeared relatively recently, they still have significant drawbacks.

The main drawback of most electronic publishing internet (not to be confused with sites of scientific journals) is due to the lack of a good review. The publishing process is no longer limited to large professional communities or major publishing houses, where there is a stable and rigorous system of peer review. Electronic publication can be made even by individuals, which leads to the spread of erroneous information. High efficiency of electronic publishing has increased the functionality of information and its available capacity. Information published in electronic form, can be interactive or multimedia that the printed form at all feasible. For documents on Web-sites, publishers like the fact that they can immediately correct mistakes, rather than wait until the next issue, to put it errata, as is the case in print. Another major advantage of information in electronic form - that it is easy to organize a search in the database. This is true regardless of whether the information is combined with a special search engine on the CD or it is published in the World Wide Web, which in this case is a very large bibliographic database (if you master its existing search engines).

Standard components of the article:

■ title

■ list of authors

■ essay,

■ introduction

■ support research

■ methodology

■ Statistical analysis of data

■ the results of the study,

■ discussion

■ Literature.


Well the title contains a large part of the contents of the article. If the title you are interested, see below for a list of authors. Occasionally in this list you will see the name of someone you trust the author, and this will determine the focus for this article. Most often, the names are unknown, then go to the next, ie read the essay.

Summary prepares author of the article, it should be informative and structured, with clearly defined sections that reflect the contents of the article.

The abstract should contain a description of the study and a brief statement of the results. Do not try to extract more abstract. Usually this is impossible, even in those journals in which abstracts have a standardized structure and well written. In journals where abstracts are not placed (or put the bad), it should look at the end of the article - whether there are findings or conclusions? They can replace the abstract. Useful if they are interesting if they for you? If you - the doctor-patient level, and the paper is devoted to patients in a specialist hospital, it is hardly possible to extend the results of such an article to your practice.


The introduction is needed to familiarize the reader with the problem under study and the rationale for the study. It provides a description of the history of problems and formulation of research objectives. Statement of purpose should contain a complete answer to the question - "What should be studied, in which individuals, and by what method?".

If you decide to read further - go to the "Methods". In fact, the "Methods" include a number of aspects of the organization and conduct of research, and should not be directly studied in detail.

Rationale for research

This section is the main and at the same time most difficult to read. That it presents information that can identify the strengths and weaknesses of the study, the possibility of using its results.

It is well known that only experiment can show that science is true. More carefully than they have been delivered, the greater the likelihood that its results are based on real-life relationship between the phenomena and not an artifact and not a coincidence. "Weight" of each fact the more, the higher the scientific method of investigation, during which the fact is received.

Method of research

Research methods in medicine are the subject of the study of science of epidemiology. In the post-Soviet countries, the epidemiology is understood as the science of infectious diseases. The medical articles, he is often incomplete. Incomplete description of methods of investigation - a sufficient basis for a skeptical attitude to the article as a whole. Classical scientific research is constructed in such a way as to eliminate the influence of random factors.

Section "Methods" should contain only the information that was available during the writing of the plan or research protocol. All information received during the study, refers to the section "Results".

Should be described in detail how participants were selected observations or experiments (patients or laboratory animals, including control groups), including inclusion and exclusion criteria of the study and description of the study population. Due to the fact that the ratio of indicators such as age and sex to the goal of research is not always clear, authors should justify their use, if these figures are mentioned in the description of the study. For example, authors should explain why only the objects of a certain age were included in the study, or why women are excluded from it. The guiding principle should be clarity and clarity on how and why the study was conducted in a certain way. If the authors mention race or ethnicity, they should specify how to measure these indicators and to substantiate their significance.


Statistical methods should be described in detail so that the knowledgeable reader with access to the original data could verify the results presented. This section should present themselves as the quantitative data and the corresponding measurement error or uncertainty (such as confidence intervals). It should also be pointed out what software was used.


It is best to present results in the text, tables or figures in a logical sequence, starting with the main or most important information. Additional material and technical details can be presented in the annex, where neither are available, and at the same time, will not interrupt the text structure. In the summation of data in the Results section numerical results are given not only as derivatives (for example, percentages) but also in the form of absolute numbers, from which we calculated these derivatives, and identifies what statistical methods were used to analyze them.


Conclusions must be clearly linked to the objectives of the study and fully confirmed the data obtained.


Although references to review articles can be an effective way in the direction of readers to the basic literature, whenever possible, readers should be given links to primary sources describing the study. On the other hand, large lists of references to original work on the presented topics will occupy very much space in the printed version.

A critical evaluation of publications

For the modern doctor's skills in critical evaluation of the publications are important and necessary because annually publishes the results of hundreds of studies that alter the established standards of medicine. Often in order to answer the clinical problem can find a variety of sometimes conflicting findings and recommendations. Therefore, one of the most important principles of evidence-based medicine is associated with a critical analysis of information: the "weight" of each fact the more, the higher the scientific method of investigation, during which the fact is received.

Another no less important task for critical evaluation of the publication is to determine the extent to which its results can be extended to a specific clinical situation (the so-called "generalizability" study).

Often, doctors are inclined to interpret the results of the extended publication. Meanwhile, the person wrongfully transferred to the results obtained in experimental animals, and the survey results narrowly limited group of patients - the patients that differ by age or disease severity, or distribute data on the effectiveness of a particular drug to the entire drug group.

Assessment tools

There are various tools that can help you learn to evaluate the article. Program development skills critical assessment of CASP (Critical Appraisal Skills Programme) to prepare such assessment tools for systematic reviews of randomized controlled trials, qualitative research, cohort studies, research, case - control, studies of diagnostic tests and studies of economic evaluations (http://

Users' Guides to Evidence-Based Practice offers the following tools to assess articles on the treatment and prevention, diagnosis, harm and prognosis ( main.asp).

Questionnaires for critical evaluation of research on therapy, diagnosis, harm, forecasts and systematic reviews developed by the Canadian Centre Evidence-Based Medicine (

Below are 3 schemes for articles from the Users' Guides to Evidence-Based Practice and general algorithm for the analysis of articles created by a Canadian group of Evidence-Based Medicine (figure 2).


Bibliographic data:




Are the results of research important?

  1. Was the selection of patients for treatment at random?

[ ] Yes

[ ] No

[ ] Do not know

  1. Were all the patients involved in the experiment,

properly interviewed and examined at the end of the experiment?

- Was a complete follow-up?

- Whether the patients were examined in groups in which they were (the analysis depending on the purpose of treatment)?

[ ] Yes

[ ] No

[ ] Do not know

3.  Was whether to use "blind" with respect to patients

physicians and scientific staff?

[ ] Yes

[ ] No

[ ] Do not know

4.  Were the groups similar at the beginning of the experiment?

- The original balance of prognostic factors (demographics, comorbidity, disease severity, other known factors, interference)?

- If there were differences, whether they have been adjusted?

[ ] Yes

[ ] No

[ ] Do not know

5. Except for experimental procedures, whether in

otherwise treated in the same group?

- Concomitant intervention?

- Infection?

- Respect for patients and treatment regimens?

[ ] Yes

[ ] No

[ ] Do not know

6.In general - are the results of the study significant?

[ ] Yes

[ ] No

[ ] Do not know

What are the results obtained?

1. How effective was the treatment?

- The absolute risk reduction?

- Relative risk reduction?

2. How accurate was the assessment of treatment results?

Confidence intervals?

Will the results of my experiment in treating my patients?

1. Are these results in treating my patients?

- Are similar whether patients on demographics, disease severity, comorbidities and other prognostic factors?

- Were there objective reasons for excluding the possibility of applying these results

[ ] Yes

[ ] No

[ ] Do not know

2. Whether all clinically important outcomes were taken into account?

Are the replacement of end points?

[ ] Yes

[ ] No

[ ] Do not know

3.Overlap if the supposed benefits of such treatment possible harm and costs?

NNT for different outcomes?

[ ] Yes

[ ] No

[ ] Do not know


Bibliographic details:




Are the results of research important?

1. Has there been an independent, "blind" comparison with the reference standard?

- Is the chosen reference standard?

- Has applied reference standard and test all patients?

[ ] Yes

[ ] No

[ ] Do not know

2.  Consistent whether the band structure of the patient real patient population for whom this diagnostic test should be used in clinical conditions?

[ ] Yes

[ ] No

[ ] Do not know

3.   Did the results of the test evaluated the decision to implement the control standard?

"Verification" or "bringing"?

[ ] Yes

[ ] No

[ ] Do not know

4.  Is there sufficient detail the method of test so that it can be played?

- Preparation of the patient?

- Carrying out the test?

- Analysis and interpretation of results?

[ ] Yes

[ ] No

[ ] Do not know

5.In general - are the results of the study significant?

[ ] Yes

[ ] No

[ ] Do not know

6. What are the results obtained? How large or small likelihood ratio?

The value of the likelihood ratio

0.1 or 10

0,1-0,2 or 5-10

0,2-0,5 or 5.2

0,5-1, 1-2

The load on the probability of pre-and post-test

- Large or significant,

- Average,

- Small, but sometimes visible,

small, rarely conspicuous

[ ] Yes

[ ] No

[ ] Do not know

7. Can these results help me in treating my patients?

A) Will provide satisfactory reproducibility and interpretation of test results in terms of my work?

[ ]Yes

[ ]No

[ ]Donotknow

B) Are these results to my patient?

- Similar distribution of disease severity?

- A similar distribution of comorbidity?

- Objective reasons for which these results apply not?

[ ] Yes

[ ] No

[ ] Do not know

C)Affect whether the results of my work?

Test and clinical thresholds?

High or low values of the likelihood ratio?

[ ]Yes

[ ] No

[ ] Do not know

D)  Will the patient better as a result of the test?

-Is it dangerous to the underlying disease, if it remains undiagnosed?

- Does accept a risk with this test?

- Is there an effective treatment?

- Will the information obtained during testing, to changes in medical practice?

[ ] Yes

[ ] No

[ ] Do not know


Bibliographic data




Are the results of research important?

1. Can I assume that the group was representative and clearly defined and that the patients in it are on the same same stage of the disease?

- Inclusion and exclusion criteria?

- Bias of selection?

- Stage of disease?

[ ] Yes

[ ] No

[ ] Do not know

2. Was follow up sufficiently long and complete?

- Reasons for follow-up was incomplete?

- Have similar prognostic factors for patients past and have not undergone follow-up?

[ ] Yes

[ ]No

[ ]Donotknow

3.Were the evaluation criteria objective and unbiased?

- Outcomes defined at the beginning of the study?

- Prognostic factors were concealed from investigators?

[ ] Yes

[ ] No

[ ] Do not know

4.Has there been an adjustment for important prognostic factors??

[ ] Yes

[ ] No

[ ] Do not know

5.In general - are the results of the study


[ ]Yes

[ ]No

[ ]Donotknow

What are the results?

1. What is the probability of certain outcomes over time?

- Survival curves?

2. How accurate are estimates of the probability?

- Confidence intervals?

Will the results of my experiment in treating my patients?

1. Similar whether the patients studied on my patients?

- Are similar whether patients on demographics, disease severity, comorbidities and other prognostic factors?

- Identify the objective reasons for excluding the possibility of applying these results?

[ ] Yes

[ ] No

[ ] Do not know

2. Whether these results directly to the selection any treatment or denial of any treatment?

[ ] Yes

[ ] No

[ ] Do not know

3.  Whether these results are useful for the encouragement of patients or

issuance of recommendations?

[ ] Yes

[ ] No

[ ] Do not know

Naturally, the use of these questionnaires leaflets reading each article is not needed. They need only one - two times to remember the process of critical evaluation.In conclusion, it must be emphasized that the modern sources of information do not contain so many articles both interesting and quality content. So, start reading, you should make sure that the time is not spent in vain.

Figure 2. The general algorithm for estimating the article.



       Case studies    Task 1.     Structured clinical questions:     Task2

7. Undertake a critical evaluation of any article from PubMed?


Read phonetically

Похожие работы, которые могут быть Вам интерестны.

1. the Fourth and fifth stages of EM. Application of the received data in practice. Approximate planning and carrying out of clinical audit. The analysis of errors and planning of repeated audit

2. История формирования всемирной сети Internet. Статистика Internet

3. MS Access. Objects of a database. Ways of creation of the table in a database. Table editing. Acquaintance by types of data

4. Methodology of Foresight Evaluation

5. Judicial system of Great Britain

6. Evaluation of methodological quality of clinical studies

7. Thefeatures of planning and formation the system of balanced indexes at railway enterprises


9. Russian case. Anti-GMO movement and its network analysis

10. Approaches and Methods in Language Teaching a description and analysis