• REQUIREMENTS FOR ARTICLE SUBMISSION
    		•
  • INFORMATION FOR AUTHORS
    		•
  • PEER-REVIEW SYSTEM
    		•
  • EDITORIAL BOARD
    		•
  • CONTACT
    		•
    < back

    CONSORT checklist and flow diagram for clinical essays

    CONSORT (Consolidated Standards of Reporting Trials) is a protocol developed by researchers and editors for the execution and reporting of clinical essays. This protocol aims to lend transparency and quality to published research.
    CONSORT developed a flow diagram and checklist to guide researchers. A simplified version of these follows below. More information can be obtained at www.consort-statement.org.
    The flow diagram below shows the four stages of a controlled randomized clinical essay: recruitment, allocation, follow up and analysis. The author should use it as a guide and aid during the execution of an essay and the production of its report, and may include it in the article to be published.

    The checklist below is an aid to be used in verifying whether your article clearly exhibits all of the relevant information.

    Verify whether your article complies with all of the items laid out below:

    Session

    Item

    Description contained

    Page number

    Title and Abstract

    1

    - How members of the sample were allocated to the interventions (e.g.: “Randomized study”, “Study with random allocation”).

     

    Introduction

    2

    - Scientific background, previous publications.

     

    Method
    Participants

    3

    - Sample selection criteria.
    - Locations and context where the data were collected.

     

    Interventions

    4

    - Precise details of the interventions applied to each group.
    - How and where these interventions were performed.

     

    Objectives

    5

    - Hypothesis and specific objective(s).

     

    Data obtained

    6

    - Clearly defined primary and secondary measuring methods.
    - When applicable, methods used to improve measurement quality (e.g.: multiple observations, training of the people responsible for measuring,...).

     

    Sample size

    7

    - How the sample size was defined.
    - When applicable, explanation of the comparisons among the groups and rules for conclusion of the study (stopping rules or stopping criteria).

     

    Randomization --
    Generation of the sequence

    8

    Methods used to generate the random allocation sequence, including the details of any restrictions (e.g.: block randomization, stratification).

     

    Randomization --
    Secrecy in allocation

    9

    - Method used to implement the random allocation (e.g.: numbered containers, central telephone), making it clear whether the sequence was kept secret until the interventions had been designated.

     

    Randomization --
    Implementation

    10

    - Who generated the allocation sequence, who recruited the participants, and who allocated them to the groups.

     

    Blindness (masking)

    11

    - Whether the participants, collaborators and researchers were kept blind to who was allocated to each group. If they were, how the masking’s success was measured.

     

    Statistical methods

    12

    - Statistical methods used to compare the primary results obtained from the groups.
    - Additional methods of analysis, such as adjusted and subgroup analyses.

     

    Results

    Participant flow

    13

    - Flow of participants at each stage (a flowchart is recommended). The flow must show, for each group, the number of allocated participants who received treatment, who completed the protocol and who were analyzed for the obtainment of primary results.
    - Describe any deviations from the original study plan, as well as what caused them.

     

    Recruitment

    14

    Dates of the recruitment and tracking periods.

     

    Initial data

    15

    - Initial basic and demographic data on each group, as well as their characteristics.

     

    Analyzed numbers

    16

    - Number of participants in each group (denominator) included in each analysis.
    - Inform whether the analysis is of the intention-to-treat type.
    - If possible, present the results in absolute numbers (e.g.: 10/20, instead of 50%)

     

    Outcomes and estimation

    17

    - For each obtained primary and secondary data group, present a summary of the results, as well as an estimate of precision and effect (e.g.: 95% confidence interval).

     

    Ancillary analyses

    18

    - Assess multiplicity by reporting any other analyses that were performed, including adjusted and subgroup analyses, indicating which were pre-specified and which were exploratory.

     

    Adverse events

    19

    All adverse or collateral events observed in each group.

     

    DISCUSSION
    Interpretation

    20

    Interpretation of the results, considering the hypothesis, potential sources of biasing or imprecision, and the risks associated with multiplicity of analyses and results.

     

    Generalizability

    21

    Generalizability (external validity) of the discoveries made during the study.

     

    Evidence in general

    22

    General interpretation of the results in the context of current evidence.

     

     

    Definition of some terms:

    • Adjusted analysis: This normally refers to an attempt to adjust differences in the groups’ basic characteristics, using specific data analysis techniques.
    • Intervention groups: Groups to which the sample participants are allocated, so that each group receives a different treatment.
    • Control group: This is usually the group that will not undergo any intervention, or will receive a placebo.
    • Allocation sequence: Sequence in which the intervention groups will be arranged in order to receive the allocated participants, or the sequence in which the participants will be arranged for later allocation. Also known as an Allocation or randomization calendar.
    • Confidence interval: Measurement of the precision of an estimated value.
    • Deterministic allocation method: An allocation method that follows pre-determined, not entirely random rules to allocate participants to the groups (e.g.: date of birth, alternation). This type of method is predictable; therefore, it may allow initial participants to be manipulated aiming to produce a bias in selection and results.
    • Generation of the allocation sequence: The procedure used to create the allocation sequence, such as a table or a random number generator. Some methods may be used to generate and apply the sequence, and are described in Randomization, below.
    • Randomization, random allocation: In a randomized essay, each participant has an equal chance of being allocated to each group.
    • Restricted randomization:  Random allocations that follow group-balancing rules. Its two most common forms are:
      • Block or permuted-block randomization: A technique for the obtainment of equally sized or proportional groups. Considering two equal groups, a block of definite and constant size (e.g., 10) of sample participants is set apart. Random allocation is then performed until one of the groups receives half of the block’s members (5). The remaining members are, then, allocated to the other group. We then proceed to the next block of ten, allocating them in the same manner, until the end of the sample.

    This can also be used for groups of differing sizes. For example, if we want one group with twice as many participants as the other (1:2 ratio), we take a block of 15 and allocate them randomly until the smaller group receives 5 members of that block, or until the larger group receives 10, whichever happens first.
          The sample size must be dividable by the size of the block, and the size of the block must be dividable by the sum of the ratio between/among the groups (if the ratio is 1:1, block size must be dividable by 2; if the ratio is 1:2, it must be dividable by 3, if the ratio is 1:1:2 – three groups -, it must be dividable by 4, and so forth).

      • Stratification or stratified randomization: The sample is initially separated into pre-groups (stratified), each with a specific characteristic (stratification can be by gender, age, weight, etc.). Then, equal parts of each pre-group are randomly allocated to the intervention groups. The goal is to obtain similar groups in relation to the stratification criterion. This is especially indicated for small samples.
    • Intention-to-treat analysis:  A data analysis strategy that considers the group as a whole as having been submitted to the initially intended treatment, regardless of some members having been excluded during the study or having received different treatments. The goal is to avoid biased or manipulated results due to participant desistance or exclusion.
    • Multiplicity: Proliferation of many possible comparisons within a study. This generally occurs when more than one measure or measuring technique is used, if measuring is done at different moments after the intervention, etc…
    • Application or performance bias: Differences in the results caused by differences in the care taken while applying the interventions to the groups.
    • Stopping criterion: A pre-defined criterion used to determine the end of a study, which may occur for several reasons: sufficient data have already been collected, the participants are being placed at risk, expected impossibility of achieving the study goals or inconclusiveness of the study, etc...