DOC

Ideally, the best available evidence is systematically obtained

By Troy White,2014-06-20 12:25
12 views 0
Ideally, the best available evidence is systematically obtained

    Delfini Group ?, LLC D

    Evidence- & Value-based Solutions For Health Care

    Clinical Improvement Consults, Content, Seminars, Training & Tools

    Ideally, the best available valid and useful evidence is systematically obtained and evaluated through doing a high quality systematic review. Most groups, however, lack the time and resources to do that demanding work, and yet need some way to summarize evidence and apply it. If you have reviewed a number of studies on the same topic and our advice is to do this by doing a systematic search it is highly recommended that you synthesize

    your findings to account for the weight of the evidence and evidence usability. The Delfini Evidence Synthesis

    Tool can make this easier for you.

    This synthesis tool, used in conjunction with the Delfini Evidence Tool Set, can provide you with the key

    essentials for doing critical appraisal work, examining results of valid studies and synthesizing the evidence. Doing a high quality systematic review is highly preferable but if you are unable to do this work, this tool can

    provide you with some assistance in using what evidence you have found.

    We strongly recommend that you review first characteristics of quality systematic reviews using the Delfini

    Systematic Review Tool or another similar resource and attempt to apply as many of those concepts as you can. Also review of the Delfini Evidence Grading Tool can be a helpful first step.

    This tool cannot and is not meant to provide all the information you need to evaluate the usefulness of study information. We provide recommendations only. Ultimately you will need to apply your own judgment in

    determining whether a study should be included in your evidence synthesis. This tool can give you key ingredients to help you make that decision and assist you in creating an evidence synthesis.

    Instructions for Electronic Tool Use are at www.delfini.org > Main Menu > Resource Center

    elfini Group ?, LLC DD

    Template ? Delfini Group, LLC, 2002-2005. www.delfini.org All Rights Reserved World Wide

    Delfini Group Evidence Synthesis & Clinical Recommendations Tool

    Goal: Present your best attempt to summarize the weight of the evidence into a conclusion about usability of the study information you have reviewed and how you

    recommend that information be applied. You wish to combine scientific strength AND usefulness of the information in order to make decisions about health care. You will

     need to apply your own judgment.

    Contents

    A. Introduction to Evidence Synthesis Steps

    B. Executive Summary & Format Suggestions

    C. Examples for Suggestions on How to Create an Evidence Synthesis D. Delfini Evidence Synthesis Table & Clinical Recommendation Table

    Before You Start….

    It may be useful, before you begin your synthesis work, to consider ideal study parameters for the

    best available evidence so that you have an effective means of comparison between optimal and the

    research you find. This information can be found in the Delfini Evidence Grading Tool.

    www.delfini.org Template ? Delfini Group, LLC, 2002-2005. All Rights Reserved World Wide. Page 2 of 13

    Delfini Group Evidence Synthesis & Clinical Recommendations Tool

A. Introduction to Evidence Synthesis Steps

    Systematic Review 1. We strongly recommend that you first acquaint yourself with characteristics of quality Attempt systematic reviews using the Delfini Systematic Review Tool or another similar guide. Do

    as many of these steps in the best possible way that you can.

    Goals & Ideals 2. Decide upon the goal of your intended clinical recommendation and anticipate parameters

    of the ideal research methods to answer this question for comparison purposes. (For ideas,

    see Ideal Research Study Parameters in the Delfini Evidence Grading Tool.)

    Ask & Acquire 3. Frame your clinical question and conduct a systematic search of the medical literature to

    identify the best possible studies.

     Document your searching and your filtering strategy.

    Selection Criteria 4. Determine inclusion/exclusion criteria for studies you plan to use in your synthesis.

    Appraise for Validity & 5. Critically appraise selected studies for both validity and usefulness of results. The goal is to Results include only those studies that represent the best available valid, usable scientific evidence.

    There are likely to be few.

     We strongly recommend the use of tools. The Delfini Study Validity & Usability Tool

    addresses both validity and results assessment in one tool.

     Grade each study. See the Delfini Evidence Grading Tool.

    Study Selection for Best 6. For studies passing both validity and results assessment, rank studies by grade of study,

    Available Valid and and select those you wish to include.

    useful Evidence The Delfini Validity and Usability Grading Scale in the Delfini Evidence Grading

    Tool is our suggested means for ranking studies and provides criteria for selecting

    studies for inclusion in an evidence synthesis, along with providing suggestions for

    grading the strength of your synthesis.

    Harms 7. Because harms may not have been sufficiently addressed in your studies of choice,

    determine if you need to do a new search specific to learning about any harms of what you

    wish to recommend. Modify your statement as needed if you have new information.

    Synthesize & 8. Summarize the best available valid and useful evidence. This may be a text statement or a Summarize table documenting characteristics of the evidence you have identified as being the best. (A

    table is included in this tool.) Again, you will have to apply judgment.

    Strength & Limitations 9. Grade the quality of the collective evidence.

    of Synthesis (Grading & 10. Document limitations of your assessment or why it differs from other syntheses using Documentation) criteria for systematic reviews.

    Clinical 11. Write a clinical recommendation based on your findings. Quantitate where possible. See

    Recommendation the Delfini Evidence Grading Tool for wording suggestions.

    Recommendation 12. Label the strength of your recommendation to make as transparent as possible and Grading & document the limitations of your assessment using criteria for systematic reviews. See the

    Documentation Delfini Evidence Grading Tool for help with grading and judgments.

    Analysis 13. Prepare a brief analysis of other considerations such as

     Efficacy versus effectiveness projection

     Applicability to which patient population and under which circumstances

     Patient perspective (benefits, harms, risks, costs, uncertainties, alternatives)

     Physician perspective (e.g., likelihood of acceptance and appropriate application)

     Patient acceptance (e.g., likelihood of patient acceptance and adherence)

     Actionability (e.g., FDA approved, affordable, do-able, implementable, fits

    circumstances of care, etc.)

    www.delfini.org Template ? Delfini Group, LLC, 2002-2005. All Rights Reserved World Wide. Page 3 of 13

Delfini Group Evidence Synthesis & Clinical Recommendations Tool

     B. Executive Summary & Format Suggestions

We recommend that a brief executive summary precede the details of your synthesis work (detailed

    instructions for which follow this table). Suggested contents are as follows:

    Categories Summary from Cochrane or Clinical Evidence

     Clinical Question

     Background Information

    General Information

     Number of Included Studies

     Number of Subjects

     Population Characteristics

     Homogeneity Information

    Efficacy Summary

     Overall Grade of your summary’s

    usefulness

     Concluding Statement

     “Insufficient Evidence” Summary

     Notes/Other

     Judgments

    Safety Summary

     Overall Grade of your summary’s

    usefulness

     Concluding Statement

     “Insufficient Evidence” Summary

     Notes/Other

     Judgments

    Summary Detail of Findings from Valid Studies

     List out a summary for each study

    included…

     Author

     Grade

     Brief critique

     Quantified results

     P-value

    www.delfini.org Template ? Delfini Group, LLC, 2002-2005. All Rights Reserved World Wide. Page 4 of 13

Delfini Group Evidence Synthesis & Clinical Recommendations Tool

    Categories Summary from Cochrane or Clinical Evidence Other

     Limitations of Your Work

     Notes/Other

Follow the Executive Summary with…

Supporting Documentation

     FDA information

     Evidence synthesis tables

     Search & filtering strategy (efficacy, harms, other)

     Selection criteria for studies

     Methods used to determine validity and usability

     Grading scheme

     Table of included studies

     Critical appraisals of included studies

     Table of excluded studies

     References

     Glossary

     Conflicts of interest

     Reviewers

     Preparers

     Date

    www.delfini.org Template ? Delfini Group, LLC, 2002-2005. All Rights Reserved World Wide. Page 5 of 13

Delfini Group Evidence Synthesis & Clinical Recommendations Tool

C. Examples for Suggestions for How to Do an Evidence Synthesis

    1. Imagine that you have conducted a literature search for a therapy question using Drug X for Condition Y.

    Your group is interested in adding another agent for women failing Drug Z. You have filtered your search to

    find the best available valid and useable evidence. You have selected three RCTs to review. They have some

    variations in their population.

    2. You have done a validity review and you have evaluated the results of those studies you have found to be

    valid. Your assessment of the outcomes and results for Drug X looks like this:

    Study 1. There is Grade A Evidence: Strong evidence of efficacy. It can be concluded that there is a

    reduction in mortality for non-pregnant women 45-60 years of age with condition Y, taking Drug X as

    compared to placebo within 5 years. The ARR for 5 years is 1.8%, 95% CI (1.25-2.5%). NNT for

    mortality reduction is estimated at NNT of 55, 95% CI (40-80)(5 years).

    Study 2. There is Grade A Evidence: Strong evidence of efficacy. It can be concluded that there is a

    reduction in mortality for non-pregnant women 50-65 years of age with condition Y taking Drug X as

    compared to placebo within 5 years. The ARR for 5 years is 2%, 95% CI (1-3%). NNT for mortality

    reduction is estimated at NNT of 50, 95% CI (33-100) (5 years).

    Study 3. Grade UVU Evidence: Uncertain Evidence (Uncertain Validity and Uncertain Usefulness [See

    Appendix for explanation] .There is uncertain evidence of efficacy (reduction in mortality) and

    uncertain validity and usefulness due to small sample size, several weaknesses in methodology which

    indicates that Drug X might be of benefit in non-pregnant women 30-44 years of age. The ARR for 5

    years is 5%, 95% CI (3-7). NNT for mortality reduction is estimated at 20, 95% CI (14-33) (5yrs).

3. You decide that you are going to summarize all three of the studies you will use the “uncertain study”

    because it is an RCT, there are few threats to validity and the results are going in a consistent direction with

    the other two good studies.

    4. Applying your own judgment, you rate the level of usability of combination of these three studies that you

    have elected to summarize. You rate the combination of the studies as Grade A, USEFUL, i.e., the weight of

    the evidence receives an A grade even though study 3 receives an individual grade of UVU. You make this

    determination because the UVU study is going in the same direction as the Grade A studies, so overall you

    deem the weight of the information useful.

    Grade of Usability Strength of Evidence

    The weight of the evidence appears sufficient to use in ? Grade A: making health care decisions. Useful ; Several well-designed and conducted studies that

    consistently show similar results

    Note: Because of the Grade UVU study, some groups might elect to give this a Grade B, which is a reasonable

    choice as well. Evidence grading involves a great deal of judgment.

    5. While assessing this information as useful, you realize that you need to apply your own judgment in

    determining how the information is to be used. Some groups might decide on a conservative approach due

    to the uncertainty of Study 1 and the likelihood of realizing results smaller than efficacy in the other studies.

    Such a group might decide to

    Recommend an interpretation of NNT=50-55 (5yrs) for ages 45-65, with a statement that there is uncertain

    evidence of benefit for women 30 44 and no evidence for women at other ages.

    Another group might decide to “keep it simple” and recommend an interpretation of NNT=50 (5yrs) for ages

    30-65. Judgment will vary between evaluators.

    www.delfini.org Template ? Delfini Group, LLC, 2002-2005. All Rights Reserved World Wide. Page 6 of 13

    Delfini Group Evidence Synthesis & Clinical Recommendations Tool

    6. You would then prepare your text summary or your evidence synthesis table (see below) and quantitative

    statements.

    7. From that you will write a clinical recommendation and document its limitations.

    www.delfini.org Template ? Delfini Group, LLC, 2002-2005. All Rights Reserved World Wide. Page 7 of 13

    Delfini Group Evidence Synthesis & Clinical Recommendations Tool

    Example of an Evidence Synthesis Table

    You may choose to use the following table instead of, or in addition to, an Evidence Synthesis Text Summary

    Statement see below for the text example.

    Evidence Synthesis Table

1. Intervention or Exposure Drug X 4 mg tid

    (include characteristics

    and parameters such as

    dosing)

    2. Grade of Evidence Grade A: Useful

    Synthesis

    3. Clinical significance: Mortality

    Is there direct evidence of

    benefit in the following

    areas?

    Morbidity, mortality,

    symptom relief,

    functioning, health

    related quality of life

    4. Population (including Women ages 30-65 (excluding pregnant women)

    disease spectrum, risk

    groups, etc. and

    documentation of

    exclusions)

    5. Indication With Condition Y

    6. Line of therapy Second lineyou have approved Drug Z

    7. Comparator Placebo

    8. Head-to-head or None

    equivalence trials

    (include grade and

    information about study

    quality)

    9. Based on study types RCTs

    10. Quality of studies: Grade A: Well-done, valid and useful RCTs for women, ages 45 to 65

     Grade Two studies

     Number of studies Total n = 15,432

     N

     Summary of validity Grade UVU: Uncertain validity and usability of RCT for women, ages 30-44

    and usefulness 1 study

     Other key details as Total n = 241

    applicable Uncertainty due to small sample size, 2% loss to follow-up and one

    patient crossed over to therapy

    11. Grade and Results Grade A: Women 45-65: NNT 50-55, 95% CI (33-100)(5 yrs) drug X vs

    (include NNT [CI] and placebo

    study time period) Grade UVU: Women 30-44: NNT 20, 95% CI (14-33) (5 yrs) drug X vs

    placebo

    12. Uncertainties Pregnant women since excluded from all studies.

    www.delfini.org Template ? Delfini Group, LLC, 2002-2005. All Rights Reserved World Wide. Page 8 of 13

    Delfini Group Evidence Synthesis & Clinical Recommendations Tool

     Women of childbearing age due to uncertainty re: fetal effects.

    13. Alternatives None for those failing Drug Z

    14. Cost Equivalent to Drug Z

    15. Review Limitations Unable to do a systematic review, but attempted as many steps as

    possible.

     Did not test for heterogeneity, but did qualitative analysis of homogeneity

    and did review CI’s, RRR and point estimates.

     Inclusion of Grade U evidence because of need for help for this

    population and no better evidence however, study is going in same

    direction as Grade A studies

    16. Harms Harms or potential harms may include skin rash. Harms data is not from

    RCTs. Harms data is from post-marketing surveillance studies (case

    reports). All studies excluded pregnant women and there are no reports of

    use specific to pregnant subpopulations.

    17. Usability statement (e.g., There is strong evidence of efficacy. It can be concluded that Drug X in

    “There is strong, comparison to placebo reduces mortality by approximately 1.8%- 2% 95%

    moderate or uncertain CI (1.25-2.5%) in women, aged 45 to 65, with Condition Y and who are not

    evidence to conclude pregnant. NNT for prevention of overall mortality is estimated to be 50-55

    efficacy.”) Considering 9%% CI (33-100) (5 yrs).

    addressing any

    exclusions, limiting There is uncertain evidence of efficacy in women, aged 30 to 44 and who are

    circumstances, results for not pregnant, as compared to placebo.

    the comparator and

    statements about risks or

    harms.

Clinical Recommendation & Other Considerations Based on Evidence Synthesis

18. Grade for Clinical Grade B: Possibly Useful

    Recommendation &

    Reason We feel the weight of the evidence is Grade A; however, our clinical

    recommendation includes advice for a younger population which is not

    strongly supported by this evidence. We apply a conservative approach for

    this reason.

    19. Recommendation for Women with Condition Y failing Drug Z and who are not pregnant

    population

    20. Under what circumstances See above

    21. Intervention (include Drug X 4 mg tid

    characteristics and

    parameters such as

    dosing)

    22. Projection of efficacy or NNT effectiveness for reduction of overall mortality is estimated to be

    effectiveness (consider approximately 50-55 (5 yrs)

    recommendation for

    downward adjustment of

    efficacy to effectiveness)

    23. Caution on Harms Patients should be advised of the possibility of rash.

    Drug should be used conservatively in pregnant women who should be

    advised of uncertain effects on fetuses due to lack of information.

    24. Actionability Yes

    www.delfini.org Template ? Delfini Group, LLC, 2002-2005. All Rights Reserved World Wide. Page 9 of 13

    Delfini Group Evidence Synthesis & Clinical Recommendations Tool

    25. Considerations from Should be well-received since there are no reasonable alternatives and good

    physician perspective safety profile.

    26. Considerations from Should be well-received since there are no reasonable alternatives for

    patient perspective patients failing Drug Z and good safety profile.

    27. Limitations of Clinical Uncertainty re: population 30-44. Chose conservative NNT for effectiveness

    Recommendation estimate. Results may be better in practice.

    28. Text Statement for Your For women with condition Y, ages 30-65 and who are not pregnant Clinical Recommendation failing therapy on Drug Z, consider use of Drug X as a second line therapy

    to reduce mortality NNT estimate =50-55 (5 yrs).

    This clinical recommendation is rated Grade B: Possibly Useful

www.delfini.org Template ? Delfini Group, LLC, 2002-2005. All Rights Reserved World Wide. Page 10 of 13

Report this document

For any questions or suggestions please email
cust-service@docsford.com