advertisement

Systematic reviews of adverse effects and other topics not – yet – covered by the Cochrane Collaboration

50 %
50 %
advertisement
Information about Systematic reviews of adverse effects and other topics not – yet –...

Published on April 29, 2008

Author: Cochrane.Collaboration

Source: slideshare.net

Description

Gerald Gartlehner speaking at the Symposium on the 10th Anniversary of the German Cochrane Centre
advertisement

Systematic Reviews of Adverse Effects and Other Topics Not Yet Covered by Cochrane Reviews Gerald Gartlehner Donau-Universität Krems, Department für Evidenzbasierte Medizin und Klinische Epidemiologie Symposium anlässlich des 10-jährigen Bestehens des Deutschen Cochrane Zentrums 2. April 2008

Objectives To further the discussion on two important issues currently left incomplete by many Cochrane reports: Adverse effects Applicability of findings

To further the discussion on two important issues currently left incomplete by many Cochrane reports:

Adverse effects

Applicability of findings

Benefits Harms A systematic review emphasizing only benefits is likely to lead to biased conclusions

Cochrane Reviews Underemphasize Adverse Effects Of 138 Cochrane reviews with data from at least 4000 participants, only 18% included data on clearly defined adverse effects (Papanikolaou et al. Am J Med 2004;117:582-589)

Of 138 Cochrane reviews with data from at least 4000 participants, only 18% included data on clearly defined adverse effects

(Papanikolaou et al. Am J Med 2004;117:582-589)

Most Cochrane reviews focus on data from RCTs 213 out of 5053 records mention “observational” OR “non-randomized” in the abstract

Most Cochrane reviews focus on data from RCTs

213 out of 5053 records mention “observational” OR “non-randomized” in the abstract

RCTs and Adverse Effects – Major Issues 1. Assessment and reporting of harms is often inadequate 2. Methodological issues limit the detection of certain adverse effects 3. Publication bias

1. Assessment and reporting of harms is often inadequate

2. Methodological issues limit the detection of certain adverse effects

3. Publication bias

Inadequate Assessment and Reporting Review of 7 interventions (Ioannidis et al. JAMA 2001;285:437-443) 61% had inadequate reporting of adverse effects and 71% of laboratory-determined toxicity Space devoted to safety was the same or less than the space given for the names of authors and their affiliations Reporting may be worse for non-pharmacologic vs. pharmacologic interventions (Ethgen et al. Ann Intern Med 2005;143:20-25)

Review of 7 interventions (Ioannidis et al. JAMA 2001;285:437-443)

61% had inadequate reporting of adverse effects and 71% of laboratory-determined toxicity

Space devoted to safety was the same or less than the space given for the names of authors and their affiliations

Reporting may be worse for non-pharmacologic vs. pharmacologic interventions

(Ethgen et al. Ann Intern Med 2005;143:20-25)

Example: Second-generation Antidepressants Sexual adverse events are common but frequently underreported Assessment in RCTs varied greatly Most studies did not use targeted questions Gartlehner et al. Comparative Effectiveness Review No 7: Agency for Healthcare Research and Quality. January 2007

Sexual adverse events are common but frequently underreported

Assessment in RCTs varied greatly

Most studies did not use targeted questions

Reporting of Sexual Adverse Events in RCTs of Antidepressants

Mean Incidence of Sexual Adverse Events in RCTs

Mean Incidence of Sexual Adverse Events in Observational Studies

Methodological Limitations of RCTs to Assess Harms Sample size Limited duration Lack of applicability (homogeneous, healthy populations)

Sample size

Limited duration

Lack of applicability (homogeneous, healthy populations)

Methodological limitations of RCTs to assess adverse effects RCTs are a good primary source to assess harms, if adverse effects : are distinctive occur frequently have a close temporal association with the intervention RCTs provide an incomplete picture of the risk of harms if adverse effects are: infrequent have a long latency or affect only subgroups of patients

RCTs are a good primary source to assess harms, if adverse effects :

are distinctive

occur frequently

have a close temporal association with the intervention

RCTs provide an incomplete picture of the risk of harms if adverse effects are:

infrequent

have a long latency

or affect only subgroups of patients

Publication Bias: Antidepressants in Children Systematic review of published RCTs Favorable risk benefit profiles for the treatment of depressive disorders in children Inclusion of unpublished trials reversed conclusions for most antidepressants (Whittington et al. Lancet 2004; 363: 1341-45)

Systematic review of published RCTs

Favorable risk benefit profiles for the treatment of depressive disorders in children

Inclusion of unpublished trials reversed conclusions for most antidepressants

How can the limitations be overcome? Observational studies may provide useful supplemental information for assessing adverse effects Often larger than RCTs Cover longer time periods Examine less selected populations

Observational studies may provide useful supplemental information for assessing adverse effects

Often larger than RCTs

Cover longer time periods

Examine less selected populations

“ Observational studies” refers to a broad range of study designs Controlled observational studies (often derived from large databases) Cohort studies Case-control studies Hypothesis-generating observational studies Case series, case reports Data-mining

Controlled observational studies (often derived from large databases)

Cohort studies

Case-control studies

Hypothesis-generating observational studies

Case series, case reports

Data-mining

Case reports 30% of primary published literature on adverse drug events is in the form of case reports (Aronson et al.Fundam Clin Pharacol 2004;57:616-21) Suicidality and SSRIs Reye syndrome and aspirin

30% of primary published literature on adverse drug events is in the form of case reports (Aronson et al.Fundam Clin Pharacol 2004;57:616-21)

Suicidality and SSRIs

Reye syndrome and aspirin

Observational Studies and Confounding Systematic error (unmeasured differences between groups) is always possible Adverse effects are often unexpected and unpredictable Confounding by indication is often not relevant

Systematic error (unmeasured differences between groups) is always possible

Adverse effects are often unexpected and unpredictable

Confounding by indication is often not relevant

Asymptomatic Carotid Artery Surgery Trial (ACAS) From Rothwell, Lancet 2005;365;82-93

From Rothwell, Lancet 2005;365;82-93

Applicability - What Decision Makers Need To Know Can it work? Will it work? In this patient? In this setting? Is it worth it? Do benefits outweigh harms/costs/inconvenience? Does it offer important advantages over existing alternatives? ( adapted from Brian Haynes ACP Journal Club)

Can it work?

Will it work?

In this patient?

In this setting?

Is it worth it?

Do benefits outweigh harms/costs/inconvenience?

Does it offer important advantages over existing alternatives?

( adapted from Brian Haynes

ACP Journal Club)

APPLICABILITY ( external validity, generalizability) Can I apply results to my population of interest? Can it work (under ideal circumstances)? Will it work (in real life)?

How can we facilitate the assessment of applicability? Better reporting of factors that may affect applicability Population characteristics Intervention and comparisons Intended and Unintended Outcomes Adverse effects Differences between trial protocol and routine practice (Rothwell, Lancet 2005;365;82-93) Standardized approach to distinguish efficacy from pragmatic studies

Better reporting of factors that may affect applicability

Population characteristics

Intervention and comparisons

Intended and Unintended Outcomes

Adverse effects

Differences between trial protocol and routine practice

(Rothwell, Lancet 2005;365;82-93)

Standardized approach to distinguish efficacy from pragmatic studies

Criteria to distinguish efficacy from pragmatic trials Population in primary care Less stringent eligibility criteria Health outcomes Clinically relevant study duration and treatment modalities Assessment of adverse events Adequate power to detect a minimally important difference from a patient perspective ITT-analysis (Gartlehner et al. A simple and valid tool distinguished efficacy from effectiveness studies. J Clinical Epidemiology 2006; 56:1040-1048)

Population in primary care

Less stringent eligibility criteria

Health outcomes

Clinically relevant study duration and treatment modalities

Assessment of adverse events

Adequate power to detect a minimally important difference from a patient perspective

ITT-analysis

Summary Balanced systematic reviews should assess both benefits and harms Broader range of data sources may be required More emphasis on the applicability of findings

Balanced systematic reviews should assess both benefits and harms

Broader range of data sources may be required

More emphasis on the applicability of findings

Add a comment

Related pages

Cochrane Reviews | Cochrane Community (beta)

Cochrane Reviews are systematic reviews of ... “To ensure that the work of The Cochrane Collaboration is relevant to ... but are often not systematic.
Read more

Cochrane Handbook for Systematic Reviews of Interventions ...

The Cochrane Handbook for Systematic Reviews of Interventions is the official document that describes in detail the process of preparing and maintaining ...
Read more

Assessment of adverse effects and applicability – two ...

... two areas not (yet) covered adequately in Cochrane ... a review of Cochrane systematic reviews found ... the risk for adverse effects if systematic ...
Read more

Homeopathic medicines for adverse effects of cancer ...

... homeopathic medicines for other adverse ... Homeopathic medicines for adverse effects of cancer treatments. Cochrane Database of Systematic Reviews ...
Read more

Echinacea for preventing and treating the common cold ...

... or reporting adverse effects did not differ ... for preventing and treating the common cold. Cochrane Database of Systematic Reviews 2014 ...
Read more

Symposium anlässlich des 10-jährigen Bestehens des ...

... Kleijnen Systematic Reviews, York, ... "Systematic reviews of adverse effects and other topics not – yet – covered by the Cochrane Collaboration”
Read more

Finding systematic reviews at PubMed Health

Finding systematic reviews at PubMed Health. ... of reviews from the Cochrane Collaboration: ... reviews on topics other than the effects of ...
Read more

Cochrane Health Promotion and Public Health Field www ...

Cochrane Collaboration. ... “Guidelines for Systematic Reviews of Health Promotion ... practitioners or other groups have not always found systematic ...
Read more