Transcript Bez tytułu slajdu

Reporting of harms in
randomized controlled
trials: what can be done?
Prof. Hanna Szajewska
The Medical University of Warsaw
Department of Paediatrics
[email protected]
Hierarchy of evidence
for questions about the effectiveness of an intervention
Systematic
review
/metaanalysis
RCT
Strong
Cohort studies
Case-control studies
Case series studies
Expert opinion without explicit critical
appraisal, theories bases on physiology or
plausibility, bench top research & animal
studies
Weak
Some figures
• An analysis of more than 1000 systematic
reviews/meta-analyses of health care
interventions
– 27% reviewed any harm data
– Only 4% primarily focused on safety
Ernst & Pittler. BMJ 2001;323:546.
Reporting harms may cause
more trouble and discredit than
the fame and glory with
successful reporting of benefits.
Cuervo LG & Clarke M. BMJ 2003
Outline
• Why it is important to consider safety
(harms) in randomized controlled
trials/systematic reviews?
• What can be done?
– CONSORT Statement
Why it is important to consider
safety in systematic reviews?
• Should probiotics be given routinely to preterm
infants to prevent NEC?
• Does DHA intake contributes to the visual
development of infants up to 12 months of age?
• Should protein hydrolysates be given for allergy
prevention to all infants regardless the risk of allergy?
• Are those interventions likely to be more beneficial
than harmful?
Why it is important to consider
safety in systematic reviews?
To make informed,
evidence based
decisions on a given
intervention
information on
benefits as well as
harms are needed
Benefits
Harms
The CONSORT
(Consolidated Standards of Reporting Trials)
Ioannidas et al. Ann Intern Med. 2004;141:781-88.
Terminology
Many terms are used interchangeably to describe
harms associated with healthcare interventions
Term
Definition
Safety
Reasonable certainty of no harm or, in some systems,
a reasonable balance between costs (harm) and
benefits
Adverse
effect
An adverse event for which the casual relation
between the drug/intervention and the event is at
least a reasonable possibility
Adverse
event
An adverse outcome that occurs during or after the
use of a drug or other intervention but is not
necessarily caused by it
Side effect
Any unintended effect, adverse or beneficial, of an
intervention
Ioannidas et al. Ann Intern Med. 2004;141:781-88.
Terminology
Harms instead of safety
• CONSORT group encourages to use the
term ‘harms‘ instead of ‘safety’
• Rationale
– ‘Safety’ is a reassuring term that may obscure the real
and potentially major ‘harms’ that interventions may
cause
Ioannidas et al. Ann Intern Med. 2004;141:781-88.
Terminology
Used interchangeably
Term
Definition
Harms
The totality of possible adverse consequences of an
intervention or therapy; they are direct opposite of
benefits, against which they must be compared
Safety
Reasonable certainty of no harm or, in some systems,
a reasonable balance between costs (harm) and
benefits
Adverse
effect
An adverse event for which the casual relation
between the drug/intervention and the event is at
least a reasonable possibility
Adverse
event
An adverse outcome that occurs during or after the
use of a drug or other intervention but is not
necessarily caused by it
Side effect
Any unintended effect, adverse or beneficial, of an
intervention
Title & Abstract
• Recommendation 1
• If the study collected data on harms
and benefits, the title or abstract
should so state.
– Example
• Maurer M et al. Efficacy and safety of omalizumab
in patients with chronic urticaria who exhibit IgE
against thyroperoxidase. J Allergy Clin Immunol.
2011 Jul;128(1):202-209.e5.
Title & Abstract
• Recommendation 1
• If the study collected data on harms
and benefits, the title or abstract
should so state.
– Cochrane Library, issue 3, 2011
– There are 686 results out of 651035 records
for: "harm in Title, Abstract or Keywords in
Cochrane Central Register of Controlled Trials”
Background
• Recommendation 2
– If the trial addresses both harms and
benefits, the introduction should so
state.
• What is known?
• What is not known?
• Why we did this study?
– RCTs that focus primarily on harms should clearly state this
interest when describing the study objectives in the Introduction
and in defining these objectives in the Methods.
Methods
Outcomes
• Recommendation 3
– List addressed adverse events with
definitions for each
• with attention, when relevant, to grading,
expected vs. unexpected events, reference to
standardized and validated definitions, and
description of new definitions
Methods
Outcomes
• Define the recorded AE (clinical and laboratory)
• Explain whether the reported AE ecompass all
the recorded AE or a selected sample
• How, why, and who selected AE for reporting
• In trials that do not mention harm-related data,
the Methods section should briefly explain the
reason for the omission
– eg. ‘the design did not include the collection of any
information on harms’
Methods
Outcomes
• Reporting of expected vs unexpected
AE
– Expectation may influence the incidence of
reported AE
– Making participants aware in the consent form
of the possibility of a specific adverse event
may increase the reporting rate of the event
Methods
Outcomes
• Standardized and validated vs.
nonvalidated scales
– Standardized
• For example, standardized case definitions of
adverse events following immunisation
Vaccine 2002;21:298-302
– Nonvalided
• Still common
Methods
Outcomes
• New definitions
– Should be explicit and clear
– Authors should describe how they developed
and validated new scales
NEJM 2006
J Pediatr 2010
Methods
• Recommendation 4
– Clarify how harms-related information
was collected
•
•
•
•
•
mode of data collection
timing
attribution methods
intensity of ascertainment,
harms-related monitoring and stopping rules, if
pertinent
Methods
• Mode of data collection
– questionnaires, interviews, tests
• Timing
– during intervention? after intervention
• attribution methods
– Whether or not an adverse event is related to the intervention
– State who makes the attribution (investigators, participants,
sponsors, or combinations)
• intensity of ascertainment,
• harms-related monitoring and stopping rules, if
pertinent
Statistical methods
• Recommendation 5
• Describe plans for presenting and
analyzing information on harms
– Coding
– Handling of recurrent events
– Specification of timing issues
– Handling of continuous measures
– Any statystical analyses
Statistical methods
• Using only descriptive statistics to report
harms
– Appropriate in most RCTs
– Most trials lack power to test harms-related hypotheses
and have no explicit prespecified harm-related
hypotheses
• Grouping of adverse events
– Describe each combination
– State whether the grouping was post hoc or a priori
• Recurrent events
– Specify whether this is count as separate events or as 1
event
Results
• Recommendation 6
– Describe for each arm the participant
withdrawals that are due to harms and
the experience with the allocated
treatment
Numbers analyzed
• Recommendation 7
– Provide the denominators for analyses
on harms
• i.e. which participants and what follow-up time
count toward total exposure to the allocated
treatment
Intention to treat analysis
is usually preferred
Rates of outcomes and ancillary
analyses for adverse events
• Recommendation 8
– Present the absolute risk for each
adverse event (specyfing type, grade,
and seriousness per arm) and present
appropriate metrics for recurrent events,
continuous variables and scale variables,
whenever pertinent
Example
Rates of outcomes and ancillary
analyses for adverse events
• Recommendation 9
– Describe any subgroup analyses and
exploratory analyses for harms
Discussion
• Recommendation 10
– Provide a balanced discussion of benefits
and harms with emphasis on study
limitations, generalizability, and other
sources of information on harms
Take home messages
• Reporting of harms-related data from
RCTs may be improved (CONSORT
document)
• Better reporting will help readers critically
appraise and interpret trial results.
A final comment…
(paraphrasing Winston Churchill on democracy)
RCTs are the worst form of evidence for
clinical practice, but for now there
appears to be no better alternative
Thank you for your
attention