- Email: [email protected]
The Need to Interpret the Results of a Clinical Trial in the Context of Other Evidence
Accepted Manuscript The need to interpret the results of a clinical trial in the context of other evidence Nicholas J. Wald, DSc(Med), Joan K. Morris, PhD PII:
S0002-9343(16)30858-0
DOI:
10.1016/j.amjmed.2016.08.015
Reference:
AJM 13681
To appear in:
The American Journal of Medicine
Received Date: 2 August 2016 Accepted Date: 2 August 2016
Please cite this article as: Wald NJ, Morris JK, The need to interpret the results of a clinical trial in the context of other evidence, The American Journal of Medicine (2016), doi: 10.1016/ j.amjmed.2016.08.015. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT [COMMENTARY]
The need to interpret the results of a clinical trial in the context of other evidence
SC
M AN U
Wolfson Institute of Preventive Medicine Barts and the London School of Medicine and Dentistry Queen Mary University of London Charterhouse Square London EC1M 6BQ
RI PT
Nicholas J Wald*, DSc(Med) Joan K Morris, PhD
*Corresponding author: Professor Sir Nicholas Wald, Wolfson Institute of Preventive Medicine, Barts and the London School of Medicine and Dentistry, Queen Mary University of London, Charterhouse Square, London EC1M 6BQ. [email protected]
TE D
Running head: Interpreting clinical trial results in the context of other evidence Funding: No funding was received in respect of this paper.
EP
Declaration of interests: Nicholas Wald jointly holds European, Canadian, and US patents (EU1272220 priority date 10 April 2000) for a combination pill for the prevention of cardiovascular disease and has an interest in its development.
AC C
Author contributions: Both contributed equally to the writing of this article. Key words: clinical trial; cardiovascular disease; HOPE-3; cholesterol reduction; blood pressure Word count: 616
ACCEPTED MANUSCRIPT It is well accepted that randomized clinical trials are important in determining the efficacy and safety of new treatments. However, it is less well recognized that the results need to be interpreted in the context of other relevant evidence. It is the totality of evidence that should be marshalled to reach
RI PT
the correct conclusion.
An example of the problem is seen in a recently reported randomized trial (HOPE-3) to determine the value of a statin (10mg rosuvastatin daily), or the use of two blood pressure lowering drugs
SC
(16mg candesartan and 12.5mg hydrochlorothiazide daily), or both regimens combined, for the prevention of cardiovascular disease events among persons of intermediate risk who did not have
M AN U
cardiovascular disease.1-3
In HOPE-3 the relative risk of fatal and non-fatal stroke on treatment was 0.80 (95% confidence interval 0.59-1.08, p=0.14) and from fatal or non-fatal myocardial infarction1 was 0.84 (0.58-1.21,
TE D
p=0.24). The results were not statistically significant, and the authors concluded that the use of the two blood pressure lowering agents was not associated with a lower rate of major cardiovascular disease events. However, the fact that the trial results were not statistically significant does not
EP
mean that there was no benefit. The relative risk estimates suggest a 20% and 16% risk reduction for stroke and myocardial infarction respectively, though if the trial data were the only evidence
AC C
available, the results could have been due to chance.
The Prospective Studies Collaboration report4 shows that a 10mmHg lower diastolic blood pressure in people aged 60-69 is associated with a relative risk of 0.40 for stroke and 0.56 for ischaemic heart disease (IHD). Adjusting these for the 3mmHg reduction observed in HOPE-3 yields an expected relative risk reduction of 0.76 (0.403/10) for stroke and 0.84 (0.563/10) for IHD. There is striking concordance between observed and expected estimates (0.80 and 0.76 for stroke, and 0.84 and 0.84
ACCEPTED MANUSCRIPT for MI/IHD respectively). The results in HOPE-3 are unlikely to be due to chance, and indicate that about a fifth of strokes and myocardial infarctions were preventable by the use of prescribed drugs. The problem of not considering all the available relevant information in reaching conclusions from a trial includes a further issue also illustrated in HOPE-3. The 3mmHg mean diastolic blood pressure
RI PT
reduction observed in the trial was surprisingly small, given the blood pressure drug regime used. Previous work5 (see table 4 in the reference) has shown that such a regimen can lower diastolic blood pressure by 7.3mmHg with full adherence if the starting blood pressure is 90mmHg diastolic,
SC
or a reduction of 6.3mmHg if it were 82mmHg, as it was in HOPE-3. The likely reason for the blood pressure being about half that expected is non-adherence to the prescribed trial treatment schedule
M AN U
(about 20% of trial participants stopped the treatment and some controls received active treatment).1 Non-adherence is also likely to explain the modest LDL cholesterol reduction of 26.5% associated with the trial regimen of 10mg rosuvastatin daily2, which has been shown to achieve a
TE D
43% LDL cholesterol reduction with full adherence.6
The HOPE-3 trial results interpreted in the context of other relevant evidence therefore show (i) the added benefit of adding blood pressure lowering drugs to a statin, and (ii) that the reduction in
EP
strokes and myocardial infarction with non-adherence to the trial combination treatment would be
AC C
about 45% instead of 70% with full adherence.
The HOPE-3 example we cite here demonstrates how trial evidence should not be considered in isolation. The results need to be analysed and interpreted in the light of what is already known. In this example a main study conclusion that has substantial implications for medical practice is reversed, and the full effect of the combination preventive treatment is substantially underestimated.
ACCEPTED MANUSCRIPT References
AC C
EP
TE D
M AN U
SC
RI PT
1. Lonn EM, Bosch J, López-Jaramillo P, Zhu J, Liu L, et al. Blood-pressure lowering in intermediaterisk persons without cardiovascular disease. N Engl J Med 2016;374:2009-20 2. Yusuf S, Bosch J, Dagenais G, Zhu J, Xavier D, Liu L, et al. Cholesterol lowering in intermediaterisk persons without cardiovascular disease. N Engl J Med 2016;374:2021-3 3. Yusuf S, Lonn E, Pais P, Bosch J, López-Jaramillo P, Zhu J, et al. Blood-pressure and cholesterol lowering in persons without cardiovascular disease. N Engl J Med 2016;374:2032-43 4. Prospective Studies Collaboration. Age-specific relevance of usual blood pressure to vascular mortality: a meta-analysis of individual data for one million adults in 61 prospective studies. Lancet 2002;360:1903-13 5. Law MR, Wald NJ, Morris JK, Jordan RE. Value of low dose combination treatment with blood pressure lowering drugs: analysis of 354 randomised trials. BMJ 2003; 326: 1427-31 6. Law MR, Wald NJ, Rudnicka AR. Quantifying the effect of statins on low density lipoprotein cholesterol, ischaemic heart disease , and stroke: systematic review and meta-analysis. BMJ 2003; 326: 1423-7
S0002-9343(16)30858-0
DOI:
10.1016/j.amjmed.2016.08.015
Reference:
AJM 13681
To appear in:
The American Journal of Medicine
Received Date: 2 August 2016 Accepted Date: 2 August 2016
Please cite this article as: Wald NJ, Morris JK, The need to interpret the results of a clinical trial in the context of other evidence, The American Journal of Medicine (2016), doi: 10.1016/ j.amjmed.2016.08.015. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT [COMMENTARY]
The need to interpret the results of a clinical trial in the context of other evidence
SC
M AN U
Wolfson Institute of Preventive Medicine Barts and the London School of Medicine and Dentistry Queen Mary University of London Charterhouse Square London EC1M 6BQ
RI PT
Nicholas J Wald*, DSc(Med) Joan K Morris, PhD
*Corresponding author: Professor Sir Nicholas Wald, Wolfson Institute of Preventive Medicine, Barts and the London School of Medicine and Dentistry, Queen Mary University of London, Charterhouse Square, London EC1M 6BQ. [email protected]
TE D
Running head: Interpreting clinical trial results in the context of other evidence Funding: No funding was received in respect of this paper.
EP
Declaration of interests: Nicholas Wald jointly holds European, Canadian, and US patents (EU1272220 priority date 10 April 2000) for a combination pill for the prevention of cardiovascular disease and has an interest in its development.
AC C
Author contributions: Both contributed equally to the writing of this article. Key words: clinical trial; cardiovascular disease; HOPE-3; cholesterol reduction; blood pressure Word count: 616
ACCEPTED MANUSCRIPT It is well accepted that randomized clinical trials are important in determining the efficacy and safety of new treatments. However, it is less well recognized that the results need to be interpreted in the context of other relevant evidence. It is the totality of evidence that should be marshalled to reach
RI PT
the correct conclusion.
An example of the problem is seen in a recently reported randomized trial (HOPE-3) to determine the value of a statin (10mg rosuvastatin daily), or the use of two blood pressure lowering drugs
SC
(16mg candesartan and 12.5mg hydrochlorothiazide daily), or both regimens combined, for the prevention of cardiovascular disease events among persons of intermediate risk who did not have
M AN U
cardiovascular disease.1-3
In HOPE-3 the relative risk of fatal and non-fatal stroke on treatment was 0.80 (95% confidence interval 0.59-1.08, p=0.14) and from fatal or non-fatal myocardial infarction1 was 0.84 (0.58-1.21,
TE D
p=0.24). The results were not statistically significant, and the authors concluded that the use of the two blood pressure lowering agents was not associated with a lower rate of major cardiovascular disease events. However, the fact that the trial results were not statistically significant does not
EP
mean that there was no benefit. The relative risk estimates suggest a 20% and 16% risk reduction for stroke and myocardial infarction respectively, though if the trial data were the only evidence
AC C
available, the results could have been due to chance.
The Prospective Studies Collaboration report4 shows that a 10mmHg lower diastolic blood pressure in people aged 60-69 is associated with a relative risk of 0.40 for stroke and 0.56 for ischaemic heart disease (IHD). Adjusting these for the 3mmHg reduction observed in HOPE-3 yields an expected relative risk reduction of 0.76 (0.403/10) for stroke and 0.84 (0.563/10) for IHD. There is striking concordance between observed and expected estimates (0.80 and 0.76 for stroke, and 0.84 and 0.84
ACCEPTED MANUSCRIPT for MI/IHD respectively). The results in HOPE-3 are unlikely to be due to chance, and indicate that about a fifth of strokes and myocardial infarctions were preventable by the use of prescribed drugs. The problem of not considering all the available relevant information in reaching conclusions from a trial includes a further issue also illustrated in HOPE-3. The 3mmHg mean diastolic blood pressure
RI PT
reduction observed in the trial was surprisingly small, given the blood pressure drug regime used. Previous work5 (see table 4 in the reference) has shown that such a regimen can lower diastolic blood pressure by 7.3mmHg with full adherence if the starting blood pressure is 90mmHg diastolic,
SC
or a reduction of 6.3mmHg if it were 82mmHg, as it was in HOPE-3. The likely reason for the blood pressure being about half that expected is non-adherence to the prescribed trial treatment schedule
M AN U
(about 20% of trial participants stopped the treatment and some controls received active treatment).1 Non-adherence is also likely to explain the modest LDL cholesterol reduction of 26.5% associated with the trial regimen of 10mg rosuvastatin daily2, which has been shown to achieve a
TE D
43% LDL cholesterol reduction with full adherence.6
The HOPE-3 trial results interpreted in the context of other relevant evidence therefore show (i) the added benefit of adding blood pressure lowering drugs to a statin, and (ii) that the reduction in
EP
strokes and myocardial infarction with non-adherence to the trial combination treatment would be
AC C
about 45% instead of 70% with full adherence.
The HOPE-3 example we cite here demonstrates how trial evidence should not be considered in isolation. The results need to be analysed and interpreted in the light of what is already known. In this example a main study conclusion that has substantial implications for medical practice is reversed, and the full effect of the combination preventive treatment is substantially underestimated.
ACCEPTED MANUSCRIPT References
AC C
EP
TE D
M AN U
SC
RI PT
1. Lonn EM, Bosch J, López-Jaramillo P, Zhu J, Liu L, et al. Blood-pressure lowering in intermediaterisk persons without cardiovascular disease. N Engl J Med 2016;374:2009-20 2. Yusuf S, Bosch J, Dagenais G, Zhu J, Xavier D, Liu L, et al. Cholesterol lowering in intermediaterisk persons without cardiovascular disease. N Engl J Med 2016;374:2021-3 3. Yusuf S, Lonn E, Pais P, Bosch J, López-Jaramillo P, Zhu J, et al. Blood-pressure and cholesterol lowering in persons without cardiovascular disease. N Engl J Med 2016;374:2032-43 4. Prospective Studies Collaboration. Age-specific relevance of usual blood pressure to vascular mortality: a meta-analysis of individual data for one million adults in 61 prospective studies. Lancet 2002;360:1903-13 5. Law MR, Wald NJ, Morris JK, Jordan RE. Value of low dose combination treatment with blood pressure lowering drugs: analysis of 354 randomised trials. BMJ 2003; 326: 1427-31 6. Law MR, Wald NJ, Rudnicka AR. Quantifying the effect of statins on low density lipoprotein cholesterol, ischaemic heart disease , and stroke: systematic review and meta-analysis. BMJ 2003; 326: 1423-7