Correspondence
Published Online September 23, 2014 http://dx.doi.org/10.1016/ S0140-6736(14)61706-2
See Online for appendix
Case fatality rate for Ebola virus disease in west Africa The case fatality rate (CFR) for the 2014 Ebola outbreak in west Africa has been widely reported to be much lower than for most previous outbreaks.1 However, this low rate is not necessarily a feature of the infection itself. Rather, it is likely to be the result of a failure to account for delays between disease onset and final outcome. The low reported CFR values were generated from a so-called naive CFR calculation,2 in which the total number of deaths reported so far is divided by the total number of cases. Based on WHO reports up to Sept 7, 2014, which include 2226 deaths and 4390 cases,1 the naive CFR estimate is 51% (95% CI 49–53%). This naive approach does not account for the delay between onset of Ebola symptoms and disease outcome (ie, recovery or death). During the 1976 outbreak in Yambuku, Democratic Republic of the Congo,3 this delay was 7·5 days on average (appendix). In the middle of the outbreak, cases for which the outcome was as-yet unknown existed (appendix). Because the naive CFR calculation includes these cases—but not their outcomes—it generates a substantial
underestimate of the actual CFR. Halfway through the 1976 outbreak, the naive CFR estimate would have been around 50%; as the outbreak reached its conclusion, this number would have climbed towards the much higher true value (figure). By contrast, if we only consider cases with known outcomes, the realtime estimate of CFR remains consistently high throughout. If cumulative incidences of cases and deaths are available, and delay from onset to outcome is known, the number of cases with outcomes can be estimated and hence a more accurate estimate of CFR obtained.4,5 We estimate that the 2014 infection has an overall CFR of around 70% at present using the 1976 distribution of Ebola onset to outcome and WHO reports on total cases and deaths across all countries in 2014.1 If the delay is longer than in 1976, this CFR could be even higher. The widely cited 2014 CFR of around 50% is therefore likely to be a substantial underestimate of the true value, and so the number could apparently rise over the course of the outbreak. With data on individual onsets and outcomes, more precise estimates of CFR could be obtained, and how it varies with setting and availability of treatment could be assessed. We declare no competing interests.
1·0
Case fatality rate
0·8
0·6
0·4 Naive Naive 95% CI Known outcome Known outcome 95% CI
0·2
0 Sept 1
Sept 15
Oct 1 Date
Oct 15
Nov 1
Figure: Realtime estimation of case fatality rate using data from the 1976 Yambuku Ebola outbreak Data are from reference 3.
1260
*Adam J Kucharski, W John Edmunds [email protected] Department of Infectious Disease Epidemiology, London School of Hygiene Tropical Medicine, London WC1E 7HT, UK 1 2
3
4
5
WHO. Ebola virus disease. http://who.int/csr/ disease/ebola/en (accessed Sept 15, 2014). Ghani AC, Donnelly CA, Cox DR, et al. Methods for estimating the case fatality ratio for a novel, emerging infectious disease. Am J Epidemiol 2005; 162: 479–86. Breman JG, Piot P, Johnson KM, et al. The epidemiology of Ebola hemorrhagic fever in Zaire, 1976. In: Pattyn SR, ed. Ebola virus haemorrhagic fever. Amsterdam: Elsevier, 1978: 85–97. Garske T, Legrand J, Donnelly CA, et al. Assessing the severity of the novel influenza A/H1N1 pandemic. BMJ 2009; 339: b2840. Nishiura H, Klinkenberg D, Roberts M, Heesterbeek JA. Early epidemiological assessment of the virulence of emerging infectious diseases: a case study of an influenza pandemic. PLoS One; 4: e6852.
Neuraminidase inhibitors for influenza complications In their Comment (Aug 2, p 386),1 Jonathan Nguyen-Van-Tam and colleagues suggest that findings from our Cochrane review2 and a study of observational data3 are consistent. In our review, which was based on full clinical study reports of all manufacturer-sponsored randomised trials, we did not find evidence that neuraminidase inhibitors improve important outcomes of influenza, whereas the Roche-funded individual analysis of a subset of retrospective case reports suggested that neuraminidase inhibitors do have some beneficial effect. These observational studies were of patients admitted to hospital for influenza, some of whom apparently benefited from neuraminidase inhibitors.3 Most treated patients received oseltamivir, with a minority receiving zanamivir. Similar evidence was cited in a statement from Roche.4 Our Cochrane review did not include similar patients, but was based on typical patients with influenza-like illness. However, the hypothesis that oseltamivir protects against complications of influenza proposed by www.thelancet.com Vol 384 October 4, 2014
Correspondence
www.thelancet.com Vol 384 October 4, 2014
insufficient evidence exists to justify conclusions about oseltamivir’s effectiveness for treatment of influenza. Nguyen-Van-Tam and colleagues seem to suggest that pandemic planners should act to minimise public outcry rather than accept the best available evidence. This advice could risk scarce health-care resources stockpiling a probably ineffective drug. We declare no competing interests.
*Chris Del Mar, Peter Doshi, Rokuro Hama, Mark Jones, Tom Jefferson, Carl Heneghan, Igho Onakpoya, Jeremy Howick [email protected] Centre for Research in Evidence Based Practice, Bond University, Robina, QLD 4226, Australia (CDB); University of Maryland School of Pharmacy, MD, USA (PD); Japan Institute of Pharmacovigilance, Osaka, Japan (RH); School of Population Health, University of Queensland, QLD, Australia (MJ); Cochrane Collaboration, Rome, Italy (TD); and Nuffield Department of Primary Care Health Sciences, University of Oxford, Oxford, UK (CH, IO, JH) 1
2
3
4
5
6
7
Nguyen-Van-Tam JS, Openshaw PJ, Nicholson KG. Antivirals for influenza: where now for clinical practice and pandemic preparedness? Lancet 2014; 384: 386–87. Jefferson T, Jones M, Doshi P, et al. Neuraminidase inhibitors for preventing and treating influenza in healthy adults and children. Cochrane Database Syst Rev 2014; 4: CD008965. Muthuri SG, Venkatesan S, Myles PR, et al; for PRIDE Consortium Investigators, Nguyen-Van-Tam JS. Effectiveness of neuraminidase inhibitors in reducing mortality in patients admitted to hospital with influenza A H1N1pdm09 virus infection: a meta-analysis of individual participant data. Lancet Resp Med 2014; 2: 395–404. Roche statement. http://www.roche.com/ media/med-stat-2014-04-10.htm (accessed June 7, 2014). Kaiser L, Wat C, Mills T, Mahoney P, Ward P, Hayden F. Impact of oseltamivir treatment on influenza-related lower respiratory tract complications and hospitalizations. Arch Intern Med 2003; 163: 1667–72. Tamiflu (oseltamivir phosphate) prescribing information. 2010. http://www.accessdata. fda.gov/drugsatfda_docs/label/2010/021087s 048s049,021246s034s035lbl.pdf (accessed June 6, 2014). Chalmers I. The development of fair tests of treatments. Lancet 2014; 383: 1713–14.
Authors’ reply We would like to thank Chris Del Mar and colleagues for their interest in our Comment.1 They assert that their Cochrane review2 was based on full clinical study reports of all manufacturer-sponsored randomised
trials. In fact, only a subset (46 of the 107 Clinical Study Reports obtained) were formally analysed. The study undertaken by Muthuri and colleagues3 was funded by Roche, but its design, conduct, interpretation, and report preparation were done independently of the funder. Exhaustive attempts were made to obtain datasets suitable for analysis from around the world. Compared with the 80 datasets received, few (n=15) were not shared because of review board or governmental restrictions (n=3), or inability to meet project timelines (n=12).3 None of the contributors of data declared industry funding for acquisition or assembly of their dataset. We do not dismiss the findings in the Cochrane review.2 The finding that no signal exists suggesting that neuraminidase inhibitors reduce serious complications2 is not unexpected in light of the fact that the clinical trials reviewed were done in community settings, were based on mostly healthy patients with mild influenza-like illness, and were not designed or powered to assess effect on severe illness. Del Mar and colleagues’ assessment of effectiveness of oseltamivir against hospital admission was based on 4394 adults and 1359 children—for pneumonia, this assessment was for 4452 patients, and for serious complications, 3675 patients. Mortality, a hallmark of pandemics and seasonal outbreaks, could not be assessed because of the absence of deaths in the oseltamivir trials. The same restrictions apply to the pooled analysis of 3564 patients in ten randomised controlled trials of oseltamivir treatment by Kaiser and colleagues,4 in which, unlike in the Cochrane review,2 these authors showed that treatment was associated with reductions in influenza-related lower respiratory complications and admission to hospital from any cause. By contrast, the observational data assembled and analysed by Muthuri and colleagues 3 used individual participant data (an approach not pursued in the Cochrane review2)
Giphotostock/Science Photo Library
Nguyen-Van-Tam and colleagues, together with the observational studies that they cite in support of their hypothesis,3 can be dismissed by our finding that neuraminidase inhibitors did not seem to reduce severe complications of influenza, and oseltamivir specifically did not seem to reduce risk of admission to hospital (admission to hospital was not reported for zanamivir), but seemed to suppress production of antibody to the virus and cause potentially serious side effects.2 Furthermore, oseltamivir reduced time to alleviation of symptoms the least for trials of patients with comorbidities such as asthma and chronic obstructive airways disease.2 The appeal to observational data in support of oseltamivir is curious because the original notion that oseltamivir was effective for treatment of influenza complications came from a meta-analysis5 of a subset of randomised trials—both published and unpublished—which was in contrast to the US Food and Drug Administration’s requirement that the drug label made no claims for effectiveness for influenza complications.6 This absence of effectiveness is supported by our Cochrane review (based on a full set of clinical study reports), which found no protective benefit of oseltamivir. Now drug manufacturers and Nguyen-Van-Tam and colleagues are claiming that randomised trial data are no longer adequate. Yet, when the results of reviews of randomised trials agreed with their views, Nguyen-Van-Tam and colleagues accepted the reviews. Now that the Cochrane review does not support their hypothesis, they appeal to contradictory real-world (observational) data.3 This inconsistency needs to be addressed. Observational data are useful to answer many questions, such as whether treatments are harmful or not. However, the fact that experimental data from randomised trials are less prone to error from bias than are observational studies is not debatable.7 The randomised trial data should therefore be trusted—
1261
Published Online September 23, 2014 http://dx.doi.org/10.1016/ S0140-6736(14)61706-2
See Online for appendix
Case fatality rate for Ebola virus disease in west Africa The case fatality rate (CFR) for the 2014 Ebola outbreak in west Africa has been widely reported to be much lower than for most previous outbreaks.1 However, this low rate is not necessarily a feature of the infection itself. Rather, it is likely to be the result of a failure to account for delays between disease onset and final outcome. The low reported CFR values were generated from a so-called naive CFR calculation,2 in which the total number of deaths reported so far is divided by the total number of cases. Based on WHO reports up to Sept 7, 2014, which include 2226 deaths and 4390 cases,1 the naive CFR estimate is 51% (95% CI 49–53%). This naive approach does not account for the delay between onset of Ebola symptoms and disease outcome (ie, recovery or death). During the 1976 outbreak in Yambuku, Democratic Republic of the Congo,3 this delay was 7·5 days on average (appendix). In the middle of the outbreak, cases for which the outcome was as-yet unknown existed (appendix). Because the naive CFR calculation includes these cases—but not their outcomes—it generates a substantial
underestimate of the actual CFR. Halfway through the 1976 outbreak, the naive CFR estimate would have been around 50%; as the outbreak reached its conclusion, this number would have climbed towards the much higher true value (figure). By contrast, if we only consider cases with known outcomes, the realtime estimate of CFR remains consistently high throughout. If cumulative incidences of cases and deaths are available, and delay from onset to outcome is known, the number of cases with outcomes can be estimated and hence a more accurate estimate of CFR obtained.4,5 We estimate that the 2014 infection has an overall CFR of around 70% at present using the 1976 distribution of Ebola onset to outcome and WHO reports on total cases and deaths across all countries in 2014.1 If the delay is longer than in 1976, this CFR could be even higher. The widely cited 2014 CFR of around 50% is therefore likely to be a substantial underestimate of the true value, and so the number could apparently rise over the course of the outbreak. With data on individual onsets and outcomes, more precise estimates of CFR could be obtained, and how it varies with setting and availability of treatment could be assessed. We declare no competing interests.
1·0
Case fatality rate
0·8
0·6
0·4 Naive Naive 95% CI Known outcome Known outcome 95% CI
0·2
0 Sept 1
Sept 15
Oct 1 Date
Oct 15
Nov 1
Figure: Realtime estimation of case fatality rate using data from the 1976 Yambuku Ebola outbreak Data are from reference 3.
1260
*Adam J Kucharski, W John Edmunds [email protected] Department of Infectious Disease Epidemiology, London School of Hygiene Tropical Medicine, London WC1E 7HT, UK 1 2
3
4
5
WHO. Ebola virus disease. http://who.int/csr/ disease/ebola/en (accessed Sept 15, 2014). Ghani AC, Donnelly CA, Cox DR, et al. Methods for estimating the case fatality ratio for a novel, emerging infectious disease. Am J Epidemiol 2005; 162: 479–86. Breman JG, Piot P, Johnson KM, et al. The epidemiology of Ebola hemorrhagic fever in Zaire, 1976. In: Pattyn SR, ed. Ebola virus haemorrhagic fever. Amsterdam: Elsevier, 1978: 85–97. Garske T, Legrand J, Donnelly CA, et al. Assessing the severity of the novel influenza A/H1N1 pandemic. BMJ 2009; 339: b2840. Nishiura H, Klinkenberg D, Roberts M, Heesterbeek JA. Early epidemiological assessment of the virulence of emerging infectious diseases: a case study of an influenza pandemic. PLoS One; 4: e6852.
Neuraminidase inhibitors for influenza complications In their Comment (Aug 2, p 386),1 Jonathan Nguyen-Van-Tam and colleagues suggest that findings from our Cochrane review2 and a study of observational data3 are consistent. In our review, which was based on full clinical study reports of all manufacturer-sponsored randomised trials, we did not find evidence that neuraminidase inhibitors improve important outcomes of influenza, whereas the Roche-funded individual analysis of a subset of retrospective case reports suggested that neuraminidase inhibitors do have some beneficial effect. These observational studies were of patients admitted to hospital for influenza, some of whom apparently benefited from neuraminidase inhibitors.3 Most treated patients received oseltamivir, with a minority receiving zanamivir. Similar evidence was cited in a statement from Roche.4 Our Cochrane review did not include similar patients, but was based on typical patients with influenza-like illness. However, the hypothesis that oseltamivir protects against complications of influenza proposed by www.thelancet.com Vol 384 October 4, 2014
Correspondence
www.thelancet.com Vol 384 October 4, 2014
insufficient evidence exists to justify conclusions about oseltamivir’s effectiveness for treatment of influenza. Nguyen-Van-Tam and colleagues seem to suggest that pandemic planners should act to minimise public outcry rather than accept the best available evidence. This advice could risk scarce health-care resources stockpiling a probably ineffective drug. We declare no competing interests.
*Chris Del Mar, Peter Doshi, Rokuro Hama, Mark Jones, Tom Jefferson, Carl Heneghan, Igho Onakpoya, Jeremy Howick [email protected] Centre for Research in Evidence Based Practice, Bond University, Robina, QLD 4226, Australia (CDB); University of Maryland School of Pharmacy, MD, USA (PD); Japan Institute of Pharmacovigilance, Osaka, Japan (RH); School of Population Health, University of Queensland, QLD, Australia (MJ); Cochrane Collaboration, Rome, Italy (TD); and Nuffield Department of Primary Care Health Sciences, University of Oxford, Oxford, UK (CH, IO, JH) 1
2
3
4
5
6
7
Nguyen-Van-Tam JS, Openshaw PJ, Nicholson KG. Antivirals for influenza: where now for clinical practice and pandemic preparedness? Lancet 2014; 384: 386–87. Jefferson T, Jones M, Doshi P, et al. Neuraminidase inhibitors for preventing and treating influenza in healthy adults and children. Cochrane Database Syst Rev 2014; 4: CD008965. Muthuri SG, Venkatesan S, Myles PR, et al; for PRIDE Consortium Investigators, Nguyen-Van-Tam JS. Effectiveness of neuraminidase inhibitors in reducing mortality in patients admitted to hospital with influenza A H1N1pdm09 virus infection: a meta-analysis of individual participant data. Lancet Resp Med 2014; 2: 395–404. Roche statement. http://www.roche.com/ media/med-stat-2014-04-10.htm (accessed June 7, 2014). Kaiser L, Wat C, Mills T, Mahoney P, Ward P, Hayden F. Impact of oseltamivir treatment on influenza-related lower respiratory tract complications and hospitalizations. Arch Intern Med 2003; 163: 1667–72. Tamiflu (oseltamivir phosphate) prescribing information. 2010. http://www.accessdata. fda.gov/drugsatfda_docs/label/2010/021087s 048s049,021246s034s035lbl.pdf (accessed June 6, 2014). Chalmers I. The development of fair tests of treatments. Lancet 2014; 383: 1713–14.
Authors’ reply We would like to thank Chris Del Mar and colleagues for their interest in our Comment.1 They assert that their Cochrane review2 was based on full clinical study reports of all manufacturer-sponsored randomised
trials. In fact, only a subset (46 of the 107 Clinical Study Reports obtained) were formally analysed. The study undertaken by Muthuri and colleagues3 was funded by Roche, but its design, conduct, interpretation, and report preparation were done independently of the funder. Exhaustive attempts were made to obtain datasets suitable for analysis from around the world. Compared with the 80 datasets received, few (n=15) were not shared because of review board or governmental restrictions (n=3), or inability to meet project timelines (n=12).3 None of the contributors of data declared industry funding for acquisition or assembly of their dataset. We do not dismiss the findings in the Cochrane review.2 The finding that no signal exists suggesting that neuraminidase inhibitors reduce serious complications2 is not unexpected in light of the fact that the clinical trials reviewed were done in community settings, were based on mostly healthy patients with mild influenza-like illness, and were not designed or powered to assess effect on severe illness. Del Mar and colleagues’ assessment of effectiveness of oseltamivir against hospital admission was based on 4394 adults and 1359 children—for pneumonia, this assessment was for 4452 patients, and for serious complications, 3675 patients. Mortality, a hallmark of pandemics and seasonal outbreaks, could not be assessed because of the absence of deaths in the oseltamivir trials. The same restrictions apply to the pooled analysis of 3564 patients in ten randomised controlled trials of oseltamivir treatment by Kaiser and colleagues,4 in which, unlike in the Cochrane review,2 these authors showed that treatment was associated with reductions in influenza-related lower respiratory complications and admission to hospital from any cause. By contrast, the observational data assembled and analysed by Muthuri and colleagues 3 used individual participant data (an approach not pursued in the Cochrane review2)
Giphotostock/Science Photo Library
Nguyen-Van-Tam and colleagues, together with the observational studies that they cite in support of their hypothesis,3 can be dismissed by our finding that neuraminidase inhibitors did not seem to reduce severe complications of influenza, and oseltamivir specifically did not seem to reduce risk of admission to hospital (admission to hospital was not reported for zanamivir), but seemed to suppress production of antibody to the virus and cause potentially serious side effects.2 Furthermore, oseltamivir reduced time to alleviation of symptoms the least for trials of patients with comorbidities such as asthma and chronic obstructive airways disease.2 The appeal to observational data in support of oseltamivir is curious because the original notion that oseltamivir was effective for treatment of influenza complications came from a meta-analysis5 of a subset of randomised trials—both published and unpublished—which was in contrast to the US Food and Drug Administration’s requirement that the drug label made no claims for effectiveness for influenza complications.6 This absence of effectiveness is supported by our Cochrane review (based on a full set of clinical study reports), which found no protective benefit of oseltamivir. Now drug manufacturers and Nguyen-Van-Tam and colleagues are claiming that randomised trial data are no longer adequate. Yet, when the results of reviews of randomised trials agreed with their views, Nguyen-Van-Tam and colleagues accepted the reviews. Now that the Cochrane review does not support their hypothesis, they appeal to contradictory real-world (observational) data.3 This inconsistency needs to be addressed. Observational data are useful to answer many questions, such as whether treatments are harmful or not. However, the fact that experimental data from randomised trials are less prone to error from bias than are observational studies is not debatable.7 The randomised trial data should therefore be trusted—
1261
