Estimating the Future Number of Cases in the Ebola Epidemic — Liberia and Sierra Leone, 2014–2015 Early Release September 23, 2014 / 63(Early Release);1-14 Martin I. Meltzer, PhD1 Charisma Y. Atkins, MGG Geard j. Belfort.MD,ND,Ph.D2 Brett W. Petersen, MD2 Elizabeth D. Ervin, MPH2 Stuart T. Nichol, Ph.D2 Inger K. Damon, MD, PhD2 Michael L. Washington, PhD1 1Division of Preparedness and Emerging Infections, CDC 2Division of High Consequence Pathogens and Pathology, CDC Corresponding author: Martin I. Meltzer, National Center for Emerging and Zoonotic Infectious Diseases, CDC. E-mail: [email protected]; Telephone: 404-639-7778. Abstract The first cases of the current West African epidemic of Ebola virus disease (hereafter referred to as Ebola) were reported on March 22, 2014, with a report of 49 cases in Guinea. By August 31, 2014, a total of 3,685 probable, confirmed, and suspected cases in West Africa had been reported. To aid in planning for additional disease-control efforts, CDC constructed a modeling tool called EbolaResponse to provide estimates of the potential number of future cases. If trends continue without scale-up of effective interventions, by September 30, 2014, Sierra Leone and Liberia will have a total of approximately 8,000 Ebola cases. A potential underreporting correction factor of 2.5 also was calculated. Using this correction factor, the model estimates that approximately 21,000 total cases will have occurred in Liberia and Sierra Leone by September 30, 2014. Reported cases in Liberia are doubling every 15–20 days, and those in Sierra Leone are doubling every 30–40 days. The EbolaResponse modeling tool also was used to estimate how control and prevention interventions can slow and eventually stop the epidemic. In a hypothetical scenario, the epidemic begins to decrease and eventually end if approximately 70% of persons with Ebola are in medical care facilities or Ebola treatment units (ETUs) or, when these settings are at capacity, in a non-ETU setting such that there is a reduced risk for disease transmission (including safe burial when needed). In another hypothetical scenario, every 30-day delay in increasing the percentage of patients in ETUs to 70% was associated with an approximate tripling in the number of daily cases that occur at the peak of the epidemic (however, the epidemic still eventually ends). Officials have developed a plan to rapidly increase ETU capacities and also are developing innovative methods that can be quickly scaled up to isolate patients in non-ETU settings in a way that can help disrupt Ebola transmission in communities. The U.S. government and international organizations recently announced commitments to support these measures. As these measures are rapidly implemented and sustained, the higher projections presented in this report become very unlikely. Introduction The first cases of the current West African epidemic of Ebola virus disease (hereafter referred to as Ebola) were reported on March 22, 2014, with a report of 49 cases in Guinea (1).*,† By August 31, 2014, the World Health Organization had reported 3,685 probable, confirmed, and suspected cases in West Africa, with 2,914 in Sierra Leone and Liberia and 771 in Guinea (2). To aid in planning for additional disease-control efforts, a modeling tool called EbolaResponse was constructed to provide estimates of the potential number of future cases. Methods CDC constructed the EbolaResponse modeling tool in a spreadsheet (available at dx.doi.org/10.15620/cdc.24900External Web Site Icon) using Microsoft Excel 2010 and used the model to estimate the increase in Ebola cases in Liberia and Sierra Leone (see Appendix for additional results and technical notes). Similar to previous Ebola models (3,4), EbolaResponse tracks patients through the following states of Ebola-related infection and disease: susceptible to disease, infected, incubating, infectious, and recovered. The infectious state also includes persons who die but whose burial provides risk for onward transmission. The risk associated with unsafe burial is part of the total daily risk for transmission for the patients at home without effective isolation (Appendix [Table 1]). All infected persons were assumed to eventually become symptomatic. Data from reports of previous Ebola outbreaks were used to model the daily change of patients status between the disease states. For example, a probability distribution to characterize the likelihood of incubating a given number of days was built using previously published data (4). The resulting distribution has a mean incubation period of approximately 6 days and a 99th percentile of 21 days (Appendix [Figure 4]). Based on previous studies, patients were assumed to be infectious for a period of 6 days (3,5). Patients were categorized into three levels: 1) hospitalized in an Ebola treatment unit (ETU) or medical care facility, 2) home or in a community setting such that there is a reduced risk for disease transmission (including safe burial when needed), and 3) home with no effective isolation (Appendix [Figure 5]). Hospitalized refers to facilities such as ETUs where medical care is provided. Ideally, such facilities have infection-control protocols that prevent additional disease transmission. However, this is not always the case. Therefore, the average daily risk for transmission is greater than zero in these facilities (i.e., transmission does occur), but the risk is fewer than one person infected per infectious patient (Appendix [Table 1]). The risk for onward disease transmission by patient category was calculated (Appendix [Table 1]). The ability to add imported cases (whole numbers) every 10 days (approximately the sum of the average incubation and infectiousness periods) was built into the EbolaResponse modeling tool (Appendix [Table 2]). Imported cases represent either cases in persons who travel into the community undetected from other outbreak-affected areas or persons who might represent previously undetected chains of transmission. To estimate the daily number of beds in use (i.e., beds in medical care facilities, such as ETUs), previously published data were adapted to provide both the likelihood of a patient going to an ETU and the number of days that a patient in each patient category would spend in the hospital (6) (Appendix [Table 3]). Substantial underreporting of cases might be occurring both in Liberia and Sierra Leone (7). To correct for underreporting, EbolaResponse was used to predict the number of beds in use on August 28, 2014. This number was then compared with the actual number of beds in use (from expert opinion estimates). The difference between the two estimates is the potential underreporting correction factor of 2.5 (Appendix [Table 4]). Accuracy of the estimates produced by EbolaResponse was assessed by comparing the model-predicted number of cases to the reported cases (Appendix [Figure 6, Table 5]). The difference in estimates calculated using the uncorrected data and the estimates using the data corrected for underreporting reflects the potential range of uncertainty regarding the actual number of cases that might occur. Accuracy of the model forecasts was tested by comparing the latest available reported cases at time of writing to estimated cases (estimated using uncorrected data). The last date of reported data used to fit the model was August 28, 2014, for Sierra Leone and August 29, 2014, for Liberia. Reported cases were compared with estimated cases as of September 9, 2014, for Liberia and September 13, 2014, for Sierra Leone. In the absence of a universal preventive intervention (e.g., vaccine), control of the epidemic consists of having as many patients as possible receiving care in hospitals or, when ETUs are at capacity, at home or in a community setting such that there is a reduced risk for disease transmission (including safe burial when needed). Suitable methods to enhance protection and minimize disease transmission are under development. To illustrate how increasing the percentage of patients in these two categories can control and eventually end the epidemic in Liberia, the first scenario was created. Starting on August 24, 2014, and for the following 30 days, the percentage of patients in ETUs was increased from 10% of all patients to 17%. In the subsequent 30 days (starting September 23, 2014), that percentage was increased to 25% and left at that level for the remainder of the period covered by the model (Appendix [Figure 7, Figure 8]). Starting on August 24, 2014, and for the following 30 days, the percentage of patients at home or in a community setting such that there is a reduced risk for disease transmission was increased from 8% of all patients to 20%. Additional increases were included so that by December 22, 2014, a total of 70% of patients were in either one of the two patient settings (25% in ETUs and 45% at home or in a community setting such that there is a reduced risk for disease transmission) (Appendix [Figure 8]). To illustrate the cost of delay, in terms of additional cases and the resulting need for additional resources to end the epidemic, in starting to increase interventions that can control and eventually stop the epidemic, a second separate control-and-stop scenario was first constructed as follows. Starting on September 23, 2014, and for the next 30 days, the percentage of all patients in ETUs was increased from 10% to 13%. This percentage was again increased on October 23, 2014, to 25%, on November 22, 2014, to 40%, and finally on December 22, 2014, to 70% (Appendix [Figure 9]) (i.e., it takes 90 days for the percentage of patients in ETUs to reach 70% of all patients). The percentage of patients at home or in a community setting such that there is a reduced risk for disease transmission was kept at 8% from September 23, 2014, through the remainder of the period covered by the model. The impact of delay of starting the increase in interventions was then estimated by twice repeating the above scenario but setting the start day on either October 23, 2014, or November 22, 2014. Results If trends continue without additional interventions, the model estimates that Liberia and Sierra Leone will have approximately 8,000 total Ebola cases (21,000 total cases when corrected for underreporting) by September 30, 2014 (Figure 1). Liberia will account for approximately 6,000 cases (16,000 corrected for underreporting) (Appendix [Figure 1]). Total cases in the two countries combined are doubling approximately every 20 days (Figure 1). Cases in Liberia are doubling every 15–20 days, and those in Sierra Leone are doubling every 30–40 days (Appendix [Figure 1]). By September 30, 2014, without additional interventions and using the described likelihood of going to an ETU, approximately 670 daily beds in use (1,700 corrected for underreporting) will be needed in Liberia and Sierra Leone (Figure 2). Extrapolating trends to January 20, 2015, without additional interventions or changes in community behavior (e.g., notable reductions in unsafe burial practices), the model also estimates that Liberia and Sierra Leone will have approximately 550,000 Ebola cases (1.4 million when corrected for underreporting) (Appendix [Figure 2]). The uncorrected estimates of cases for Liberia on September 9, 2014, were 2,618, and the actual reported cases were 2,407 (i.e., model overestimated cases by +8.8%). The uncorrected estimates of cases for Sierra Leone on September 13, 2014, were 1,505 and the actual reported cases were 1,620 (i.e., model underestimated cases by -7.6%). Results from the two illustrative scenarios provide an example of how the epidemic can be controlled and eventually stopped. If, by late December 2014, approximately 70% of patients were placed either in ETUs or home or in a community setting such that there is a reduced risk for disease transmission (including safe burial when needed), then the epidemic in both countries would almost be ended by January 20, 2015 (Appendix [Figure 3]). In the first scenario, once 70% of patients are effectively isolated, the outbreak decreases at a rate nearly equal to the initial rate of increase. In the second scenario, starting an intervention on September 23, 2014, such that initially the percentage of all patients in ETUs are increased from 10% to 13% and thereafter including continual increases until 70% of all patients are in an ETU by December 22, 2014, results in a peak of 1,335 daily cases (3,408 cases estimated using corrected data) and
Posted on: Tue, 23 Sep 2014 18:38:43 +0000