Ebola Outbreak Finally Receding in Sierra Leone; CDC Modeling Was Incredibly Accurate

Back in late September, just a week before Ebola panic hit a peak in the US when a patient in Dallas was diagnosed with the disease, the CDC produced a remarkable study in which they modeled the expected number of Ebola cases both with and without intervention. That study received a huge amount of press coverage, primarily because the model predicted that without intervention by public health authorities, as many as 1.4 million people could be infected. By contrast, with a program of isolating infected patients and educating survivors on proper burial techniques, the model showed that the outbreak would be much less widespread. The modeling projected cases through yesterday’s date, January 20.

Less reported in the media at the time was the projected number of cases under the scenario of intervention. The model predicted an actual number of cases between 25,000 and 30,000 by this week and a reported number of cases of nearly 10,000. Here are the two projections placed alongside one another:

CDC modeling of projected number of Ebola cases without (left) and with (right) improved patient isolation and safe burial practices.

CDC modeling of projected number of Ebola cases without (left) and with (right) improved patient isolation and safe burial practices.

The latest data from WHO indicate just over 21,000 cases as of January 11. That is a remarkable achievement by the team that developed the model. The observed actual number of reported cases fell squarely within the range predicted by the model. With the influx of health professionals to the region to provide care for infected patients, it seems likely to me that the correction factor applied in the CDC model to correct from the reported number of cases to the actual number would be very different now, so that the reported number and actual number would be much closer to one another, making the prediction even more accurate.

Last time I posted on progress in stopping the spread of the virus, we saw that the rate of appearance of new cases was dropping rapidly in Liberia but was still accelerating in Sierra Leone. The good news is that the improved practices have finally been implemented sufficiently in Sierra Leone that the rate is now dropping there. Here are the plots of weekly new cases in the two countries from the latest WHO Situation Report:

Weekly number of new cases of Ebola in Liberia (left) and Sierra Leone (right). Control of the virus was achieved about two months later in Sierra Leone than in Liberia.

Weekly number of new cases of Ebola in Liberia (left) and Sierra Leone (right). Control of the virus was achieved about two months later in Sierra Leone than in Liberia.

Although the battle is not yet over, all indications are that the outbreak is well past the worst phase and should end soon. Considering how closely the CDC model predicted the eventual size of the outbreak with the control measures that were implemented, it seems safe to say that the world would have witnessed a truly horrific level of spread of the virus had improved safety measures not been implemented. As of the January 14 WHO Situation Report, a total of 825 health care workers have been infected, with 493 of them dying. Without their sacrifices, many more would have been lost.

Ebola Outbreak Receding in Liberia, Still Strong in Sierra Leone

Back in late September, the press had a field day with a mathematical model developed by CDC that estimated that if left unchecked, the Ebola outbreak in West Africa could wind up infecting over 1.4 million people. Almost missed in the hysteria over that high number was the fact that this same model predicted that even with key public health measures (patient isolation, monitoring of at-risk population who had contact with infected people and safe burial practices) falling short of 100% implementation, the outbreak could be brought under control around January of next year.

Word has been leaking out for a while now that the rate of new Ebola infections in Liberia is falling. Reports in the Washington Post on October 29 and November 3 told us as much. A chart in the WHO Situation Report for November 5 drives home just how dramatic the decline in new cases has become:

WHO Ebola Situation Report November 5, 2014

WHO Ebola Situation Report November 5, 2014

As can be seen in the chart, the rate of new infections for the two most recent weeks is less than one fourth the rate at the peak of the outbreak. Unfortunately, the news for Sierra Leone is not as good. While the rate of new infections may be leveling off, it is not yet falling appreciably:

WHO Ebola Situation Report November 5, 2014

WHO Ebola Situation Report November 5, 2014

Digging into the WHO report a bit further, we can find some evidence for how this dramatic drop in new cases has been brought about. We see that 52% of cases are now isolated. The WHO target for December 1 has been set at 70%, with a target of 100% by January 1. When it comes to management of dead bodies, though, the December 1 target has already been surpassed. WHO reports that 87% of the dead are being “managed in a safe and dignified manner” while the targets were set at 70% for December 1 and 100% for January 1. Also, although no benchmarks are reported, WHO states that 95% of registered contacts were reached daily (although in the text of the report, there are suggestions this number may be somewhat overstated).

It should come as no surprise that progress in implementing these basic measures has had a huge impact on bringing down the rate of new infections. It fits perfectly with the CDC mathematical model and it also addresses the known biology of Ebola infections. Patients are most infectious at or near death, so establishing safe burial practices is vitally important. Conversely, identifying infected individuals through daily monitoring of the at-risk population and then isolating infected individuals once symptoms begin means that far fewer people are exposed to people producing large amounts of virus.

Sadly, those who remain exposed are the health care workers who are providing care to those who are infected. Despite shortages of equipment and supplies, WHO and other organizations are doing their best to overcome those shortages and to beef up training to reduce risk to these brave people on the front lines in the work to control the virus. As of this November 5 report, 546 health care workers have been infected, with 310 of them dying. Only four new infections were reported for the week ending November 2, so it is hoped that this rate is also dropping.

Had the alarmists who insisted that this was a new super-strain of Ebola capable of airborne transmission (or even a strain developed in a bioweapons laboratory), it is doubtful that these basic public health measures would have had such a dramatic impact on the rate of new infections. Perhaps those folks can go back to railing about chemtrails or the evils of vaccines, because basic boring science appears to be on the road to controlling the current outbreak before all of mankind succumbs.

In the meantime, we are at about two weeks into the three week incubation period both for anyone “exposed” by Craig Spencer or for Kaci Hickox (or anyone she “exposed”) to show symptoms. No reports of transmission so far, and the odds of any cases showing up are dropping very rapidly from the already very low levels where they started.

New CDC Hospital Ebola Guidelines Fall Short of WHO Guidance on Personnel Flow

I’m either a lone voice in the wilderness or just another angry old man shouting at clouds on this, but, to me, the issue of personnel flow inside a facility treating a patient for Ebola is critical. Texas Health Presbyterian Dallas got that issue terribly wrong in the case of Thomas Duncan, and now, although they provide very good guidance on the issue of personal protective equipment and its use, new guidelines just released by CDC sadly fall short of correcting the problem I have highlighted.

The issue is simple and can even be explained on a semantic level. If a patient is being treated in an isolation ward, that isolation should apply not only to the patient but also to the staff caring for the patient. As I explained previously, National Nurses United complained that health care workers at Texas Health Presbyterian Dallas treated Duncan and then continued “taking care of other patients”.

Allowing care providers to go back to treating the general patient population after caring for an isolated patient is in direct contradiction to one of the basic recommendations by WHO in a document (pdf) providing guidance for treatment of  hemorrhagic fever (HF, includes Ebola):

Exclusively assign clinical and non-clinical personnel to HF patient care areas.

By exclusively assigning personnel to care of the isolated patient, then the isolation is more complete.

The new CDC guidelines, released on Monday, offer updated recommendations on the types of personal protective equipment (PPE) to be used and how it is to be used. The guidelines also stress the importance of training on effective PPE use prior to beginning treatment of an Ebola patient. Unfortunately, though, the guidelines still leave open the possibility of health care workers moving between the isolation area and the general patient population.

In the preparations before treatment of an Ebola patient commences, the guidelines state:

Identify critical patient care functions and essential healthcare workers for care of Ebola patients, for collection of laboratory specimens, and for management of the environment and waste ahead of time.

And then once treatment begins, we have this:

Identify and isolate the Ebola patient in a single patient room with a closed door and a private bathroom as soon as possible.

Limit the number of healthcare workers who come into contact with the Ebola patient (e.g., avoid short shifts), and restrict non-essential personnel and visitors from the patient care area.

So the facility is advised to identify the “essential” workers who will provide care to an Ebola patient and to limit the number of personnel coming into contact with the patient. And even though the patient is to be in an isolated room, the guidelines still fall short of the WHO measure of calling for the Ebola treatment staff to be exclusively assigned. Precautions for safely removing the PPE are described, but once removed, the workers presumably are free to go back to mixing with the general patient population. Hospitals are cautioned against allowing large numbers of care providers into the room and to avoid “short shifts”, but there still is no recommendation for workers to be exclusively assigned to the isolation area.

The first thing that comes to mind in this regard is to question whether the CDC recommendations fall short of the WHO call for exclusive assignment in order to allow US hospitals avoid the perceived expense of dedicating a handful of personnel to treatment of a single patient. Is the ever-constant push to reduce personnel costs responsible for this difference between CDC and WHO guidelines? In the US healthcare system, it appears once again that MBA’s can carry more weight than MD’s on critical issues.

Texas Hospital Violated Basic Precaution in WHO Ebola Patient Treatment Guidelines

The incompetence of Texas Health Presbyterian Hospital Dallas is staggering. In following today’s rapidly developing story of a second nurse at the hospital now testing positive for Ebola, this passage in the New York Times stands out, where the content of a statement released by National Nurses United is being discussed (emphasis added):

The statement asserted that when Mr. Duncan arrived by ambulance with Ebola symptoms at the hospital’s emergency room on Sept. 28, he “was left for several hours, not in isolation, in an area where other patients were present.” At some point, it said, a nurse supervisor demanded that Mr. Duncan be moved to an isolation unit “but faced resistance from other hospital authorities.”

The nurses who first interacted with Mr. Duncan wore ordinary gowns, three pairs of gloves with no taping around the wrists, and surgical masks with the option of a shield, the statement said.

“The gowns they were given still exposed their necks, the part closest to their face and mouth,” the nurses said. “They also left exposed the majority of their heads and their scrubs from the knees down. Initially they were not even given surgical bootees nor were they advised the number of pairs of gloves to wear.”

The statement said hospital officials allowed nurses who interacted with Mr. Duncan at a time when he was vomiting and had diarrhea to continue their normal duties, “taking care of other patients even though they had not had the proper personal protective equipment while providing care for Mr. Duncan that was later recommended by the C.D.C.”

From the context of both the New York Times article and the nurses’ statement, it seems most likely that this movement of nurses from treating Duncan to treating other patients took place during the period after Duncan was admitted to the hospital and before the positive test result for Ebola was known. However, from the nurses’ statement showing that at least some of the personnel on duty realized Duncan almost certainly had Ebola, proper isolation technique should have been initiated immediately.

And that movement of nurses from a patient who should have been in isolation back into the general patient population is a huge, and obvious, error. Consider this publication (pdf) put out in August by the World Health Organization, summarizing precautions to be taken in care of Ebola patients. The very first page of actual content, even before the section labeled “Introduction”, is a page with the heading “Key messages for infection prevention and control to be applied in health-care settings”. The page lists nine bullet points about dealing with ” hemorrhagic fever (HF) cases” (hemorrhagic fever diseases include Ebola). Here is the third entry on that list:

Exclusively assign clinical and non-clinical personnel to HF patient care areas.

There really is no point in saying a patient is isolated if staff are freely moving back and forth between the isolation area and the general patient population. I’m wondering how long it will be until there is a whole new management team at Texas Health Resources, the parent firm for the hospital.

CDC Director: “Zero Risk” to Passengers on Flights With Dallas Ebola Patient

Although not yet widespread, panic and disinformation are emerging surrounding the case of the first patient to have been diagnosed with Ebola while in the United States. The worst of the hysteria surrounds the fact that Thomas E. Duncan flew from Liberia to the United States on a trip that required 28 hours, ending at the Dallas-Fort Worth airport on September 20. It is known that Duncan was screened in Liberia and found not to have a fever when he boarded the first flight. Persons infected with Ebola but not yet exhibiting symptoms are incapable of spreading the disease, primarily because the disease spreads through direct contact of mucous membranes or open wounds with bodily fluids and symptom-free patients are not yet vomiting or having diarrhea, so no virus-carrying fluids are being produced or at risk of being spread in ways that other people will come into contact with them.

On Tuesday, the Director of the CDC, Dr. Tom Frieden, stated outright that there is zero risk to passengers who were on a flight with Duncan:

A national public health official today said there was “zero risk of transmission” of Ebola on a commercial airline flight that a Dallas patient who has tested positive for the disease flew on from Liberia earlier this month.

Centers for Disease Control and Prevention Director Tom Frieden said today in a live briefing from Atlanta that the person — a male who remained unnamed — showed no symptoms before boarding the plane and was not contagious. The CDC doesn’t “believe there is any risk to anyone who was on the flight at that time,” he said.

Despite Frieden’s clear statement that other passengers face no risk, the press continued to hound CDC and the airlines until Duncan’s itinerary was released. While CBS was moderately responsible in their coverage of the flight information, the Daily Mail asked breathlessly in their headline whether YOU were on a flight with Duncan. Even more incredibly, stocks in US airlines were dumped yesterday in response to the news of Duncan’s flights:

Investors were also selling stocks following news that the first case of Ebola had been diagnosed in the U.S. Investors dumped airline stocks and bought a handful of drug companies working on experimental Ebola treatments.

The story of just how Duncan became infected is a sad one. On September 15 (recall that he left Liberia on the 19th and arrived in Dallas the 20th), Duncan helped neighbors take their 19-year-old daughter to the hospital. Sadly, the hospital was already overwhelmed with patients and she was turned away, only to die early the next morning after returning:

In a pattern often seen here in Monrovia, the Liberian capital, the family of the woman, Marthalene Williams, 19, took her by taxi to a hospital with Mr. Duncan’s help on Sept. 15 after failing to get an ambulance, said her parents, Emmanuel and Amie Williams. She was convulsing and seven months pregnant, they said.

Turned away from a hospital for lack of space in its Ebola treatment ward, the family said it took Ms. Williams back home in the evening, and that she died hours later, around 3 a.m.

Mr. Duncan, who was a family friend and also a tenant in a house owned by the Williams family, rode in the taxi in the front passenger seat while Ms. Williams, her father and her brother, Sonny Boy, shared the back seat, her parents said. Mr. Duncan then helped carry Ms. Williams, who was no longer able to walk, back to the family home that evening, neighbors said.

The hospital in Dallas where Duncan is being treated has received a lot of criticism because he first went there on September 26 but was sent home when only exhibiting a low grade fever:

When Mr. Duncan first arrived at the hospital last Friday, six days after he had arrived in America, he told a nurse that he had come from West Africa. Public health officials have been urging doctors and nurses to be on the alert for Ebola in anyone who has been in Guinea, Liberia or Sierra Leone. But information about Mr. Duncan’s travel was not “fully communicated” to the full medical team, said Dr. Mark Lester, executive vice president of Texas Health Resources, the parent organization that oversees Texas Health Presbyterian Hospital.

As a result, that information was not used in the clinical diagnosis and Mr. Duncan was sent home, with the diagnostic team believing he simply had a low-grade fever from a viral infection, Dr. Lester said.

Those with whom Duncan had contact from the time of the onset of his symptoms until he returned the hospital on September 30 28 (corrected; September 30 was when tests confirmed Ebola after he returned to the hospital on September 28) in much worse condition and was then isolated are being monitored for signs that they may be infected:

Officials said Wednesday that they believed Mr. Duncan came into contact with 12 to 18 people when he was experiencing active symptoms and when the disease was contagious, and that the daily monitoring of those people had not yet shown them to be infected.

The incubation period (the time between exposure to the disease and the onset of symptoms in an infected person) for Ebola varies from 2 to 21 days. Recall that Duncan was exposed on September 15 and visited the hospital for the first time on September 26, so his incubation period was around eleven days. We are now around six days into the time since Duncan first visited the hospital, so those with whom he came into contact will need to be monitored for for another two weeks or so until at least 21 days have passed since their last contact with Duncan.

While there is some chance that one or more of those with whom Duncan had contact while he was contagious will become infected, as long as everyone who was in contact with him during that critical period is under observation now, there is virtually no chance of the disease spreading outside that small group of people. And you can rest assured that nobody from any of the flights Duncan was on will come down with disease from exposure to him.

CDC Modeling Demonstrates Importance of Intervention in Ebola Outbreak

Helpful graphic from WHO illustrating precautions to prevent infection while traveling. Click on image to see a larger version.

Helpful graphic from WHO illustrating precautions to prevent infection while traveling. Click on image to see a larger version.

As the Ebola outbreak in West Africa continues to grow, fresh attention was focused on it yesterday when the CDC announced that in a mathematical model they developed of the outbreak, failing to intervene in spread of the virus could lead to as many as 1.4 million people infected by late January. Somewhat lost in the response to the “wow factor” of a projection of over a million people being infected is that the model also very powerfully demonstrates how the viral outbreak can be contained simply through moderate adoption of the most basic aspects of an infection control program.

First, to review from my previous Ebola post, Ebola is only transmitted when bodily fluids of infected or dead individuals come into contact with broken skin or mucous membranes.

The key to preventing spread of the virus is for those who care for infected patients, whether they are health care workers at a hospital or family members in the home, is preventing contact with fluids from the patient. CDC has prepared an informative guidance document for how health care workers can control the spread of Ebola in their facilities. The key steps are to provide protective clothing to cleaning staff, use an effective disinfectant, avoid re-use of materials with pourous surfaces and dispose (as regulated medical waste) of all textiles, linens, pillows and mattresses that may be contaminated.

Because practices such as these are routinely implemented in US health facilities when patients with high risk infectious diseases are being treated, there is little to no chance of Ebola spreading within the US. As noted in the previous Ebola post, the extreme poverty of the health care systems in the affected countries in Africa is what has allowed the disease to spread, as health care facilities there simply cannot afford the materials they need for implementing safe practices.

Here is the output of the model for Ebola spread in Liberia and Sierra Leone if infection control is not implemented beyond the current level. As noted in the NYTimes article linked above, the current estimate is that 18% of patients in Liberia and 40% of patients in Sierra Leone are treated in facilities that prevent spread of the virus. The model predicts both the number of infected patients in the two countries and the number of beds devoted to care of those patients (“corrected” means that the estimate for number of infected individuals is corrected for the assumption that 2.5 times more patients are infected than have been officially reported):
no intervention

As noted above and widely cited in the press yesterday, if the virus outbreak is left unchecked, the model predicts a cumulative total 1.4 million infected patients in the two countries by January 20 (many of whom are dead by then) and a need for up to 100,000 beds for treatment of these patients.

The good news that is buried in the CDC model is that stopping the virus outbreak does not require implementation of virus control measures for treatment of every infected patient. In the graphs below, we see the output from the model under the assumption that viral control practices start to be implemented now and expand to a level of 70% of infected patients (25% of them in hospitals and 45% in home treatment) being treated under safe practices by December:

Note that the cumulative number of cases levels off between 25,000 and 30,000 and the total number of beds needed peaks at around 13,000  1300 before dropping rapidly.

This model demonstrates very clearly that the highest priority for stopping the Ebola outbreak should be rapid and widespread implementation of basic infection control practices. Spreading this information into homes where patients are being treated is key. Convincing families of the importance of removing infected clothing and bedding seems likely to be the pivotal aspect of the public information campaign. Help from the West will be essential in providing the huge amount of disposable protective clothing and the necessary cleaning and disinfecting supplies. Replacement clothing, linens, mattresses and pillows should be provided as many of the affected families will be hard-pressed to replace these items under the already difficult conditions of an infected family member.

Further good news is that these projections were based on conditions in August and there is reason to believe that the situation may already be getting better. From the Times, again:

The caseload projections are based on data from August, but Dr. Thomas R. Frieden, the C.D.C. director, said the situation appeared to have improved since then because more aid had begun to reach the region.

“My gut feeling is, the actions we’re taking now are going to make that worst-case scenario not come to pass,” Dr. Frieden said in a telephone interview. “But it’s important to understand that it could happen.”

Let’s hope that Dr. Frieden is correct.

No, We Aren’t All Going to Die Because Ebola Patients Are Coming to US for Treatment


Scary, color-enhanced electron micrograph of Ebola virus particles. Creative Commons license courtesy of Thomas W. Geisbert, Boston University School of Medicine.

With the death toll now over 700 in an Ebola outbreak that has been building since February, Americans are suddenly up in arms about the virus, but only because it was announced yesterday that up to two Americans infected with the virus may be transported to Atlanta for treatment. Yes, the virus is especially deadly, with a death rate of 70-90% of infected patients, but the virus does not spread particularly efficiently and is not airborne. Writing at, biologist Laurie Garrett points out a disaster scenario for the virus. Rather than an outbreak in the US, which seems extremely unlikely, Garrett outlines how the virus could spread in the much more densely populated Nigeria rather than the more remote areas of Guinea, Sierra Leone and Liberia where it is now concentrated.

Before getting into the details of the current outbreak and its possible spread to Nigeria, a little background on the virus. From the World Health Organization, we have this information on how the virus spreads:

Ebola is introduced into the human population through close contact with the blood, secretions, organs or other bodily fluids of infected animals. In Africa, infection has been documented through the handling of infected chimpanzees, gorillas, fruit bats, monkeys, forest antelope and porcupines found ill or dead or in the rainforest.

Ebola then spreads in the community through human-to-human transmission, with infection resulting from direct contact (through broken skin or mucous membranes) with the blood, secretions, organs or other bodily fluids of infected people, and indirect contact with environments contaminated with such fluids. Burial ceremonies in which mourners have direct contact with the body of the deceased person can also play a role in the transmission of Ebola. Men who have recovered from the disease can still transmit the virus through their semen for up to 7 weeks after recovery from illness.

Of particular relevance to the two patients who may be transported to Atlanta for treatment (they work for Samaritan’s Purse, an aid organization) and the tragic death of Sheik Umar Khan, Sierra Leone’s top Ebola doctor, the information from WHO continues:

Health-care workers have frequently been infected while treating patients with suspected or confirmed EVD. This has occurred through close contact with patients when infection control precautions are not strictly practiced.

The fact that these health care givers become infected because standard infection control precautions are not strictly practiced in no way should suggest that they are uninformed or careless. Instead, Garrett points out in her article the stark realities facing health care providers in the three countries where the outbreak rages:

To show how ill-equipped these nations are to battle disease, per capita spending on health care, combining personal and governmental, amounts to only $171 a year in Sierra Leone, $88 a year in Liberia and $67 a year in Guinea, according to the Kaiser Foundation.

For those who want more detail on the virus, this succinct summary of the structure of the Filovirus family of viruses and their mode of operation is very informative.

For the panic-motivated hypochondriacs among us, initial symptoms of this virus mimic the onset of most other viral infections.

The most recent update from WHO on the outbreak can be read here. The update summarizes the assistance that is being provided to the countries where the outbreak is ongoing. Significantly, WHO is not advocating travel restrictions at this time.

Returning to Garrett’s article, she points out the factors that would lead to chaos should Ebola spread in Nigeria:

Were Ebola to take hold in that country [Nigeria], spreading from person-to-person in a densely populated, chaotic city such as Lagos, the worldwide response would swiftly spin into uncharted political and global health territory.

Consider the following: Nigerian physicians are on strike nationwide; hundreds of girls have been kidnapped from their schools and villages over the past six months by Boko Haram Islamist militants — and none has been successfully freed from their captors by the Abuja government.

Nigeria is in the midst of national election campaigning. President Goodluck Jonathan’s government is, at best, weak. The nation is torn apart by religious tension, pitting the Muslim north against the Christian south. Islamists in the north have long distrusted Western medicine. They have opposed polio vaccination and have kidnapped and assaulted central government health providers.

Garrett’s plea is for an already-planned African summit on Monday to be used to develop a coordinated plan for dealing with the virus:

One way or another, Obama must take advantage of Monday’s Africa summit to press the case for calm and appropriate responses. These would include specific post-Ebola financial commitments to Liberia, Sierra Leone and Guinea.

The possibility that the epidemic might take hold in Nigeria must be confronted, and plans of action must be considered. The world cannot afford to make decisions in the heat of panic about such things as international airport closures, withdrawal of foreign oil workers, negotiations for outbreak responses with northern imams, hospital and clinic infection control training across thousands of Nigerian health facilities, deployment of international assistance teams for rapid diagnostics and lab assistance and countless other contingencies.

Sadly, Garrett points out important information on the damage that has already been done in this outbreak:

When this Ebola epidemic eventually ends, the health budgets of these nations [Liberia, Sierra Leone and Guinea] will have been bankrupted, and many of their most skilled and courageous physicians, nurses, Red Cross volunteers and hospital workers will have perished.

Let’s hope that Monday sees the beginning of stronger coordination to put more resources where they are needed to halt the spread of this ongoing disaster.