NEWS & EVENTS

Going viral

FEATURE - 28TH JUNE 2017

How Dr Felicity Fitzgerald paused her PhD to help fight against the worst Ebola outbreak in history

Words: Mark Riddaway
Images: Michael Duff / Orlando Gili

Dr Felicity Fitzgerald’s decision to leave the satisfying plod of a lab-based PhD and head into the maelstrom of the world’s worst-ever Ebola outbreak was arrived at overnight. “I just decided one night in September 2014,” she explains, with characteristic breeziness. “It made sense to me: I am a children’s doctor who specialises in infectious diseases, I’d always wanted to do humanitarian work, I was single, I had no kids, I had no dependents. This was what I should be doing. So, off I went.”

Her destination was Freetown, the sprawling capital of Sierra Leone and a city at the heart of the grim epidemic that swept through west Africa between 2013 and 2016: the first time that the deadly and highly contagious Ebola virus had taken a grip of such a densely-populated urban environment. At least 11,310 people were killed by the disease in Sierra Leone, Guinea and Liberia over the course of two and a half years, although this number is almost certainly a gross underestimate. The international response to the crisis was, in retrospect, slow and half-hearted. Felicity Fitzgerald’s most definitely wasn’t.

With the blessing of her supervisors at UCL (“although I did say it would be six weeks, then came back seven months later”), her first step was to volunteer as a clinician through the King’s Sierra Leone Partnership, a British NGO that provides support to the Sierra Leonean health system. But Felicity, then in the second year of a PhD investigating HIV in children, is a researcher to the core, and what she found when she arrived in Freetown persuaded her that her epidemiology skills could be put to good use in forging a better understanding of this infamous but esoteric illness.

Felicity’s subsequent research project, conducted with a skeleton team (including two Sierra Leonean medical students) in the middle of a catastrophic epidemic, has since been written up in two papers. One focused on children who tested positive for Ebola, the second looked at outcomes for children admitted to Ebola units with possible symptoms, but who tested negative. It is no exaggeration to suggest that her research—one of the largest cohort studies of its kind, and one that will be of great significance in any future outbreak—has seen this previously callow doctor transformed into a world expert on Ebola transmission in children: a heart-warming coda to an otherwise relentlessly tragic story.

  • "After a fortnight,I was suddenly the paediatric Ebola expert’’

Mundane initial symptoms
Just the word Ebola gives people the shivers. It sounds like the premise for a schlocky horror film: a disease discovered in rural Africa, carried by bats, passed between humans through blood and bodily fluids, and with a fatality rate of around 50 per cent. Part of the terror it engenders comes from the descent into violent sickness being prefaced by the most mundane of initial symptoms. The long tick-list of early signs—some of which might be present, others not—include a temperature, an upset stomach, a cough. That tight chest and sore throat? It could be a common cold, or it could be the precursor to a swift and quite horrific death.

Was Felicity frightened? “You’d be stupid not to be,” she says. But that fear was at least tempered by a growing understanding of the virus’s transmission. “This was the first time we’d seen Ebola on this scale, and it was becoming increasingly clear that it wasn’t a disease like flu, where you sneeze and the whole tube carriage gets it—you really have to go some to get Ebola,” she explains. “Yes, we’d be at risk, because we were doctors and nurses dealing with sick patients, and the sicker you get with Ebola, the more infectious you are. And then you’re a corpse and you’re really, really infectious. But I knew that I was going to be wearing the most protective equipment possible, and I knew that if I got sick I would probably be air-lifted out.”

Her courage was, she says, dwarfed by that of the local health workers. “We were choosing to go out there. The Sierra Leoneans often hadn’t been paid for months and had been looking after their friends and peers who were dying, knowing they could go the same way, and yet were still coming to work every day and being cheerful and working so, so hard. Nobody was slapping them on the back and saying they were heroes. We all got medals; no one gave them medals. 
It was both humbling and infuriating.”

On her arrival in Freetown, Felicity began working in an Ebola holding unit. It was to these local centres that any patient displaying symptoms of Ebola would be admitted for testing—if the test came back positive, they would be sent on to one of the treatment centres in the major hospitals. With results taking an average of two and half days to come through, and with a backlog of beds at the treatment centres hindering departures, the primary role of the clinicians at the unit was a public health one. “If we tried to cram more people into our holding units, we increased the chances of hospital-acquired infections,” explains Felicity. “If you put people with malaria or appendicitis next to people with Ebola, and they’re all vomiting and having diarrhoea, you need to be really worried about nosocomial transmission of Ebola.” 

  • "It felt like they were all dying, because we could give them so little care. not being able to give them all the things that could usually be provided in a children’s hospital was pretty heart-breaking. But it seems we weren’t doing such a terrible job"

Limited options
Options for clinical care were limited. Ebola patients can lose a vast amount of fluids—10-15 litres of diarrhoea per day—so dehydration is a massive problem, but providing intravenous fluids in an over-crowded, under-staffed unit comes loaded with risks. “Ebola patients get really disturbed and confused, so they often pull out their lines, leave the sharp in the bed, then wander around the unit trailing Ebola blood, which is really, really dangerous for staff and other patients.” As a result, at the peak of the crisis, the most the team could do was offer oral hydration salts and paracetamol, together with malaria medicines and broad-spectrum antibiotics. “We were giving treatment, but we were dubious as to what impact we were having on mortality rates.” 

This uncertainty about the unit’s impact nagged at Felicity: “We had all these questions: what are the nosocomial transmission rates, what are the mortality rates, what modifiable factors are there, what can we do better to limit the number of people dying? Because we had this constant transfer of patients from one place to another, nothing was joined up—you couldn’t know what the normal course of Ebola might be. We would see the patients who died with us, but we wouldn’t see the patients who survived in the treatment centres, and the treatment centres had a survival bias because they missed the really sick people who died with us. If you couldn’t put all that data together, you couldn’t see what Ebola really looked like. But writing things down came so far down everyone’s list of priorities. Collecting patient data was simply not a consideration—we didn’t have the time.”

On a brief return to London for Christmas, Felicity decided to establish a research project to tackle some of these unknowns. “I persuaded my PhD bosses that I needed to come back and set up an epidemiological cohort looking at children with Ebola. I discussed this with Save the Children, who were able to fund and support the work. I also worked with a wonderful academic from the London School of Hygiene and Tropical Medicine, Dr Shunmay Yeung, who supervised me remotely.”

The search for data
The questions she was looking to answer were fairly simple. The hard part was finding enough clean, reliable data to make the study meaningful. “I returned in January, but didn’t officially start collection until March, by which time the peak had passed, so most of my cohort was retrospective rather than prospective,” says Felicity. “I was relying on data that had already been collected.”

In an Ebola holding unit, nothing is simple. A task as seemingly straightforward as getting information about a patient—heart rate, respiratory rate, weight, temperature—from the red zone (the isolation area) to the green zone (where the administration of the unit is carried out) can, in the face of rigid infection control protocols, be rendered herculean. “Some people shouted over the fence, but that takes time and you can mishear things,” says Felicity. “One unit had a Perspex sheet between the zones; you’d hold your notes up against the sheet and someone the other side would copy them down.” More technologically sophisticated methods proved useless due to the high concentration of chlorine used to keep red zones as sterile as possible. “We tried baby monitors and walkie talkies, some people had tablets, but everything would die.”

The chlorine affected the gathering of data, as well as its transfer. “You felt like you were blind, deaf and dumb in there,” explains Felicity. “As a paediatrician, weighing children is so important, because that allows you to measure how dehydrated they are, but the scales would just die. You couldn’t use a stethoscope because of infection prevention control. Even clock batteries would die in a couple of days, so you didn’t have a clock to count someone’s heart rate with. Even taking the pulse of a small, dehydrated child through three layers of gloves is really difficult.” Those records that were successfully collected had to be kept in paper form, as the unreliability of the power supply precluded the use of digital storage. Some of these paper records were prematurely destroyed, others probably deserved to be.

Joining up the records of Ebola-positive patients who were transferred from holding units to treatment centres was a further source of complexity, partly a result of the prevalence in Sierra Leone of a small number of very similar names. “Practically everyone seems to be called Mohamed Sesay or Fatmata Sesay!” laughs Felicity. “Fatmata could be Fatima or Fatumata or Fatma. People’s ages can be just as hard to pin down—although it’s a little bit easier with children. It was often very difficult to know if we have got the matching right.”

Basic outbreak stuff
Until January 2015, there had been no system in place for providing patients with a unique identifier. “You had all these patients at different holding units, all being given a sequential patient number by the unit, then they would move to a treatment centre and would take their patient number with them. You could end up with 11 different patients with the number 67, five of whom probably had the same name! Eventually, the American Center for Disease Control started distributing forms which had a sticker on the top with a genuinely unique identifier, and you could use the sticker on blood samples and in your admissions book. That’s pretty basic outbreak stuff, but for six months it wasn’t being done.”

All this is a long-winded way of saying that patient data for the Sierra Leone Ebola outbreak was, in Felicity’s understated phrasing, “really messy”. To find meaning in the numbers, Felicity needed to wade through the mess, assess the reliability of the record keeping, and find those disparate strands of paperwork that might knit together to tell the story of a child’s journey through the system. This entailed a lot of travel, a lot of frustration and a huge amount of work, all while continuing to assist on the clinical side. But it paid off: by the end of the process Felicity had managed to uncover meaningful data for a cohort of 1,054 children who had been admitted to 11 Ebola holding units in the western area of Sierra Leone during 2014-2015.

After analysing the data, her conclusions were striking. Of the 630 children discharged from the holding units after testing negative for Ebola, only three were known to have been readmitted with the disease: less than 0.5%. This suggested a nosocomial transmission rate vastly lower than anyone would have feared, particularly given the challenges involved in getting sick, distressed children to follow isolation protocols. 

Also encouraging was Felicity’s discovery that Ebola-negative children who passed through the units were no more likely to die from other illnesses than those who were cared for in children’s hospitals in previous years. “It felt like they were all dying, because we could give them so little care. You could be managing a child who you knew deep down had cerebral malaria and not Ebola, and not being able to give the care they needed—intravenous sugars, transfusions, all the things they could usually provide in a children’s hospital—was pretty heart-breaking. But it seems we weren’t doing such a terrible job.” 

Looking back
For the 309 children who tested positive for Ebola, the mortality rate was a distressingly high 57%. Felicity tried to ascertain whether there were any modifiable factors that influenced who died and who didn’t—the presence of a parent, the use of intravenous fluids, the transfer distance between the holding unit and the treatment centre—but nothing could be found. “We know that younger children were more likely to die and that children who had diarrhoea when they presented were more likely to die, but that’s all we can really say.” She was also able to draw out information that will enable a more specific case definition to be drawn up, narrowing down the number of children admitted to holding units in a future outbreak.

Looking back, her big regret is that she missed out on seeing what happened to those children turned away from the holding units at the gate. “When people came to the gate, you’d go through a tick-box list of symptoms,” she explains. “If they ticked three or more, they got admitted. Those pieces of paper where we had some details but the people didn’t get admitted, they just got binned. You missed the chance of finding your false negative rate from that initial test. My data is only for children who screened positive at the gate—we don’t have data for the ones who may have had Ebola but went back into the community. It’s a real shame. I could have done that at the time, said: ‘Let’s not throw that paper away.’”

For this contagiously positive and energetic young doctor (if you’ll forgive the use of the word contagious in this context), the whole experience was highly rewarding, despite the daily trauma. “It was so much fun,” she says. “That might sound bizarre: describing working in an Ebola outbreak as fun. Coming from the NHS, if you’re doing research you’re really pleased if you’ve worked out something that means you’re going to cut down your admission days with asthma by one day a year, or something like that. That’s a really significant finding. But with this Ebola outbreak, there was a complete evidence vacuum. That’s why, having been there for a fortnight, I was suddenly the paediatric Ebola expert. It was terrifying, but also really intoxicating.”

A major accolade
Since returning to the UK, Felicity has presented her findings at the Academy of Medical Sciences, the Marylebone-based fellowship of medical researchers (“It was a wonderful opportunity and I love talking about the subject; I think I’m a bit of a show off”), and has been awarded the Academy’s Lord Leonard and Lady Estelle Wolfson Prize: a major accolade. 

But much as she has enjoyed engaging with the research community in London, she took even more pleasure from a return to Sierra Leone. “One of the bits of the whole project that I’m most pleased about is that I got to present my study to the nurses and doctors and community health officers and cleaners who were working in the units,” she says. “I was able to say, look, the transmission rates were really low. And that is down to you guys doing an incredible job. Your blood, sweat and tears paid off. Children survived because of you.”