Something Else Altogether — Kerbside Consult 20/2026

A reasonable diagnosis

It is easy, twenty-seven years later, to ask how Nipah could have been mistaken for Japanese encephalitis.

That is the wrong question.

The first cases appeared around pig farms near Ipoh in Perak in late 1998. JE was endemic in Malaysia. Pigs were recognised amplifying hosts. Encephalitis around pig farms therefore fitted an established disease remarkably well.

On 16 November 1998, the Malaysian Doctors Only Bulletin Board System — DOBBS — carried a request from colleagues in Ipoh for help managing what was described as Japanese B encephalitis.

Doctors exchanged information. WHO material was circulated. Vaccine schedules were discussed. Prof Lam Sai Kit of the University of Malaya offered assistance from his WHO collaborating centre.

But there was already uncertainty.

On 26 November, Lam wrote that laboratory confirmation in the fatal cases had "not been conclusive."

A reasonable working diagnosis had acquired its first uncomfortable exception.

What JE should have looked like

Around this period, Prof Lam wrote an overview of Japanese encephalitis for Berita MMA. Reading it today provides an unusually useful benchmark.

Historically, Malaysian JE predominantly affected children. Pigs were important amplifying hosts, but they did not develop the severe respiratory and neurological illness later being reported from affected farms. Transmission to humans was through Culex mosquitoes.

The response therefore followed logically from the diagnosis: control the mosquitoes and vaccinate people at risk.

But as the outbreak moved into Negeri Sembilan, the clinical picture became increasingly difficult to explain.

At University Hospital, Kuala Lumpur, Prof C.T. Tan, Assoc Prof Adeeba Kamarulzaman, Dr Goh Khean Jin and their colleagues saw their first patient from the outbreak on 26 February 1999.

Even before confirmation of a new virus, they recognised that something did not fit.

The patients were too old. Many had been vaccinated against JE. Cases clustered among people who worked directly with pigs. Sometimes members of a family who handled the animals became ill while someone who remained at home did not.

And the patients were reporting something else. Their pigs were sick. Some were coughing. Some developed neurological illness. Some died.

The UH clinicians put the contradiction rather more simply:

"Pigs are not sick in JE."

Sometimes the observation that breaks a diagnosis is not sophisticated. It is simply the fact that should not be there.

"Something else altogether?"

By 9 March, Malaysian doctors were openly questioning the diagnosis on DOBBS.

One doctor listed four concerns: vaccinated patients were still being admitted; older people were affected; many patients were negative for JE IgM in blood and cerebrospinal fluid; and pigs themselves were dying.

Then came the question. Were we dealing with a new strain of the virus —

"or something else altogether?"

It was.

I wonder if anyone is doing an epidemiology study on all cases with suitable controls.....the answer is there....

Two days later, Prof Lam addressed another important question. Why were vaccinated people developing encephalitis? Before concluding that the JE vaccine had failed, he argued, we first had to prove that JE was actually causing those deaths.

That remains good diagnostic reasoning. A working diagnosis is necessary. It is not a conclusion.

When the diagnosis determines what you do

The distinction was not academic.

If the disease was JE, mosquito control and vaccination made sense. But if humans were acquiring a new infection directly from pigs, the response had to be very different.

There was another concern being discussed at the time. During the JE response, pigs were being vaccinated, and questions were raised on DOBBS and Cybermed about whether needles reused from animal to animal might mechanically transmit the newly recognised infection.

Whether reused needles actually contributed significantly to Nipah transmission was never established. Contemporary veterinary evidence more clearly supported rapid pig-to-pig spread through pharyngeal and bronchial secretions, while movement of infected pigs helped carry the outbreak south.

The larger lesson remains: an intervention that makes sense for one diagnosis may become irrelevant — or even potentially harmful — when the diagnosis is wrong.

The diagnosis begins to break

On 17 March, notes from a University Hospital symposium circulated on DOBBS.

Thirty patients were described. Twenty-seven were men. Their mean age was 33.9 years.

The epidemiological anomalies could now be reduced to three observations: Adults. Pig handlers. Sick pigs.

The laboratory evidence was also becoming increasingly difficult to reconcile with JE. Some early results did support JE, and genuine JE infection may have been present in a minority of patients. But JE could no longer explain the outbreak as a whole.

When specimens from 13 patients were subsequently examined at the CDC, only one showed evidence of recent JE infection. The other 12 had evidence of infection with the newly recognised Hendra-like virus.

The laboratory wasn't necessarily wrong. The interpretation had been incomplete.

A new virus had been isolated from cerebrospinal fluid in Vero cell culture. The infected cells fused together, producing striking syncytia. Electron microscopy showed something morphologically very different from Japanese encephalitis virus. It appeared to be a paramyxovirus.

One DOBBS response captured the moment rather less cautiously:

It appears that we were vaccinating against the wrong virus!

The virus that wasn't JE

At the centre of the breakthrough was Dr Chua Kaw Bing, working within the Department of Medical Microbiology, University of Malaya — led by Prof Lam Sai Kit.

In early March 1999, Chua and the University of Malaya team isolated a previously unknown virus from the cerebrospinal fluid of encephalitis patients. The virus produced striking syncytia in Vero cells. Electron microscopy showed the morphology of a paramyxovirus. It reacted with antibodies against Hendra virus — but it was not Hendra.

Malaysia had discovered a new virus.

Prof Lam Sai Kit and the wider University of Malaya team were closely involved, while subsequent identification and characterisation brought Malaysian investigators together with scientists from the CDC and Australia.

I was in direct discussion with Chua and Lam while this was happening. Shortly before 29 March 1999, they gave me an electron micrograph of the newly isolated virus for publication on Vads Corner/Cybermed and in the Malaysian Medical Journal.

Their fact sheet on my Cybermed page, dated 29 March, was still cautiously titled: "Factsheets about a new Paramyxovirus isolated during the viral encephalitis outbreak in Negri Sembilan."

There was no long history attached to that image then. We were looking at a virus that had only just been discovered.

Electron micrograph of the newly isolated Nipah paramyxovirus, March 1999. Department of Medical Microbiology, University of Malaya. — **The virus that wasn't JE — March 1999.** Electron micrograph of the newly isolated Malaysian paramyxovirus, supplied to me by Dr Chua Kaw Bing and Prof Lam Sai Kit shortly before 29 March 1999 for publication on Vads Corner/Cybermed and in the *Malaysian Medical Journal*. Image: Department of Medical Microbiology, University of Malaya; courtesy of Dr Chua Kaw Bing and Prof Lam Sai Kit.

To the best of my recollection, this was the first published image of the new virus anywhere in the world.

At first it was described as a new Malaysian paramyxovirus, or a "Hendra-like virus." On 10 April 1999, the isolate was officially called Nipah virus, named after the village associated with the patient from whom the virus had been isolated.

Nipah virus.

The pigs point the way

Singapore supplied an important independent clue.

Eleven abattoir workers developed Nipah infection after handling pigs imported from Malaysia. One died.

Different country. Same disease. Same exposure. The pigs.

Subsequent Malaysian clinical work confirmed what the UH doctors had been seeing. In a series of 94 patients with Nipah encephalitis, the mean age was 37 years. Ninety-three per cent had direct contact with pigs, 74% had been immunised against JE, and 32% died.

Veterinary investigations completed the picture. The pigs themselves were suffering a newly recognised respiratory and neurological disease. The clue in the pigsty was explaining the patient in the ICU.

Today we would call this One Health: clinicians, veterinarians, epidemiologists and laboratory scientists examining different parts of the same event. In 1999 the terminology mattered less than the principle.

No single discipline held the whole answer.

Finding the virus wasn't the end

Giving the virus a name did not answer every question. Where had it come from? How had it entered pigs? Which animals carried it? Could humans transmit it to one another? How far had it spread? And what did a positive antibody test in an animal actually mean?

By 20 May, the contemporary Ministry of Health figure I recorded in Cybermed was 101 deaths. By the end of the outbreak, the figure most widely cited in the scientific literature was 265 cases of encephalitis and 105 deaths in Malaysia. Around 1.1 million pigs were destroyed as the outbreak was brought under control.

Later Malaysian surveillance reports used somewhat different totals when JE and dual-positive classifications were included — another reminder that even case definitions and attribution continued to evolve.

Behind the numbers were families, farms, livelihoods and communities.

A positive test is a finding. What that finding means is a different question.

The virus had changed. The diagnostic problem had not.

A year later: Nipah antibodies did not necessarily mean active Nipah infection.

Early in the outbreak: JE antibodies did not necessarily mean JE explained the outbreak.

A positive antibody test could show that an animal had encountered the virus. It did not, by itself, establish when that infection occurred or whether the animal was currently infectious.

In June 2000, pigs in Perak and Sarawak were again being investigated for Nipah antibodies, with the possibility of further culling. On Cybermed I returned to a question that had troubled the outbreak from the beginning: what exactly did a positive test mean?

A year later, the same problem resurfaced.

The diagnostic lesson also continued after Nipah had been identified.

From the pigs to the bats

By May 1999, fruit bats were becoming the leading suspect as the natural reservoir. Antibodies had been found in Pteropus bats. But finding antibodies was not the same as finding the virus itself. The distinction mattered.

Chua did not stop with discovering the human virus. He and his Malaysian colleagues went looking for where it had come from. They collected urine beneath colonies of Island flying foxes and swabbed fruit partially eaten by the bats.

They eventually obtained three Nipah virus isolates from Malaysian Island flying foxes, Pteropus hypomelanus — two from bat urine and one from a swab of partially eaten fruit. Genetic analysis confirmed the virus.

The trail that had begun with a pig farmer's cerebrospinal fluid had led back to the bats.

Our understanding of transmission evolved too. During the Malaysian outbreak, infection was overwhelmingly associated with infected pigs and there was no convincing evidence of sustained person-to-person transmission. Later outbreaks in Bangladesh and India demonstrated that Nipah can indeed spread between people.

That is how knowledge of a new disease develops. A newly discovered pathogen does not arrive with its textbook already written. The textbook is written afterwards.

The Malaysian contribution

There is another part of this story I particularly want remembered.

In June 1999, I worried in Cybermed that a disease discovered in Malaysia might eventually be described mainly by scientists elsewhere.

It would be a shame if we see articles after articles on Nipah virus outbreak coming from CDC and not from our own people.

Fortunately, I was wrong.

By June 2000, Cybermed was already pointing readers to Malaysian-led work appearing in The Lancet, the New England Journal of Medicine, Science and the Journal of Infectious Diseases.

Chua Kaw Bing led the landmark Lancet report describing fatal encephalitis among Malaysian pig farmers and the previously undescribed Hendra-related paramyxovirus. He subsequently led the major Science paper characterising the virus.

Malaysian clinicians documented its clinical syndrome. Malaysian pathologists described its characteristic vasculitis and microinfarction. Malaysian veterinary investigators characterised the disease in pigs. Chua and his colleagues later isolated the virus from Malaysian fruit bats.

International collaboration — with the CDC, Australian laboratories and others — was crucial. But the discovery itself happened here.

Nipah was a Malaysian discovery.

Medicine thinking aloud

Reading the old DOBBS and Cybermed pages twenty-seven years later is revealing for another reason.

There was no WhatsApp. No social media. No smartphone in every doctor's pocket. Yet Malaysian doctors were already discussing an unfolding epidemic online.

A doctor in Ipoh could ask colleagues for help. WHO information could be circulated. Clinicians could report unusual observations. Symposium notes could be shared. Doctors could question whether the diagnosis was wrong before the scientific papers had even been written.

Those exchanges capture something polished journal articles rarely can: medicine thinking aloud while the answer was still unknown.

There was uncertainty. There was speculation. Some ideas proved correct and others did not. That wasn't a failure of medicine. That was medicine working as it should.

"There is a lesson to be learnt…"

By October 1999, the peer-reviewed literature had begun to catch up with what Malaysian clinicians, veterinarians and scientists had been observing.

Chua and colleagues reported in The Lancet the new virus isolated from Malaysian pig farmers with fatal encephalitis. The pathology demonstrated widespread vasculitis and microinfarction. Singapore investigators documented Nipah infection among abattoir workers exposed to Malaysian pigs.

In Cybermed I wrote:

There is a lesson to be learnt ....................

I left the sentence unfinished. Twenty-seven years later, perhaps it is time to finish it.

The lesson is not that Japanese encephalitis was a foolish diagnosis. It wasn't. Medicine needs working diagnoses. Without them, we cannot investigate, treat or act.

The lesson is knowing when a reasonable diagnosis has accumulated enough unreasonable exceptions that it must be questioned.

Adults instead of children. Pig handlers rather than neighbours. Illness despite JE vaccination. Sick and dying pigs. Laboratory evidence that did not fit. And finally, a different virus growing in a Malaysian laboratory.

Back in March 1999, while the deaths were still occurring, I asked:

Did we ask questions when the deaths were increasing?

Twenty-seven years later, perhaps that remains the most important question.

Nipah taught Malaysia many things about bats, pigs, viruses, laboratories and emerging infections. But perhaps its most enduring lesson was simpler.

WHEN THE FACTS STOP FITTING THE DIAGNOSIS, QUESTION THE DIAGNOSIS — NOT THE FACTS.

Sources & further reading

Historical sources — Vads Corner/Cybermed, 1998–2000

Scientific and authoritative sources

Published 20/2026 · 17 September 2026 · No corrections to date · Corrections policy