Brooklyn measles outbreak of 2013: A case study of the cost of antivaccine pseudoscience

A month and a half ago, I discussed a study showing where the “hotspots” of vaccine refusal leading to low vaccine uptake are in the US. As I said at the time, these are the places where antivaxers will be most likely to “make measles great again” sometime in the near future. My personal prediction is that the most likely state to suffer a big measles outbreak soon is Texas, given how antivaxers have co-opted conservative, libertarian, anti-regulation politics and portrayed the right to refuse vaccines as a matter of personal freedom and parental rights, thus politicizing what was previously an issue with strong bipartisan support, an issue that used to be as close to completely nonpartisan as there could be. Wherever the next outbreak appears, we’ve had multiple outbreaks of vaccine-preventable diseases over the last several years, most recently the outbreak among Somali immigrants in Minnesota, which was fueled by antivaccine godfather Andrew Wakefield and his white, affluent antivaccine followers spreading their message among the Somalis. What I haven’t discussed that much is the public health cost of these outbreaks, which brings me to a recent study of the public health impact of the Brooklyn measles outbreak of 2013 published earlier this week in JAMA Pediatrics. The study shows how even a relatively “small” outbreak sparked by one person can cost substantial resources.

The study, not surprisingly, sparked a number of news stories, such as these stories in the New York Post, Reuters, Healio, and MedPage Today. The outbreak began in March 2013, when a single infected child returned to New York from a trip to London. Of course, at this point, I can’t help but note that London and the UK have long suffered inadequate uptake of the measles-mumps-rubella (MMR) vaccine, thanks to the small case series Andrew Wakefield published in The Lancet in 1998, which was later shown by Brian Deer to have been fraudulent. Ultimately, then Dr. Wakefield lost his license to practice in the UK (was “struck off” the medical register, in the parlance of the UK’s General Medical Council), after which he moved to the US, took up residence in Austin, Texas, where he ran a quack clinic for autistic children; that is, until he became too toxic even for the antivaxers running the clinic. Since then, he’s been falling into even more disrepute, even going so low as to speak on a cruise for conspiracy nuts. Of late, however, he made an unfortunately successful “documentary” (successful at promoting antivaccine views, not at making a lot of money) VAXXED and was last seen canoodling with supermodel Elle Macpherson after having separated from his long-suffering and dedicated wife Carmel. (What I can’t figure out is where he got $1.4 million to buy Carmel a “dream house” before leaving her to “find himself” and get a makeover by a supermodel.) Be that as it may, he’s arguably more responsible for more suffering from measles than any single person in the world.

But back to the Brooklyn measles outbreak and the study by Jennifer Rosen et al from the New York City Department of Health and Mental Hygiene and the National Center for Immunization and Respiratory Diseases, Centers for Disease Control and Prevention. Basically, Rosen et all performed an epidemiological assessment and cost analysis of the Brooklyn measles outbreak.

So how did the outbreak begin? I mentioned that an unvaccinated child returned from London on March 13, 2013. Thanks to Wakefield, London is a place where this adolescent could easily have caught the measles, and that’s what happened. Worse, the outbreak started in two Brooklyn neighborhoods, Williamsburg and Borough Park, each of which is home to large Orthodox Jewish communities. In fact, every single case was a member of a Brooklyn Orthodox Jewish community:

From March 13, 2013, through June 9, 2013, a total of 58 case patients with measles were detected in 2 neighborhoods of Brooklyn, New York, with 6 generations of transmission (Figure). The index case patient presented to a health care professional with symptoms of measles after returning from a trip to London. The case was reported to the New York City DOHMH by a commercial laboratory only once the measles IgM test results had returned positive, 8 days after the clinician initially considered a diagnosis of measles. Two additional case patients resided outside of New York City and were excluded from analyses. Orthodox Jewish persons accounted for 100% of the case patients.

Surprise! Surprise! All but one of the people who developed the measles were unvaccinated:

The median age of the case patients was 3 years (range, 0-32 years). None of the case patients had documentation of receipt of measles-containing vaccine prior to their presumed exposure to measles virus. Of the 58 case patients, 45 (78%) were at least 12 months old and were unvaccinated owing to parental refusal or intentional delay. Twelve case patients (21%) were less than 12 months old and therefore were too young for routine immunization with measles-containing vaccine. One case patient (1%) was an adult who verbally reported a history of receiving measles-containing vaccine as a child; although no vaccination documentation was available, a high IgG avidity test result obtained at the CDC was consistent with a prior history of vaccination.

This is about as clear-cut as it gets; there isn’t even wiggle room for antivaxers to invoke their misleading trope that there were “more vaccinated than unvaccinated children affected” while ignoring attack rates (i.e., the actual risk of being affected as a percentage of the vaccinated or unvaccinated population, which always shows the unvaccinated to be a a much higher risk of developing the measles during an outbreak).

As you might expect, there were complications, including one case of pneumonia and one miscarriage. I’d say that Brooklyn was lucky, because usually the complication of pneumonia usually occurs at a higher than one in 58. It’s usually more like one in twenty.

Now here’s where the real work and expense comes in from an outbreak like this:

A total of 3,351 exposed contacts were identified, excluding the 58 case patients who developed measles. Among those contacts, 2,214 (66%) had evidence of immunity to measles based on receipt of 2 documented doses of measles-containing vaccine, having a positive measles IgG titer, or birth before 1957; 376 contacts (11%) had received 1 dose of measles-containing vaccine; 335 contacts (10%) were susceptible; and immunity status was unknown for 426 contacts (13%). The MMR vaccine was administered within 3 days of initial exposure to 114 contacts who were 6 months or older, and immunoglobulin was administered within 6 days of initial exposure to 77 infants younger than 6 months or to infants aged 6 to 11 months who had not received MMR prophylaxis. Immunoglobulin was not administered to immunocompromised or pregnant persons.

That was a lot of work to track down all those contacts, figure out who was at risk, administer MMR prophylaxis, including vaccine and immunoglobulin, as appropriate. The total cost came to:

Total direct costs were $394 448 ($62,102 for inputs and $332,346 on compensated personnel time) (Table 3). The incremental cost was $73,135, or 19% of the total direct costs. Salaries and fringe benefits combined accounted for $321,313 (97% of personnel costs). Overtime pay totaled $11,033 (3% of personnel costs). The majority (99%) of the cost of inputs was attributed to advertising ($29,425), MMR vaccine ($17,590), laboratory supplies and testing ($9,316), and courier service ($4,886).

To be honest, I was surprised that it wasn’t more. Similar cost analyses performed after the Somali measles outbreak in Minnesotal finally burned itself out last year estimated a cost of $1.3 million to contain the outbreak, with $900,000 coming from the Minnesota Department of Health and the rest coming from Hennepin County, which dispatched 89 staff members to interview affected families at a cost of about $400,000. During the height of the outbreak, health officials were asking for $5 million for an emergency fund to deal with the outbreak, as well as other infectious diseases, such as Zika and syphilis. Of course, the Minnesota measles outbreak was larger, sickening more people (79), sending 22 to the hospital. (Also, the Brooklyn measles outbreak only sent three people to the hospital, and in 31% of cases no medical care was sought.) but it wasn’t that much larger. I’m just speculating, but likely the reason for the more than three-fold difference in cost between Minnesota and New York included the much tighter geographic area in Brooklyn where the cases occurred and needed to be investigated compared to Hennepin County and the added expenses of investigating measles cases among a population of immigrants.

The authors did note that dealing with outbreaks is expensive. The Brooklyn outbreak was one of the largest in the last few decades and was resource-intensive to deal with. It was, however, within the cost estimates in the literature: $2,685 to $22,375 for small outbreaks, from $47,732 to $208,829 for medium outbreaks, and from $280,829 to $1,640,789 for large outbreaks.

Also, Minnesota undertook a massive immunization program, while in the surrounding communities in Brooklyn, MMR uptake was very high, with New York’s five boroughs having an overall MMR uptake rate of 97%. If you want to see an example of how effective herd immunity can be, just take a look at this outbreak. It raged on among several families in the Orthodox Jewish communities in Brooklyn with low MMR uptake but outside of those communities:

The report also displays the effectiveness of a strong vaccination program, said Jason Schwartz, an assistant professor with the Yale School of Public Health.
New York City has a measles/mumps/rubella (MMR) vaccination rate near 97 percent, Schwartz explained. This kept the outbreak confined to a group of people who apparently eschewed the vaccine for religious reasons.

“Even in the largest city in the world, in densely populated neighborhoods in Brooklyn, those individuals who were vaccinated around these clusters were unscathed,” Schwartz said.

The rapid public health response also limited the spread:

But 335 exposures occurred in people completely unprotected against measles, officials found.

About 114 unvaccinated children and adults who’d been exposed to measles were given an MMR vaccine shot to ward off infection, the report said. There were 77 infants too young for immunization who received a dose of measles antibodies, Zucker added.

These post-exposure vaccinations kept the measles case count from being even higher, Zucker said.

This case study of the public health response to a measles outbreak illustrates three main points. First, it shows how easily measles can tear through an unvaccinated population. As mentioned before, these particular families of Orthodox Jews apparently eschewed vaccines for religious reasons, even though Jewish rabbis and leaders emphasize that there is nothing in the Jewish faith that requires foregoing vaccines and thus urge their fellow Jews to have their children vaccinated. Second, it emphasizes how important high levels of vaccination are. If the surrounding New York City environs hadn’t had such a high MMR uptake rate, well over the level needed for herd immunity, it is very likely that this outbreak would have sickened many more children. If this were Romania or countries in Europe with low MMR uptake, the results could have been disastrous. We already see this in the massive measles outbreaks that are ongoing in countries with low MMR uptake. Finally, this study portrays yet another example of why a robust public health infrastructure is so important for a rapid and effective response to outbreaks like the ones in Brooklyn, Minnesota, and Disneyland. Unfortunately, unless we can find a more effective means of dealing with antivaccine misinformation and vaccine hesitancy and refusal, our public health officials and frontline doctors, nurses, and other professionals will continue to have too much work too do.