Vaccination at the National Institute of Pediatrics. Image via Sanofi Pasteur.

Experts are calling for a broader view of how to measure the value of vaccination to public health. Contrary to popular belief, vaccines do not need to be 100% effective to significantly reduce the public health burden of major diseases.

In the 20th century, vaccines against common and highly infectious diseases such as measles were developed – saving millions of lives every year.

However, while the research that led to those vaccines was undoubtedly complex and ground-breaking, diseases such as measles were low-hanging fruit compared to dengue, malaria, and other vector-borne illness.

Impossible task?

Indeed, the challenge was so great that some considered it technically impossible to invent vaccines against them, prompting public health authorities to focus on mosquito control measures.

But they were wrong. Science has risen to the challenge and developed vaccines that protect against these devastating diseases.

The trouble is that while measles and hepatitis A vaccines protect almost everyone who receives their shots, many other vaccines do not achieve such high efficacy.

We accept success rates for cancer medication, for example, that are well below 50% but past experience has raised our expectations of vaccination.

With the arrival of the first dengue vaccine – and as scientists work on malaria and Zika vaccines – some epidemiologists and economists want a rethink of how vaccine impact is evaluated.

If a disease causes tens of millions of deaths, isn’t a modestly effective vaccine going to have a huge impact on the contribution of disease to reduced public health?

To borrow an extreme analogy, would you rather have 10% of Bill Gates’ fortune or 100% of your own?

Measuring impact

Vaccine efficacy (VE): the percentage reduction of disease in a vaccinated group of people compared to an unvaccinated group

Vaccine-preventable disease incidence (VPDI): the reduction in disease rates that can be achieved by vaccine in a given context; higher is better.

Number needed to vaccinate (NNV): the number of people that need to be vaccinated to avoid a case of disease; lower is better.

The public health value of the dengue vaccine – beyond efficacy

Ask a public health expert or a Health Minister in a low or middle-income country what they expect from a new vaccine and the answer is clear: they want to reduce the burden of disease in their country.

The number of cases of disease – ranging from mild to severe – and the number that can be prevented by vaccination are relevant pieces of information. But what decision-makers with limited budgets really want to know is how many severe or debilitating cases will be avoided and how this will benefit the health system as a whole.

Vaccine efficacy is an important factor but so too are the numbers of deaths, severe clinical cases, hospitalization and outpatient visits, and lost working or school days.

Health ministers are looking at the bottom line but this, says Dr. Bradford Gessner of the French non-profit Agence de Médecine Préventive (AMP), is not the kind of information that clinical trials present. These studies are designed to get through regulatory approval and are focused on demonstrating vaccine efficacy and safety only.

Disease burden

In a new research paper co-authored with Annelies Wilder-Smith of Lee Kong Chian School of Medicine in Singapore, Dr. Gessner proposes a broader view of assessing the value of new vaccines.

“Finding the true value of a vaccine is complicated,” Dr. Gessner told Break Dengue. “If dengue fever were in the US at the levels seen in, for example, Brazil the vaccine would immediately be widely used. Dengue, malaria, and typhoid are in lower and middle-income countries where decisions must be made based on impact and so burden of disease matters a great deal.”

Dr Gessner and Dr Wilder-Smith propose a more nuanced way of measuring this impact that goes beyond vaccine efficacy. Vaccine-preventable disease incidence (VPDI) is a way to measure the reduction in disease rates (or incidence) achievable by vaccine (higher is better), while calculating the number needed to vaccinate (NNV) to prevent a single case of disease (lower is better) is another way to show the potential for vaccines to reduce the burden of disease.

“VPDI takes efficacy and applies it to the background incidence of disease as measured when a vaccine is not used. It can help decision-makers look at the burden reduction that a vaccine can deliver,” says Dr. Gessner. His latest study shows that the dengue vaccine has favorable VPDI and NNV compared to vaccines currently used in national publicly funded immunization programs in Latin America and Asia.
For example, all national immunization programs now include Hib vaccine, in large part to prevent Hib pneumonia. In Asia, the VPDI for Hib vaccine against pneumonia is similar to that for dengue vaccine against severe hospitalized dengue while in Latin America it is similar to the VPDI for all hospitalized dengue cases.

Another example: “The pneumococcal conjugate vaccine reduces all-cause severe pneumonia by 10-15%. But pneumonia is very common and the impact of severe diseases is immense so reducing it by this amount is a great investment.”

Dr. Gessner and Prof Wilder-Smith argue a similar case for dengue fever. The first dengue vaccine, Dengvaxia (CYD-TDV) by Sanofi Pasteur, has been registered for use in individuals 9-45 years of age living in endemic areas.

The vaccine works around 65% of the time against a very common illness, and is even more effective against severe forms of the disease, and for this reason has high public health value in dengue endemic populations. “The vaccine has a higher impact in populations with a high dengue disease burden,” adds Prof Wilder-Smith who is Director of Global Health and Vaccinology at her university.

Big picture

Preventing cases of disease is a great outcome for an immunization program but this doesn’t do justice to the true impact of public health measures.

“There are all sorts of ways to measure public health value beyond just preventing a case of illness in an individual,” explains Dr. Gessner. “You should also factor in the decrease in disease transmission and a reduction in unpredictable impacts on the health system.”

Diseases such as pneumonia or malaria are present at a high but fairly steady – rate in countries where they are endemic. While these diseases have a devastating effect on those infected, the health system is relatively well-equipped to deal it.

“Systems adapt,” Dr. Gessner says. “People are put in place and the capacity is there to deal with the high but constant rate of disease. But diseases that cause unpredictable outbreaks or epidemics can be very disruptive.”

He points to the Ebola epidemic where the shock to the health system had immeasurable knock-on effects on people not directly affected by the outbreak.

“The West African Ebola outbreak undoubtedly caused many deaths by disrupting the health system. Its biggest impact may well have been not indirectly killing people infected by the virus but by disrupting care and thus increasing deaths from vaccine preventable diseases through the reduction in immunization coverage, maternal deaths during childbirth through reductions in antenatal care, malaria deaths by disrupting malaria prevention programs, and so on.”

That is why Gessner describes measles vaccines as “probably the greatest investment” available to developing countries: it is a dangerous disease, measles outbreaks disrupt health systems and prevention is inexpensive. “While dengue has a less dramatic impact on mortality, dengue and measles otherwise share many of the same characteristics of epidemic-prone diseases,” he says.

Better data

Dr. Gessner says that future vaccine clinical trials will need to be conducted – and their results presented – in ways that are meaningful to public health decision-makers. “Clinical trials were not designed to show Ministers of Health how to maximize public health,” he says.

He would like to see trials that look at the indirect effects of vaccination. For example, immunization programs for Hib, rotavirus and measles have all been linked to less disease even in unvaccinated people by reducing the circulation of the infecting agent.

Individually randomized trials – the gold standard in clinical research – may not be up to the task. Dr. Gessner would like to see more ‘cluster randomized trials’ where entire villages or neighborhoods are vaccinated en masse and compared to similar villages or neighborhoods that do not receive the vaccine. That way, he says, scientists could look at the direct and indirect impact of mass immunization.

Some of these issues were already creeping up the public health agenda at WHO level in response to the Ebola outbreak and are now being applied to handling the Zika outbreak. The approval of the dengue vaccine in several countries adds further momentum.

A new era awaits

With dengue fever continuing to impose an enormous burden on countries around the world the arrival of a new vaccine against the disease is accelerating a shift in how epidemiologists, economists, and policymakers think.

The relatively blunt ways used to value measles and pertussis vaccines decades ago are being replaced by a more sophisticated approach that puts the various dimensions of public health burden firmly in the spotlight.