Predicting disease severity based on history is pure puffery: So what's a responsible planner to do?

(CIDRAP Business Source Osterholm Briefing) – It's my job to stay current—even anticipate—what the world of pandemic influenza will look like over the next months. But you will never hear me predict severity. Never. You deserve better than that! The severity of past pandemics predicts nothing about the current pandemic. Worse, it distracts our attention from what we must do right now.

If anyone could accurately forecast the onset and severity of pandemic H1N1 cases in your community, business locations, and international supply chains, then preparing for a next wave of H1N1 would surely be made easier. Isn't going to happen—despite some of my professional colleagues' recent attempts to predict that future.

Now it's one thing to overlay (for purposes of illustration) the morbidity and mortality statistics from previous pandemics on today's population to fathom the scope of a pandemic's power, to understand the range of ensuing problems. And in many years of making presentations, I've used such examples; they wake people up, get them thinking, and maybe encourage them to act. But that's as far as we can go. And that's a far cry from forecasting the severity of a completely unpredictable new virus based on the history of others.

Don't get me wrong, influenza experts do need to lay out as clearly as possible evidence of where we see this H1N1 pandemic going so that we can all execute responsible pandemic preparedness plans that are grounded in flexibility. Your guiding principle must remain the following: Expect the unexpected.

Severity is only one part of the equation that determines a pandemic's impact. How widespread the transmission is is another. If you're feeling like a ping-pong ball with all the new reports and predictions about severity coming out, let me help you stop the senseless distraction so you can focus on your priority tasks.

Blundering into the past

To be blunt, any attempt to use past pandemics to predict severity of the current one is pure puffery. Let me explain why.

Just this week, two of my colleagues (and friends) from the National Institutes of Health (NIH) wrote a piece in the Journal of the American Medical Association (JAMA) suggesting that there are assumptions about the course of the 1918 influenza pandemic (also caused by an H1N1 influenza strain) that may be leading to misperceptions of what the current H1N1 has in store for the world.

Researchers, including noted influenza historian John Barry—another friend and colleague—have compelling data showing that soldiers in US military training camps in April and May of 1918 who experienced a relatively mild outbreak of "flu-like illness" were largely protected against the much more serious illness that occurred later in the fall and early winter. We didn't have the capability to isolate viruses back then, so no one will know with absolute certainty that Barry and his co-researchers are right. Furthermore, because illness was mild, finding genetic material from anyone whose deaths were caused by the pandemic virus is virtually impossible.

But based on epidemiologic data, I, for one, think Barry is correct, that there was a mild wave spring in 1918 that preceded the much more serious wave later that year. My colleagues from the NIH disagree with that conclusion and believe that today's planners are assuming that the current H1N1 pandemic will worsen in terms of disease severity this fall, based on what they believe are the faulty conclusions of what happened in 1918.

That's not the end of the story. Two days ago, Helen Branswell, one for the world's leading influenza journalists, wrote a story about the JAMA piece and interviewed other infectious disease experts for their opinion. It's worth a read.

As you might expect, every expert had an answer but no more data to support their cases than my NIH colleagues. One of the experts interviewed for Branswell's piece is convinced that what we'll see during the next wave of H1N1 will look much more like what happened in 1957 with the H2N2 pandemic. In that pandemic, there was no evidence that the virus got more virulent (able to cause more severe disease) during the second wave of illness.

So why do I even mention this JAMA article and get into the nuances of what influenza experts are predicting? Because both the JAMA and Branswell pieces are circulating in the preparedness world, leading many to believe that we are capable of predicting what will happen in the upcoming weeks. Nothing could be further from the truth. I see two reasons why you should interpret with great caution the JAMA report and the comments of the experts in Branswell's piece.

Reason 1: The virus itself is unpredictable

Nobody really knows what is going to happen when this next wave of H1N1 is overlaid on the traditional "enhanced influenza transmission period" of our northern hemisphere winter season.

Yes, the current pandemic strain of H1N1 could either mutate or reassort (ie, mix genes with an existing influenza virus) and change in its ability to cause serious disease. But remember that means it could become either more severe or less severe. We just don't know.

Please—view with skepticism anyone who tells you that he or she knows what this virus will do when it comes to severity. And if someone tries to sell you a previous pandemic occurrence as the model for what will happen this year, say "thank you" and move on. We can view pandemic history with interest. I sure do. But would I use that history to predict whether or not today's pandemic will be a model that differs from or resembles what we saw in 1889, 1918, 1957, or 1968? No. I know I can't give you a definitive answer. What I can tell you is that not one of the pandemics I just mentioned was similar to any of the others.

Reason 2: It's the economy, my friends

The current pandemic is the first to occur in a world with a global just-in-time economy. We have precious little to no surge capacity for many of the critical products and services we count on every day, including many life-saving medical supplies and drugs. What's more, as part of the global nature of our new economy, most of the products or their key components originate in Asia. And their supply chains are long and fragile.

Remember: We do know that pandemic H1N1 is disproportionately hitting people younger than 50 years old, regardless of where they live. Unfortunately, neither China nor India will have much H1N1 vaccine to administer to citizens, including their workforces. Even if the current H1N1 influenza virus changes little over the next months, just consider the sheer numbers of people who'll be affected by this easily transmitted virus. We do know we're going to see greater transmission. The very real possibility of substantial illness among workers and their family members and resulting absenteeism for several months is staggering.

Bear in mind I'm not even talking about severity.

Bottom line for your organization

Don't get caught up in the esoteric debate of some academic flu experts about what is going to happen over the next 6 months. Trust me, they don't know. I promise to alert you as soon as information that does merit your attention becomes known.

So here's our imperative: We must plan as if things could be worse, even if no more than 30% of our population under age 55 will get sick with H1N1 and limit our workforce for weeks. Such a scenario doesn't turn on the virus's changing. Also, don't forget the backdrop against which this pandemic is unfolding—the global just-in-time economy. None of us have any idea how this factor will change what happens with H1N1 infection in our communities.

In the end, this fact may be the wild card my virologist friends know little about. But you must consider it in your organization's preparedness plans.

Newsletter Sign-up

Get CIDRAP news and other free newsletters.

Sign up now»


Unrestricted financial support provided by

Bentson Foundation Unorthodox Philanthropy logo and text 'Leading Underwriter'3M logoGilead 
Grant support for ASP provided by


  Become an underwriter»