Ebola Screening at Airports is Feel-Good Technology, Not Science

October 14, 2014

Monday, October 13, 2014

Ebola Screening at Airports is Feel-Good Technology, Not Science

Ebola_Virus_TEM_PHIL_1832_lores

By Nancy Thorner & Ed Ingold – 

As reported on Sunday, October 12, Ebola has spread in the U.S. The second American case of the deadly Ebola virus was confirmed in tests on a health care worker who treated the first individual stricken with Ebola, Thomas Eric Duncan, who died on Wednesday, October 8th at the Texas Health Presbyterian Hospital in Huston, Texas. This infection took place despite the use protective suits and face masks. Because of the strain the new infection imposes on their resources,the Texas Health Presbyterian Hospital has placed their emergency room on “referral status.” This means they aren’t accepting any new emergency patients. All are being referred to other hospitals. Such chaos exists after only two Ebola cases.

In an interview with Neil Cavuto on Fox Business News, October 9, 2014, CDC Director, Tom Frieden, MD., insisted that a travel ban is not the right answer. “We don’t want to isolate parts of the world, or people who aren’t sick, because that’s going to drive patients with Ebola underground, making it infinitely more difficult to address the outbreak.” Using the analogy of forest fires, Dr. Frieden continued,“When a wildfire breaks out we don’t fence it off. We go in to extinguish it before one of the random sparks sets off another outbreak somewhere else.” Really? The prime method of fighting forest fires is to prevent their spread by creating fire breaks, even setting backfires. Without new fuel, forest fires quickly burn out. Fences work on fires, as would restricting all non-essential travel.

While WHO still soft-pedals the issue, it reports facts which can’t be dismissed.  According to experts in the WHO, Ebola can be spread by coughs and sneezes. It remains viable in air as long as it remains moist. By current definition, only diseases like influenza, which remains viable in air even when dry, are designated “airborn.” The CDC is parsing words on the definition of “direct contact.” Likewise the virus remains viable for several days on moist surfaces, clothing and body fluids. There is no solid evidence whether or not an asymptomatic victim is infectious. Not enough is known at present. The CDC is willing to assume that since we can’t prove victims are infectious before symptoms appear, there’s no need to be concerned.

America’s response, as of Saturday, Saturday, Oct. 11 to the Ebola epidemic in western Africa is to inspect all arrivals, at least by air travel (Screening has already been instituted at Kennedy International Airport for travelers from Liberia, Sierra Leone and Guinea), by a cursory inspection for certain symptoms (e.g., fever) and a supplementary questionnaire. We are told that this will virtually eliminate the possibility of Ebola spreading to the United States, negating the need to limit flights to and from the affected region. However it appears that public health officials, including the CDC, are simply repeating policies emanating from the White House rather than citing good science. It is a long-established principle that you cannot “inspect” quality into a system, that inspection merely alerts you to the presence of a problem. This is the fallacy of inspection.

This principle is rooted in the science of quality control.  Quality control began to take a much more scientific approach took form during World War I due to problems in procurement of vast quantities of materiel from a multitude of suppliers. The concept of statistical quality control arose from those efforts, with significant advances during and after World War II. Walter Deming (aka W. Edwards Demings) was a prime mover in this area, and played a key role in the re-development of manufacturing in Japan following the war. Deming’s work in Japan is responsible for their reputation for high quality after a rocky start.  His success there was met by skepticism in his home country, the United States, where quality control was dominated by tradition – If enough customers complain, we’ll fix it

One of the early facts uncovered by the scientific approach was you cannot achieve quality control by inspection alone. You cannot cull out bad products that so that only good products get shipped. How bad is it? Visual inspection will only detect 25% to 35% of bad products passing through the inspection station. Instrumentation helps a little, but human fatigue and inattention remain significant factors. Where stakes are high, as in the production of food and pharmaceuticals, three or more inspectors look at the same items in order to improve the odds of finding defects. On the other hand, 1% to 2% of good products will be rejected. Re-inspection of rejected items is costly because the percentage of bad product is much higher than in the normal production stream, and require much greater care and effort.

Under the principles developed by Deming and others, inspection is a tool to identify problems, not solve them. Once identified, the root cause of the problem must be identified and solved. If you don’t direct your main effort to solving the problem, you are fooling yourself, and confusing others. No conscientious manufacture would deliberately send products down the line in hopes that inspection would identify and cull the bad ones. That doesn’t mean it never happens.

There are multiple implications of these principles to screening airline passengers. Only passengers directly from affected countries will be screened. The potential for infections are much higher for these countries, hence more likely to be missed. Secondly, people will take indirect routes in order to bypass these inspections. Finally, inspections will be instituted at only five major air ports out of hundreds of potential entry points. The inspection itself will only identify potential victims by the presence of a fever (or other symptoms).

Since there are many causes for fever, the percentage of false identifications will be large, causing disruption to air traffic, not to mention the lives of those culled out of the crowd. The incubation period of Ebola is long, from five to twenty-one days (why not twenty or twenty-two?), so most people infected will pass through undetected, and possibly contact many people before the disease surfaces. The resources needed to respond to an active case of Ebola are enormous. A few dozen cases would overwhelm our system, much less hundreds of cases.

Under government procurement procedures, failing inspection leads to increased inspection levels (and higher cost), ultimately to disqualification of the supplier. In the proposed health screening procedures, there is no intervention plan, just a reaction.

In order to limit our exposure to a possible epidemic, non-essential travel to and from the afflicted regions be curtailed. There should be no casual travel or immigration. Contrary to the administration’s talking points, this has no effect on humanitarian aid, any more than our current ban on travel for political reasons. We would simply have to regulate such travel, and quarantine those returning. Such regulation would require a lot of resources for monitoring and a likelihood of treating subsequent infections. The cost and administration of these measures, and the even greater cost of failure, makes sending thousands of troops into harm’s way impractical.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s