Clinical Trial + Math Errors = Research Critical Incident

Use of Artificial Turf in Playing Fields – Issue 60
January 16, 2017
The CFO’s role in preventing a university’s biggest financial losses
February 15, 2017

Clinical Trial + Math Errors = Research Critical Incident

Invoice drawn addition black white

By Harry Rosenthal

Of the many classes of university liability risk exposures protected by Unimutual, one of the most serious regards research involving human subjects.  Often called clinical trials, this class of research holds great promise in the development of new medications and medical devices as well as furthering our understanding of the human body.  On the other hand, such research can endanger subjects, resulting in serious injury, disfigurement or even the death of the participant.  While such adverse outcomes are very rare, they are often widely reported in the international media, interrupting valuable research, adversely impacting reputations and helping to invigorate the “anti-science” community, who use such failures to illustrate their views.  Human trials gone wrong can bring a sudden halt to promising research, and such trials are very common throughout the global university sector including Australia.  Unimutual protects hundreds of such trials each year.

A high-profile clinical trial incident in the UK has recently come before the courts, and is worthy of examination by the Australian research community.  Northumbria University, in the UK has a student population of 30,000 and an annual budget of close to AU$420 million.The research involved the use of university students who participated in a program to study the effects of caffeine on sports performance.

The protocol required students to ingest elevated amounts of caffeine and perform basic exercise activities, which were monitored by the researchers.  As part of the project, two of the students were accidentally administered the equivalent of 300 cups of coffee and immediately began to suffer adverse side effects.  Both students were in their 20s and studying a sports science degree at Northumbria.  Upon taking the mixture of caffeine and orange juice, they immediately suffered dizziness, blurred vision and rapid heartbeats. As a result they both were taken to a hospital and each spent days in the Intensive Care Unit, receiving dialysis and suffering significant weight loss. Both students recovered and were able to resume their studies.

A review of the incident indicated several contributing factors:

  1. During the trials, the researchers switched from using a commercial caffeine tablet to using pure caffeine in powder form. The researchers were now required to prepare the caffeine dosages themselves.
  2. As a result of using the new materials, the research technicians performed their own calculations based on the students’ height and weight, to determine the amount of caffeine powder to administer in orange juice.
  3. The technicians made a math error in the dosage calculations, giving the students 30.6g and 32g respectively, instead of the prescribed 306 mg and 320mg. The decimal points were wrongly inserted resulting in a dosage 100 times greater than prescribed in the experiment.

The court found that the students could have died if not for the prompt medical attention they received.  It was determined the technicians administering the trial were not properly trained and the academic overseeing the experiment did not check the calculations.  The university was fined AU$668,000 over the incident and made to pay AU$45, 000 in court costs.

(See: Northumbria University ‘life-threatening’ caffeine test fine – BBC or University fined £400k after caffeine test nearly kills two students – Express)

Math errors are quite common in the scientific world, as almost all scientific disciplines rest on a foundation of mathematics.  Certainly in the space program, for example, math errors are familiar to us.  Many will recall the famous loss of the NASA Climate Orbiter in 1999, where a programmed burn of the craft’s main engines for 16 minutes, 23 seconds was designed to place the spacecraft in to orbit around Mars. Unfortunately, as the $AU136 million spacecraft orbited behind Mars, it was never seen again. An investigation in to the craft’s disappearance revealed that the Florida engineering team used Imperial measurements in their calculations and the Texas Jet Propulsion lab used the conventional metric system in their calculations.  The spacecraft probably burned up in the Martian atmosphere. Ironically a short time later, a sister spacecraft, the Polar Lander crashed, for unclear reasons, but differing conversion units was not suspected.

Another well-known math error was Air Canada flight 143 in 1983, where a commercial Boeing 767 airliner literally ran out of fuel at 41,000 feet halfway on its trip from Montreal to Edmonton.  Before take-off it was discovered the fuel gauge on the aircraft was not working, so fuel was loaded and reported to the crew. Unfortunately the ground crew reported the fuel amounts in gallons and the flight crew assumed they were in the usual units of measure:  litres.  The plane was able to safely glide to a safe landing at an industrial airport, with no fuel on board.

Math errors are serious and are of constant concerns in hospitals and medical centres when medication dosages are calculated. In clinical trials, they can adversely affect the accuracy of the data and in extreme cases, such as the one cited above, expose the subjects to significant harm. We encourage all researchers to seek support in dosage calculations, and ensure staff are trained in the calculation of the required dosages and that these are properly checked.