Three Tales About How Science Can Save Lives

by March for Science

This post is from Jorge Munoz, a data scientist. This piece illustrates how science works, and how science can inform policy to save lives. 


Wash Your Hands Before Delivering a Baby

On March 13, 1847, Jakob Kolletschka died from an illness with symptoms similar to those of childbed fever, also known as “The Doctor’s Plague.” Kolletschka was a Professor of Forensic Medicine at the Vienna General Hospital in Austria and he acquired the plague after being accidentally pinched with a scalpel by a medical student who was performing an autopsy. This unfortunate event triggered a eureka moment in his friend Ignaz Semmelweis. Semmelweis had been trying to solve the mystery of why so many women were dying of this illness at one of his clinics. While in contrast the rate was lower at other clinics and even for women who gave birth literally in the streets.

Poor women gave birth at a hospital - with a mortality rate of over 10 percent - not only because the service was free, but there were additional benefits later for children born there, such as childcare and food. The women were subjects for the training of physicians and midwives. Semmelweis meticulously went over all the differences between the two hospitals including methods and procedures. He concluded that the only significant difference was that the birthing assistants at one clinic were doctors and midwives at the other.

When his friend Kolletschka died, Semmelweis realized the real difference was that the midwives were only delivering babies, while doctors-to-be were performing autopsies and delivering babies. When his suggestions that medics wash their hands in between was adopted, the mortality dropped to below 2 percent (Semmelweis 1861). The statistics he collected before the measure was adopted are shown in Fig. 1.

Figure. 1. Table from Semmelweis’ 1861 book Die Ätiologie, der Begriff und die Prophylaxe des Kindbettfiebers (The etiology, the concept, and the prophylaxis of childbirth fever) showing the difference in mortality rate at the “division for doctos” and the “division for midwives” (fourth and last column from the left) between 1841-1846, before Semmelweis implemented his solution. Creative Commons BY-NC 3.0 DE

Figure. 1. Table from Semmelweis’ 1861 book Die Ätiologie, der Begriff und die Prophylaxe des Kindbettfiebers (The etiology, the concept, and the prophylaxis of childbirth fever) showing the difference in mortality rate at the “division for doctos” and the “division for midwives” (fourth and last column from the left) between 1841-1846, before Semmelweis implemented his solution. Creative Commons BY-NC 3.0 DE

You would think that a discovery of this magnitude would change minds right away - particularly because it does not require that much effort to wash your hands before delivering a baby and the hypothesis is easy to test. Yet, Semmelweis’ work came before the germ theory of disease had been proved by the experiments of Louis Pasteur and Robert Koch and most medics still believed that the disease was the result of individual imbalances in the body or that it was caused by ‘bad air.’ The main argument of the medical establishment against Semmelweis’ recommendations was that they were based on a statistical correlation instead of a well-developed scientific theory (Hauzman 2006).

Semmelweis was not able to explain why or how washing your hands decreases your chances of getting sick, but he saved hundreds of lives at Vienna General Hospital with his observation. Once germ theory was developed, society benefitted from knowing the why and how to develop mechanisms and chemicals against not just childbed fever, but many other infectious diseases.

Science May Take Time But is Our Best Way to The Truth

This tale illustrates how sometimes realizing the truth takes time. But the strength of the scientific method is that truth is eventually reached and scientists surrender their opinions to the evidence. Thanks to the germ theory, millions of lives have been saved.

Smoking Can Kill You

On March 8, 1881, James Bonsack got a patent for a ‘Cigarette Machine’ that could roll up to 200 cigarettes per minute (U.S. Patent 238,640 1881) and revolutionized the tobacco industry by allowing the mass production of cheap cigarettes. Smoking cigarettes became very popular during WWI in both America and Europe, when the tobacco industry encouraged the public to send cigarettes to soldiers in the trenches and governments provided tobacco with the intention of maintain troop morale (Johnson 2014).

When the soldiers came back from the war they brought their newfound ‘habit’ back and cigarette smoking became widespread in society. With the increased smoking of cigarettes came an increased incidence of lung and other types of cancer, as recorded in death certificates and other public records. The correlation was picked up by many in the medical community, but the causation was far from obvious as there were also a population increase, better detection and diagnostic methods, and more pollution due to industrialization.

Starting in 1929, Fritz Lickint performed and published case-series studies in which he demonstrated that lung and throat cancer patients were especially likely to be smokers, and showed that the incidence was markedly different in men and women because smoking was more popular among men (Lickint 1953). Lickint also found that being around smokers increases the chances of acquiring cancer and coined the term ‘passive smoker.’

Figure 2. Difference in mortality rate between smokers and non-smoker U.S. Veterans from the 1964 Report on Smoking and Health. Image is public domain because it is a work prepared by an officer or employee of the United States Government as part of that person’s official duties under the terms of Title 17, Chapter 1, Section 105 of the US Code.

Figure 2. Difference in mortality rate between smokers and non-smoker U.S. Veterans from the 1964 Report on Smoking and Health. Image is public domain because it is a work prepared by an officer or employee of the United States Government as part of that person’s official duties under the terms of Title 17, Chapter 1, Section 105 of the US Code.

You would think that public health institutions or governments would alert the public about the risk of smoking, even if the actual mechanism by which cancer was being caused was not known. But this is where the story takes an interesting twist because the Nazis actually accepted Lickint’s conclusions and started an anti-tobacco campaign (Proctor 1999). Lickint himself was demoted from his post at the Chemnitz hospital and constricted into the army as an aide because of his political views.

Other countries were slower to warn their populations and in fact encouraged smoking. For example, many tobacco companies sent tobacco for free to American soldiers during WWII. Eventually the US government caught up and in 1964 the Surgeon General published the Report on Smoking and Health, a review of 7,000 scientific articles by an Advisory Committee that concluded, among other things, that smokers have an abnormally high mortality rate and that there is a causative link between smoking and lung cancer (Terry et al. 1964). A plot from that report is shown in Fig. 2. The report incentivized hundreds of thousands of people to quit smoking, saving or extending many lives. Still the actual mechanism by which carcinogens in cigarettes produce cancer was not demonstrated until the 1990’s when genetic experiments became common (Denissenko et al. 1996).

The Scientific Method Saves Lives

This tale illustrates how the systematic study of an observation and the gradual accretion of evidence and knowledge –  the scientific method – can save lives and improve our well-being. When we are able to put together an explanatory and predictive theory based on fundamental knowledge, the benefits multiply. In this case, the mutation theory allows us to find other carcinogens in the environment and develop medical treatments.

Carbon Dioxide is Heating Up the Planet

In the Austral summer of 1982-1983, a Soviet-French collaboration drilled and later analyzed a 6,800-foot ice core (a very long and relatively thin cylinder) from the Vostok station in Antarctica--the coldest place on Earth) (Barnola et al. 1987). This and other collaborations had been digging out ice cores from the Arctic, Antarctica, and Greenland starting in the early 1970s, but this one was the deepest ice core ever dug. The length matters because it allowed them to reconstruct the composition of the Earth’s atmosphere from air bubbles trapped in the ice for over 150,000 years, which covers several glacial periods. The analysis consisted of cutting out pieces from different parts of the core, crushing them, and analyzing the gases trapped in tiny bubbles in the ice, including water vapor.

The main goal of these paleoclimatologists was not to find the direct connection between carbon dioxide and the temperature of the Earth, but that’s what they achieved. In Fig. 3 you can see on the left axis the temperature differential DT (DT = 0 is the temperature of the Earth at the time of the publication) and on the right axis you can see the carbon dioxide concentration. It is evident that the peaks in temperature and carbon dioxide concentration coincide. What really spooked everybody is that the carbon dioxide concentration had been increasing every year since Charles Keeling and others started measuring it in 1957 (Briggs 2007) and continues to increase to this day. Regarding temperature, the global warming could still not be unambiguously measured in the 1980s, but the trend is very clear now: sixteen of the 17 warmest years on record have occurred since the year 2000 and the years 2014, 2015 and 2016 set a new record (Mooney 2017).

Figure 3. Correlation between the temperature of the Earth in degree Celsius above or below the temperature at the time of publication (bottom curve, left axis) and carbon dioxide concentration in the atmosphere (top curve, right axis) as a function of time in thousands of years ago, from Barnola, et al. Nature 329, 408 (1987).Reprinted by permission from Macmillan Publishers Ltd: Nature 329, 408, copyright 1987.

Figure 3. Correlation between the temperature of the Earth in degree Celsius above or below the temperature at the time of publication (bottom curve, left axis) and carbon dioxide concentration in the atmosphere (top curve, right axis) as a function of time in thousands of years ago, from Barnola, et al. Nature 329, 408 (1987).Reprinted by permission from Macmillan Publishers Ltd: Nature 329, 408, copyright 1987.

Correlation is not causation and scientists are a skeptical crowd, so the Barnola et al. results by themselves did not convince scientists that global warming is human-made, but they did convince everybody that that “global climate system and carbon cycle are intensely interactive” (Sundquist 1987). Either increased carbon dioxide concentration causes global warming or global warming causes an increase in carbon dioxide. The key question to be answered was: are there natural sources of carbon dioxide that can account for the measured increase in atmospheric carbon dioxide or does it come mainly from human-made sources? Scientist organized the Intergovernmental Panel on Climate Change (IPCC) to answer this question and learn more about climate and global warming. Their first report came out in 1990 (Houghton et al. 1990) and was the basis for the United Nations Framework Convention on Climate Change. Evidence has been piling up since then and shows that the combustion of fossil fuels and deforestation have increased the concentration of carbon dioxide in the atmosphere by 43% since the industrial revolution (NOAA 2005).  The Fifth Assessment of the IPCC released in 2015 concluded that it is extremely likely that human influence is the dominant cause of global warming (Pachauri 2015).

Climate is complex and we don’t yet have a simple theory, like in the case of germs to explain the results, but we have several models that have a record of accurate predictions and the statistical evidence of global warming is solid. The scientific method has uncovered the truth and it is our decision whether we will start ‘washing our hands or continue smoking.’

The Way of Science

Imagine that you had a biased coin that landed heads 80 percent of the time, but you did not know that ahead of time. How would you go about to determine whether it is biased or not, or to what degree? What would be enough evidence to convince you one way or the other? In this case, the experiment is easy to set up: toss the coin several times and keep track of how many times it lands ‘heads.’ With only a couple of tosses (or experiments) you would not know whether it is biased or not (and you should not make decisions based on so little data), but if it lands heads 82 out of the first 100 times you start to collect enough data to make a determination, as shown in Fig. 4. Each time you repeat the experiment you update your previous estimate so that it incorporates the new knowledge and your new estimate improves. This is called Bayesian statistics and is how science works. The most powerful kind of knowledge comes in the way of a mathematical theory or model that predicts the outcome of an experiment before it happens based solely on things you know beforehand. For the coin toss this previous knowledge could be the side that is starting up, the force that you are going to apply to the coin, the launch angle, etc. Note that you don’t need to reach the mathematical theory phase before your results are useful: you could set up a whole casino with only statistical knowledge, but you definitely have to be more careful.

Figure 4. Probability distribution resulting from a coin that lands heads 80 percent of the time. Before tossing the coin even once, we don’t know anything (blue); after tossing the coin 10 times we start to suspect it is biased but we can’t be too sure of the value (red); after 100 times, we can be pretty convinced that the coin is biased towards 80 percent heads (green).

Figure 4. Probability distribution resulting from a coin that lands heads 80 percent of the time. Before tossing the coin even once, we don’t know anything (blue); after tossing the coin 10 times we start to suspect it is biased but we can’t be too sure of the value (red); after 100 times, we can be pretty convinced that the coin is biased towards 80 percent heads (green).

Science in general is not that different from the coin toss case and the job of scientists is to set up experiments that tell you the most about the world, come up with mathematics to describe what is happening, iterate until theory converges with reality, and keep each other honest. In the three cases above I show the way science usually happens: It goes from 1) an initial observation and statistical evidence to 2) the development of a theory that explains the observations and generalizes to other cases. In the case of childbed fever and smoking the loop has been closed: we know what happens, why and how to prevent it. The case of global warming is exciting [JR1] because the loop is closing but there is still a lot to discover. Yet, in Climate Change 2013: The Physical Science Basis, the IPCC reviewed 9,200 peer-reviewed studies that all show the same story: the coin is biased, global warming is happening (IPCC 2013). The question now is how we use this science to best serve individuals and societies.


About the Author

Jorge Munoz is a data scientist in Intel IT Labs at Intel Corporation, he designs algorithms to analyze and simulate complex systems. He has a PhD from Caltech in Materials Physics, his doctoral research was on the quantum mechanical origins of the phase stability of materials. He can be contacted via Twitter with the handle @abnormalxy. The opinions expressed are his own and don't necessarily represent Intel's positions, strategies, or opinions.


Citations

  • Semmelweis, I. P. (1861). Die Ätiologie, der Begriff und die Prophylaxe des Kindbettfiebers. Leipzig: Hartleben.
  • Hauzman, E. E. (2006). Semmelweis and his German contemporaries. 40th International Congress on the History of Medicine, ISHM 2006. Retrieved from https://web.archive.org/web /20080530013805/http://www.ishm2006.hu/abstracts/files/ishmpaper_093.doc
  • Bonsack, J. A. (1881). U.S. Patent No. 238,640. Washington, DC: U.S. Patent and Trademark Office.
  • Johnson, N. K. (2014). World War I, Part 5: Tobacco in the Trenches. Retrieved from https://pointsadhsblog.wordpress.com/category/nick-johnson/
  • Lickint, F. B. (1953) Ätiologie und Prophylaxe des Lungenkrebses als ein Problem der Gewerbehygiene und des Tabakrauches. Dresden: Steinkopff.
  •  Proctor, R. N. (1999) The Nazi War on Cancer. Princeton: Princeton University Press.
  • Terry, L. L. et al. (1964) Smoking and Health: Report of the Advisory Committee to the Surgeon General of the United States. Washington, DC: Department of Health, Education, and Welfare
  • Denissenko, M. F., Pao, A., Tang, M. S., & Pfeifer, G. P. (1996). Preferential Formation of Benzo[a]pyrene Adducts at Lung Cancer Mutational Hotspots in P53. Science, 18 (5286), 430-432.
  • Barnola, J. M., Raynaud, D., Korotkevich, Y. S., & Lorius, C. (1987). Vostok ice core provides 160,000-year record of atmospheric CO2. Nature 329 (01 October), 408-414.
  • Briggs, H. (2007, December 2). 50 years on: The Keeling Curve legacy. BBC News. Retrieved from http://news.bbc.co.uk/2/hi/science/nature/7120770.stm
  • Sundquist, E. T. (1987). Ice core links CO2 to climate. Nature 329 (01 October) 389-390.
  • Mooney, C. (2017, January 18). U.S. scientists officially declare 2016 the hottest year on record. That makes three in a row. The Washington Post. Retrieved from https://www.washingtonpost.com/news/energy-environment/wp/2017/01/18/u-s-scientists-officially-declare-2016-the-hottest-year-on-record-that-makes-three-in-a-row
  • Weart, S. R. (2008) The Discovery of Global Warming: Revised and Expanded Edition. Cambridge: Harvard University Press.
  • Houghton, J. T., Jenkins, G. J., & Ephraums, J. J. (Eds.). (1990). Climate Change: The IPCC Scientific Assessment. Geneva: Intergovernmental Panel on Climate Change.
  • National Oceanic and Atmospheric Administration. (2005). After two large annual gains, rate of atmospheric CO2 increase returns to average. Washington, DC: NOAA. Retrieved from http://www.noaanews.noaa.gov/stories2005/s2412.htm
  • Pachauri, R. K. (Ed.). (2015) Climate Change 2014 Synthesis Report. Geneva: Intergovernmental Panel on Climate Change.
  • Intergovernmental Panel on Climate Change. (2013). Climate Change 2013: The Physical Science Basis. Geneva: Intergovernmental Panel on Climate Change.