Florence Nightingale: how did a nurse change army hospitals with figures and a diagram?

Can you convince a minister with a single table? If the table has hundreds of rows of figures, probably not: the minister will not read it to the end. But if you turn the figures into a picture that can be understood at a glance, the situation changes. In the middle of the 19th century the English nurse Florence Nightingale proved this.

Who was Florence Nightingale?

Britannica describes her as a British nurse, statistician and social reformer. She was born on 12 May 1820 in the Italian city of Florence and died on 13 August 1910 in London.

The turning point in her life came during the Crimean War. Nightingale set out from England on 21 October 1854 with an officially authorized group of 38 women nurses and in early November reached the barrack hospital at Scutari (in the area of present-day Istanbul) (Britannica; the National Archives of the United Kingdom). Observing the state of the hospital, she came to an important conclusion: for soldiers' lives, hospital conditions matter just as decisively as battle wounds.

Why is she called a statistician?

Nightingale tried to prove her observations not with emotion but with figures. According to the Royal Statistical Society, in 1858 she became a member of the Statistical Society of London (today's Royal Statistical Society) and was the society's first female member. She remained a member until the end of her life, that is, for more than 50 years. Among those who signed her membership application was the well-known medical statistician of the time, William Farr.

The collaboration with Farr is of particular importance. As the history of statistics page of the University of York states, Farr helped Nightingale refine the series of charts on which her statistical reputation rests. In other words, the famous diagrams were not the product of one person but the joint product of a nurse's experience and professional statistical knowledge.

What did the "rose diagram" show?

In 1858 Nightingale had her large report on the health, efficiency and hospital administration of the army privately printed; the report had been prepared to be presented to the Secretary of State for War at his request. It contained a coloured chart entitled "Diagram of the Causes of Mortality in the Army in the East" (History of Information).

The diagram was not an ordinary bar chart. According to the description in the Lehigh University Libraries exhibition and in Britannica's encyclopedia for students: - the circle is divided into twelve segments with equal angles, each segment representing one month; - the colours distinguish the causes of death; - the angles are the same, but the radii differ: the quantity is expressed through the area of the segment.

As a result, it could be seen at a glance how much greater death from disease was than death from wounds. According to an article in the Science Museum journal, in February 1855 a sanitary commission headed by Dr John Sutherland and the engineer Robert Rawlinson was appointed to investigate and remedy conditions in the Crimea. Nightingale split the months into two circles: according to the analysis of the statistician Michael Friendly and his co-author, this made it possible to compare deaths before and after the commission's arrival side by side. In this way the picture became evidence for the claim that "sanitation saves lives".

What is the secret of drawing a circular diagram correctly?

There is a small but important mathematical subtlety here. The area of a segment depends on the square of the radius. So if the number of deaths is 4 times larger, the radius must be drawn not 4 but 2 times larger. Otherwise the picture exaggerates the truth. In Nightingale's diagram the data is given precisely through area, which is why it is still shown as an example in data visualization classes today.

Then and now

Example: showing infections at the "Shifo-Nur" clinic in a circle

The example below was compiled by the author; the figures are hypothetical.

Malika, the head nurse of the hypothetical "Shifo-Nur" clinic in Namangan, counted the cases of hospital-acquired infection over three months: 40 cases in month 1, 90 in month 2 and 160 in month 3. She wants to draw a diagram in Nightingale's style.

In month 3 there are 4 times as many cases as in month 1 (160 ÷ 40 = 4). In a correctly drawn diagram the radius of the month 3 segment is 2 times larger and its area 4 times larger. If Malika made the radius directly proportional to the number, the area would come out 16 times larger and the management would perceive the problem as 4 times more serious than it really is. The correct diagram, on the other hand, shows the conclusion "cases are rising, hygiene measures need to be strengthened" without exaggeration.

Where does this stand in history?

Link with the previous period: in the 17th century John Graunt looked for regularities in the London Bills of Mortality, and in the 19th century Quetelet measured society through the "average man". Nightingale turned these figures into pictures and brought them to the decision-maker. At the next stage, in the 20th century, the sampling method and mathematical statistics put the collection of data itself on a scientific basis.