Is there a more appropriate way to interpret/look at this Maternal Mortality Rate data? My fiancee and I are discussing maternal mortality rates in the US. According to the data, from a May 2017 NPR article, 26.4 out of every 100,000 women die during child birth in the US. My fiancee wanted to put this into percentage, which is .000264%. He is claiming that using a percentage is more appropriate for this matter and that saying 26.4 out of 100,000 is being used because it sounds like a bigger number and is more dramatic. He is claiming that looking at it as a percentage is more effective, I am claiming that it can be looked at that way, but it seems easier to interpret more appropriate to look at it the first way. Why is it that NPR chose to show it as 26.4 out of 100,000 instead of .000264? Is it more appropriate to show the statistic one way or the other or can they be looked at the same? Let me know if you have any questions.. we are not very math savvy and the conversation is getting heated, it would be nice to understand why statistics are shown certain ways. For instance sometimes a pie chart makes more sense than putting data into a bar graph. So why would one look at it as a percentage rather than a ratio? http://www.npr.org/2017/05/12/528098789/u-s-has-the-worst-rate-of-maternal-deaths-in-the-developed-world
 A: First, it is standard practice in public health to use a base for a rate other than percent. If you go to the Centers for Disease Control and Prevention web site or that of its statistical agency, the National Center for Health Statistics you will find that many tables of rates use per 10,000 or per 100,000.
Most rates use a single base over many years and most use a base that is pertinent to their rarity in the population. For example, the overall death rate is usually quoted as per 10,000 and death rates due to specific causes are usually quoted as per 100,000.
REFERENCE: Maternal mortality rates from 1915 through 2003 can be found here: https://www.cdc.gov/nchs/data/series/sr_03/sr03_033.pdf
This is kept the same over many causes of death and over many years. These rates have been used since about 1910. This makes it easier to recognize and interpret patterns and changes without having to spend time and effort interpreting the scale.
One reason to choose a particular scale is because numbers are easier to read if they are between about 10 and 100. But many death rates fall below this and some fall above it if a basis of 100,000 is used. 
Many people, in my personal experience, have difficulty reading and interpreting numbers that have zeros after the decimal. I do not know if this reflects a known phenomenon in perception or cognition. But this phenomenon does exist.
In past, when tables were produced by typesetters and statistical clerks, a base such as percent or decimal would have introduced many errors into published tables.
A logical argument for a different base doesn't hold much water. Percent is not special here and decimals also have a better claim. But neither claim is especially good or even better than standard practice.
If you are a professional in public health then the use of unusual, variable, or unfamiliar bases for rates is confusing to readers. So representing another basis, e.g., percent, as superior is unfounded and misleading. 
Now, to be specific about maternal deaths. Over the last 60-70 years there have been several million births each year in America. In many recent years, to give this some context, there have been around 4 million, give or take a couple of hundred thousand. 
After the discovery and development of antibiotics, particularly sulfa drugs in the 1920s, maternal deaths dropped steadily. This decrease was fairly dramatic until the 1950s and more slowly since then. Since the 1950s there have been fewer than about 1000 maternal deaths a year and in many of those years there have been about 300 to 500 a year. (Apparently, this number has been rising lately) This is, of course, a very small number of the women who give birth.
So, the maternal death rate is exceedingly small by any measure. However it did more than double from 1990 to 2013, increasing from 12 to 28 deaths per 100,000. This increase is likely due to assignable causes and is not random variation. In other words, our health care system and our social service system did a better job 30 years ago than they do today.
So the basis of 100,000 for maternal deaths is useful in concluding that the US has a real problem in the recent increase in maternal deaths.
Knowing the exact number of such deaths is also useful as an indicator of how good our maternal care system is. A large fraction of maternal deaths are completely preventable. That is, it is a kind of mortality that could still be greatly reduced, even if only back to the rates that prevailed 30 years ago.
As a separate, totally apolitical observation about current events, if America once again changes its health insurance system in the direction that seems likely I predict that the maternal death rate will rise even higher. This seems pretty obvious with only limited statistical theory or analysis.
