Skip to 0 minutes and 10 secondsBARBARA MCPAKE: Why do the number and type of health workers matter? This relationship drawn here from a group of Asian countries between the number of skilled birth attendants in a country and the maternal mortality ratio holds for other regions and time periods. And this applies not only to maternal mortality but also to infant and under five mortality, though the relationships aren't as steep. Availability of health workers has a direct impact on health outcomes. I'm sure it's not a surprise that there's a strong correlation between a country's GDP per capita and the density of its health workforce. For example, the country with the most nurses and midwives for every 1,000 people, about 20, is Monaco.

Skip to 0 minutes and 51 secondsAnd the country with the least, with only 0.078, or less than one per 10,000 people, is Somalia. Countries with a lower rate than 0.5, or one nurse or midwife per 2,000 people, are concentrated in Africa. Where the data are slightly surprising are the lower rates in Northern America than much of Europe, Australia, and New Zealand, and the very low rates in the richer countries of South America, Argentina and Chile. Before looking at the next map, pause for a minute and think about the likely explanation. One explanation you may have considered is that these countries may have higher levels of other health personnel that can substitute for nurses and midwives.

Skip to 1 minute and 33 secondsThe equivalent map for physicians suggests that Argentina makes up for its shortage of nurses and midwives with a very high density of physicians. You might notice concentration of high densities of physicians in communist or formerly communist countries. The highest density is in Cuba, although it's too small to pick out on the global map. I think that the high physician density in Sudan is a data entry error. Here I've calculated and graphed the ratio of nurses and midwives to physicians for all countries.

Skip to 2 minutes and 3 secondsThe analysis confirms that Chile and Argentina are indeed at one extreme of the range with around one nurse for every seven doctors and are the only two countries who have less than one nurse or midwife for every four physicians. At the other end of the spectrum is Kiribati with nearly 23 nurses for every doctor. And a number of African countries, including Liberia and Malawi, are also at that end of the figure, with 19 and 18 nurses per doctor respectively. In the middle are health systems with quite different levels of resources. For example, both Norway and Sudan have ratios of about three nurses for every doctor, while Spain and Vietnam are fairly close to ratios of one to one.

Skip to 2 minutes and 44 secondsWhat do you think might be the implications of having different ratios between cadres of health professionals? We need to think about health workers who are non-clinicians too. What do you notice about this table? The data are rather out of date, coming from the 2006 World Health Report. But an analysis of more current data would probably not produce a different conclusion. Generally as densities increase, the proportions of managers increases. And that implies that if clinical workers are scarce, managers and support workers are even more scarce. We tend to focus on front-line health workers when we talk about the health workforce crisis. But these data suggest that there is an even greater crisis in the shortfall of management and support workers.

Skip to 3 minutes and 27 secondsAfrica has the lowest proportion of health management and support workers. Globally, one third of health workers are management and support. In the Americas, nearly half. In Africa, it's less than one fifth. This means that clinical workers in Africa tasked with providing health care to the largest populations with the greatest health needs get much less managerial support to do this. We've just been looking at the highly inequitable distributions of workforce numbers between countries. Within countries, the situation further exacerbates those inequities. Here are a few examples of measures of maldistribution of workforce within countries.

Skip to 4 minutes and 6 secondsIn more detail in Zambia, this map, also from the 2006 World Health Report, shows the vast areas of the country where there are fewer than one health service provider of any kind per 1,000 population. Munga and Maestad analysed the distribution of health workers in Tanzania. If distribution were equitable, each district would have a proportion of health workers in line with its proportion of population. The red 45-degree line would show both population share and health worker share. The blue line shows the reality. For example, at point x, it shows that the least well staffed districts with 20% share of the population have only about 8% of the health workers.

Skip to 4 minutes and 47 secondsThe curve here is the blue line where the further it stretches from the red line, the more unequal is the distribution. Here is the same analysis undertaken by Munga and Maestad broken down by cadre. Guess which cadre is the pale blue line? It's doctors, of course, which are the most unevenly distributed cadre.

The composition and distribution of the health workforce

In this video, we looked at the composition and distribution of health workers and how they differ across countries. While looking at the adequacy of health workforce, it is important to not just focus on numbers and densities of health workers, but also the cadre-mix: the ratios between different cadres of health professionals and their distribution.


What kinds of systems level factors would influence the cadre-mix of health professionals in a particular setting and/or their distribution? For example, what might explain the much higher ratio of physicians to nurses in countries like Argentina and Chile? Or higher number of nurses and midwives compared to physicians in other countries? Use the comment box below to list a few system level influences.

Share this video:

This video is from the free online course:

Health Systems Strengthening

The University of Melbourne

Get a taste of this course

Find out what this course is like by previewing some of the course steps before you join: