health workers:

health workers: a global profile 

chapter one

a global profile
in this chapter

1 Who are the health workers?

4 How many health workers are there?

7 How much is spent on the health workforce?

8 Where are the health workers?

Health workers are people whose job it is to
9 Are there enough health workers?
protect and improve the health of their com-
13 Addressing the shortage how much will it cost?
munities. Together these health workers, in
15 Conclusion
all their diversity, make up the global health
workforce. This chapter gives an overview of what is known about them. It describes
the size and distribution of the workforce, and some of its characteristics, including
how much it costs. It shows that there is a substantial shortage of health workers
to meet health needs, but that shortages are not universal, even across low income
countries. The chapter then considers how much it would cost to scale up training to
meet this shortfall and pay health workers subsequently.
The data used in this chapter are drawn from many different Who are the health workers?
sources, with varying degrees of completeness by country and by This report defines health workers to be all people en-
year. WHO supplemented this information with written requests to gaged in actions whose primary intent is to enhance
numerous agencies as well as with special country surveys request- health. This meaning extends from WHOs definition of
ing information about the numbers and types of health workers the health system as comprising activities whose pri-
and training institutions. Substantial gaps remain, however, in the mary goal is to improve health. Strictly speaking, this
information, and the picture painted here is based on incomplete means that mothers looking after their sick children
data which means that it needs to be interpreted carefully. and other unpaid carers are in the health workforce.
Jean Mohr/WHO
 The World Health Report 2006

They make important contributions and are critical to the functioning of most health
systems. However, the data available on health worker numbers are generally limited
to people engaged in paid activities, so the numbers reported in this chapter are
limited to such workers.
Even then, the definition of a health action for classifying
This report defines paid workers is not straightforward. Consider a painter
employed by a hospital: the painters own actions do not
health workers to be all improve health, though the actions of the painters employer,
the hospital, do. Then take the case of a doctor employed by
people engaged a mining company to care for its employees: the actions of the
doctor improve health, though the actions of the employer do
in actions whose not. A classification system that considers the actions of the

primary intent is to individual alone, or those of the employer alone, cannot place
them both in the health workforce.
enhance health In principle, the report argues that the actions of the
individual are most important, so that the painter is not a
health worker while the mines doctor is. However, in practice,
it is not yet possible to fully apply this rule because much of
the data on health worker numbers do not provide sufficient detail to allow people
directly engaged in improving health to be separated from other employees (1). For
this reason, the report takes a pragmatic approach and includes all paid workers
employed in organizations or institutions whose primary intent is to improve health
as well as those whose personal actions are primarily intended to improve health but
who work for other types of organizations. This means that the painter working for a
hospital is included as is the doctor working for a mine. WHO is working to devise a
more detailed, standard classification system for health workers that should permit
the gold standard definition to be applied in the future (see Box 1.1).
The system of counting used in this chapter allows two types of health workers
to be distinguished. The first group comprises the people who deliver services
whether personal or non-personal who are called health service providers; the
second covers people not engaged in the direct provision of services, under the term
health management and support workers (details are given in Box 1.1; see also Box
1.2). The report sometimes presents information for different types of health service
providers, although such detail is often available only for doctors and nurses. Further
explanation of the sources of the data, classification issues, and the triangulation
and harmonization applied to make the data comparable across countries is found
in the Statistical Annex.
The available data do not allow reporting on the people working for a part of their
time to improve health, such as social workers who work with mentally ill patients. In
addition, the report has chosen not to include workers in other types of occupations
who contribute in vital ways to improving population health, if their main function
lies elsewhere. This category includes, for instance, police officers who enforce
seat-belt laws. Finally, current methods of identifying health workers do not allow
unpaid carers of sick people or volunteers who provide other critical services to be
counted. This exclusion is simply because of a lack of data, and all these valuable
contributions are acknowledged in subsequent chapters.
Furthermore, official counts of the health workforce often omit people who deliver
services outside health organizations, for example doctors employed by mining
companies or agricultural firms, because they classify these employees under the
 health workers: a global profile 

Box 1.1 Classifying health workers

The third version of the International Standard Classification used to define the different types of economic activity in a
of Occupations (ISCO), an international classification sys- country. In ISIC, health is considered a separate industry.
tem agreed by members of the International Labour Orga- Vast numbers of workers with different training and oc-
nization, was adopted in 1987 and is known as ISCO-88 (2). cupational classifications are found in the health industry:
Many national occupational classifications, and most cen- many more than the health service providers themselves.
suses and labour force surveys, use one of the three ISCO These include professionals such as statisticians, com-
versions. Because the system is used to classify all types of puter programmers, accountants, managers and admin-
workers, the breakdown provided for health workers is not istrators and also various types of clerical staff as well as
very detailed, so many ministries of health have developed support staff such as drivers, cleaners, laundry workers
their own classification systems. WHO is now working on a and kitchen staff. Examples of the various types of occu-
process to devise a more detailed, standard classification pations included for the health industry classification in
system for health workers that is consistent with the ISCO. the South African census are provided below.
This work coincides with the update of ISCO-88, which is Some health service providers work in industries other
expected to be ready in 2008. than health, such as mining or manufacturing. According-
The table below shows the health-specific occupation- ly, for this report, health workers include all occupations
al classification used in the South African census of 2001, listed under the health industry, plus people in occupa-
which is typical of many countries using a three-digit ISCO tional groups 15 working in other industries.
coding system (four-digit codes break down each of the The report groups health workers into two categories
categories listed into subcategories). Note that traditional that map directly into the ISCO codes. People covered by
healers are part of the official occupational classification occupational codes for groups 15 in the table are health
and are included in counts in this report where data are service providers; other workers in the health industry
available. are called health management and support workers.
At the same time, another internationally agreed clas- This is shown in the figure, where health workers make
sification system the International Standard Industrial up the first three of the four occupational boxes.
Classification of all Economic Activities (ISIC) is commonly

Occupational classifications for the health industry, South African census, 2001
ISCO groups of health service providers Type ISCO code no.
1. Health professionals (except nursing) Professionals 222
2. Nursing and midwifery professionals Professionals 223
3. Modern health associate professionals (except nursing) Associates 322
4. Nursing and midwifery associate professionals Associates 323
5. Traditional medicine practitioners and faith healers Associates 324
Examples of other occupations involved in the health industry
6. Computing professionals Professionals 213
7. Social science and related professionals Professionals 244
8. Administrative associate professionals Associates 343
9. Secretaries and keyboard operating clerks Clerks 411
10. Painters, building structure cleaners and related trades workers Craft and related trades workers 714
Data source: (2).

Health workers in all sectors

Sector Health sector All other sectors
Health workforce
Health service Health management Health service
Occupation providers and support workers providers
All others

Professionals Professionals e.g. physician

e.g. doctor, nurse e.g. accountant in a hospital employed in mining
Associates Associates company
e.g. laboratory e.g. administrative
technician professional in a hospital
Other community Support staff
e.g. traditional e.g. clerical workers, drivers in a hospital
practitioner Craft and trade workers
e.g. painter in a hospital
 The World Health Report 2006

The global health industries that employ them. An accurate count of such workers
is difficult to obtain, but they make up between 14% and 37%

workforce is of all health service providers in countries with available census

data. Excluding them from official counts results in a substantial
conservatively underestimation of the size of the health workforce and its
potential to improve health. Such undercounting also prevents
estimated to be just consideration of the complex labour market links between
different sectors that could inform planning, recruitment,
over 59 million retention and career paths.

workers How many health workers

are there?
The work undertaken for this report allowed WHO to update
the information contained in its Global Atlas of the Health Workforce (3) for some
countries and to find previously unavailable data for others. Data are generally more
complete for health service providers than for health management and support work-
ers but, using the best available information from various sources, a conservative
estimate of the size of the health workforce globally is just over 59 million workers
(see Table 1.1). This figure is conservative in so far as it is likely to undercount health
workers outside the health industry in countries where census information is not
available.
Health service providers account for 67% of all health workers globally, though
only 57% in the Region of the Americas. A breakdown by the level of national income
in a country shows that health management and support workers slightly outnumber
health service providers in high income countries, while the opposite is the case in
low and middle income settings where health service providers typically constitute
over 70% of the total health workforce.
Within the category of health service providers, attention is often focused on
the ratio of nurses (and midwives) to doctors, though the exact numbers and mix
necessary for a health system to run efficiently and effectively remains unclear
(48). The number of nurses per 1000 doctors for a typical country is highest in the
WHO African Region, partly because of the very low number of doctors per 1000
population in that region. The ratio is lowest in the Western Pacific Region. There is
also considerable heterogeneity among countries within regions. For example, there
are approximately four nurses per doctor in Canada and the United States of America,
while some other countries in the Region of the Americas, such as Chile, El Salvador,
Mexico and Peru, have more doctors than nurses. Similarly, in the European Region,
there is nearly one physician for every nurse in Bulgaria, Portugal and Turkey, but
around five nurses for each physician in Norway and the United Kingdom.

Box 1.2 The invisible backbone of the health system: management and support workers

People who help the health system to function but do not whole. Health management and support workers provide
provide health services directly to the population are often an invisible backbone for health systems; if they are not
forgotten in discussions about the health workforce. These present in sufficient numbers and with appropriate skills,
individuals perform a variety of jobs, such as distributing the system cannot function for example, salaries are
medicines, maintaining essential buildings and equipment, not paid and medicines are not delivered.
and planning and setting directions for the system as a
 health workers: a global profile 

Table 1.1 Global health workforce, by density

Health management and
Total health workforce Health service providers support workers
Density Percentage of Percentage of
(per 1000 total health total health
WHO region Number population) Number workforce Number workforce
Africa 1 640 000 2.3 1 360 000 83 280 000 17
Eastern
2 100 000 4.0 1 580 000 75 520 000 25
Mediterranean
South-East Asia 7 040 000 4.3 4 730 000 67 2 300 000 33
Western Pacific 10 070 000 5.8 7 810 000 78 2 260 000 23
Europe 16 630 000 18.9 11 540 000 69 5 090 000 31
Americas 21 740 000 24.8 12 460 000 57 9 280 000 43
World 59 220 000 9.3 39 470 000 67 19 750 000 33
Note: All data for latest available year. For countries where data on the number of health management and support workers were not avail-
able, estimates have been made based on regional averages for countries with complete data.
Data source: (3).

Information has also been obtained on the relative availability of dentists and
pharmacists, though fewer countries report this information. There is close to parity
between the number of pharmacists and doctors in the South-East Asia Region,
substantially more than in the other regions. The ratio of dentists to doctors is
highest in the Region of the Americas. These data should be interpreted carefully,
however, because of the difficulties involved in counting dentists and pharmacists,
many of whom are likely to work in the private sector.

Public and private sector workers

Most data on the distribution of health workers between the public and private
sectors describe who is the primary employer of the worker rather than where the
money to pay the salary or most of the workers income comes from. This informa-
tion suggests that the majority of health service providers in low and middle income
countries report their primary site of employment as the public sector: over 70% of
doctors and over 50% of other types of health service providers. Insufficient infor-
mation is available from high income countries to allow similar analysis; it is likely,
however, that the proportions officially employed by governments are lower, because
many providers are officially in private practice despite much of their income coming
directly from the public purse. This is also true for providers employed by faith-based
and nongovernmental organizations in many settings.
These broad averages hide considerable variation across countries with the same
level of income or in the same geographical region. For example, while 70% of
doctors in sub-Saharan Africa are officially employed in the public sector, in six
countries in the region more than 60% of them are formally employed in the private
sector. Furthermore, even in countries where the public sector is the predominant
employer, public sector employees often supplement their incomes with private work
or receive a large part of their income directly from patients rather than from the
government (911). The data presented here on the relative importance of the public
sector need, therefore, to be supplemented with information on health expenditures,
as discussed below.
 The World Health Report 2006

Sex and age of health workers

Figure 1.1 illustrates the average distribution of women health service providers
across countries. Insufficient information is available on the sex distribution of health
management and support workers for them to be included. Men continue to dominate
the medical profession, while other health service providers remain predominantly
female. Notable exceptions exist, however. Mongolia, the Rus-

The proportion sian Federation, a number of other former Soviet republics and
Sudan report more female than male doctors. Moreover, women
of female doctors in are making substantial progress in some regions. The propor-
tion of female doctors in Europe increased steadily during the
Europe increased 1990s, as did the proportion of female students in medical
schools (12). In the United Kingdom, for example, women now
steadily during constitute up to 70% of medical school intakes (13).
From the limited information that exists on the ages of
the 1990s health workers in different settings, no general patterns can
be observed, though some information is available for specific
countries. For example, an increase in the average age of the
nursing workforce over time has been noted in a number of
OECD countries, including the United Kingdom and the United States (14, 15). Policies
relating to the official age of retirement are considered in Chapter 5.
It has not been possible to document trends over time in the mix of health
professionals or their characteristics in enough countries to allow a global analysis.
Systems for recording and updating health worker numbers often do not exist, which
presents a major obstacle to developing evidence-based policies on human resource
development.

Figure 1.1 Distribution of women in health service professions,

by WHO region
100
90
Women (% of health service providers)

80
70
60
50
40
30
20
10
0
Africa Americas South-East Asia Europe Eastern Western
Mediterranean Pacific
Doctors Nurses Others
Data source: (3).
 health workers: a global profile 

Table 1.2 Proportion of government health expenditure paid to health workers

Wages, salaries and allowances of
employees as percentage of
general government health Number of countries
WHO region expenditure (GGHE) with available data
Africa 29.5 14
South-East Asia 35.5 2
Europe 42.3 18
Western Pacific 45 7
Americas 49.8 17
Eastern Mediterranean 50.8 5
World 42.2 64
Note: Grouped proportions are simple averages of the country proportions, showing the ratio in a typical country in the region.

How much is spent on the

health workforce?
The large numbers of health workers in the world make up an important part of
the total labour force. In general, the relative importance of the health workforce
is higher in richer countries than in poorer ones and can account for up to 13%
of the total workforce. Payments of salaries and other benefits to health workers
are also a significant component of total government health expenditure (including
capital costs) (Table 1.2). A typical country devotes just over 42% of total general
government health expenditure to paying its health workforce, though there are
regional and country variations around this average (16). For example, governments
in Africa and South-East Asia typically devote lower proportions than do those in
other regions.
Information on the non-government (i.e. private) sector by itself is not available.
Data have been obtained, however, from 43 countries on the
share of total health expenditure (including capital costs)
from all sources, government and non-government, paid in
salaries and other allowances. On average, payments to the
A typical country
health workforce account for just under 50% of total health devotes just over
expenditure, suggesting that payments to health workers in the
non-government sector make up a higher proportion of total 42% of total general
expenditures than in the government sector. However, there is
little overlap between the 43 countries described here and those government health
included in Table 1.2 because of the way data are reported by
different countries, so this information should be interpreted
expenditure to paying
carefully. It should also be remembered that payments made by
households directly to providers, and which are not captured in
its health workforce
official records of salaries, are not included in this analysis.
Trends over time (19982003) in the ratio of wages, salaries and allowances
to government health expenditure are available for only 12 countries. Trends in the
share of total health expenditure paid to health workers as wages, salaries and
allowances are available for another 24. Neither set of figures shows any consistent
pattern. The share rose in some countries and fell in others, and the average across
all countries is remarkably stable.
 The World Health Report 2006

Where are the health workers?

Uneven distribution across the globe
Health workers are distributed unevenly (17). Countries with the lowest relative
need have the highest numbers of health workers, while those with the greatest
burden of disease must make do with a much smaller health
workforce. This pattern is summarized in Figure 1.2, where the
Countries with the vertical axis shows burden of disease, the horizontal axis the
number of health workers, and the size of the dots represents
lowest relative need total health expenditure. The Region of the Americas, which

have the highest includes Canada and the United States, contains only 10% of
the global burden of disease, yet almost 37% of the worlds
numbers of health health workers live in this region and spend more than 50%
of the worlds financial resources for health. In contrast, the
workers African Region suffers more than 24% of the global burden of
disease but has access to only 3% of health workers and less
than 1% of the worlds financial resources even with loans
and grants from abroad.

Uneven spread within countries

Within regions and countries, access to health workers is also unequal. For example,
Viet Nam averages just over one health service provider per 1000 people, but this
figure hides considerable variation. In fact, 37 of Viet Nams 61 provinces fall below
this national average, while at the other extreme one province counts almost four health
service providers per 1000 (20). Similar variations exist in other countries (21).
Many factors influence the geographical variation that is observed in health
worker density. Areas with teaching hospitals and a population that can afford to pay
for health services invariably attract more health workers than regions without such
facilities or financial support. As a result, health worker density is generally highest
in urban centres where teaching hospitals and high incomes are most common.
Although the extent of urbanization increases across countries with increasing
income, in countries of all income levels the proportion of health professionals living
in urban areas exceeds the proportion of the general population found there. This
is particularly the case for doctors, as shown in Figure 1.3, where the red dotted
line shows that, while under 55% of all people live in urban areas, more than 75%
of doctors, over 60% of nurses and 58% of other health workers also live in urban
areas.
In many countries, female health service providers are particularly scarce in rural
areas, a situation that may arise in part because it is unsafe for female workers to
live alone in some isolated areas. The picture may well be different if traditional birth
attendants and village volunteers could be included in the calculations, as these are
the domains of women in many countries, but this information is rarely routinely
available. Moreover, there are some notable exceptions. For example, Ethiopia and
Pakistan are among the countries that have actively sought to recruit and train female
health workers in rural areas: they are called health extension workers in Ethiopia
and lady health workers and lady health visitors in Pakistan.
WHO is developing a tool to help countries to identify their health service
resources, including where their health workers are to be found (see Box 1.3).
 health workers: a global profile 

Figure 1.2 Distribution of health workers by level of health

expenditure and burden of disease, by WHO region
35

30 South-East Asia
% of global burden of disease

25 Africa

20 Western Pacific

15 Europe
Americas

10 Eastern Mediterranean

0
0 5 10 15 20 25 30 35 40 45
% of global workforce
Data sources: (3, 18, 19).

Are there enough health workers?

How important is the uneven geographical distribution of health workers within
countries? Perfect equality is not feasible, and in some cases it is not even desirable.
For instance, teaching hospitals must be strategically located, and a concentration of
certain types of health workers around hospitals can be completely acceptable. But
while some degree of geographical variation in health worker numbers is appropri-
ate, the question remains: to what degree?
Although available data do not allow a simple response, coverage rates of key
interventions are generally lower in areas with relatively low numbers of health
workers, compared with those that have higher concentrations. This relationship
can be observed across countries and within them. For example, researchers have
recently found that countries with a higher density of health workers attain higher
levels of measles vaccination and coverage with antenatal care (2325).

Figure 1.3 Ruralurban distribution of health service providers

80

75
% in urban localities

70

65

60
Line of spatial equality
(% of world population
55 in urban localities)
Doctors Nurses Others
50 Data sources: (3, 22).
10 The World Health Report 2006

The correlation between the availability of health workers and coverage of health
interventions suggests that the publics health suffers when health workers are
scarce (20, 21, 2527). This raises the more fundamental issue of whether there are
enough health workers. Methodologically, there are no gold standards for assessing
sufficiency. The following section examines sufficiency from the perspective of
essential health needs.

Box 1.3 Where are the health workers? Service Availability Mapping
To help national decision-makers obtain information rapidly, systems/serviceavailabilitymapping/en/index.html.
WHO is working with countries to develop a tool called Ser- A rapid version of a national SAM has been applied
vice Availability Mapping (SAM). Using WHOs Health Mapper in a dozen countries, providing a rich picture of services
(a Geographic Information System-based software package) across districts. Data on human resources include the
and a questionnaire loaded on personal digital assistants, density and distribution of health workers by major cadre
district health teams collect critical information on health and training exposure in the last two years, unfilled posts
resources, public health risks and programme implementa- and absentee rates. The figure shows the density of doc-
tion, in order to provide updated maps of health services. tors, clinical officers, registered and enrolled nurses and
For more information, see http://www.who.int/healthinfo/ midwives, combined, per 1000 population in Zambia.

Source: Ministry of Health, Zambia, in collaboration with WHO.

Map production: Public Health Mapping and GIS/WHO.
 health workers: a global profile 11

Figure 1.4 Population density of health care professionals

required to ensure skilled attendance at births
100
Coverage of births by skilled birth attendants (%)

80
Minimum desired level of coverage

60

40
Threshold estimate
(2.28)
20
Lower bound Upper bound
(2.02) (2.54)
0
0 1 2 3 4
Doctors, nurses and midwives per 1000 population
Data sources: (3, 30, 31).

Needs-based sufficiency
Various estimates of the availability of health workers required to achieve a package
of essential health interventions and the Millennium Development Goals (including
the scaling up of interventions for HIV/AIDS) have resulted in the identification of
workforce shortfalls within and across mostly low income countries. In the HIV/AIDS
literature, scaling up treatment with antiretrovirals was estimated to require between
20% and 50% of the available health workforce in four African countries, though less
than 10% in the other 10 countries surveyed (28). In more general terms, analysts
estimated that in a best case scenario for 2015 the supply of health workers would
reach only 60% of the estimated need in the United Republic of Tanzania and the
need would be 300% greater than the available supply in Chad (29). Furthermore,
The world health report 2005 estimated that 334 000 skilled birth attendants would
have to be trained globally over the coming years merely to
reach 72% coverage of births (18).
To achieve a global assessment of shortfall, the Joint
WHO estimates a
Learning Initiative (JLI), a network of global health leaders,
launched by the Rockefeller Foundation, suggested that, on
shortage of more than
average, countries with fewer than 2.5 health care professionals
(counting only doctors, nurses and midwives) per 1000
4 million doctors,
population failed to achieve an 80% coverage rate for deliveries nurses, midwives
by skilled birth attendants or for measles immunization (24).
This method of defining a shortage, whether global or by and others
country, is driven partly by the decision to set the minimum
desired level of coverage at 80% and partly by the empirical identification of health
worker density associated with that level of coverage. Using a similar threshold
12 The World Health Report 2006

Figure 1.5 Countries with a critical shortage of health service providers

(doctors, nurses and midwives)

Countries with critical shortage

Countries without critical shortage Data source: (3).

method and updated information on the size of the health workforce obtained for
this report, the JLI analysis has been repeated for skilled birth attendants (see Figure
1.4). A remarkably similar threshold is found at 2.28 health care professionals per
1000 population, ranging from 2.02 to 2.54 allowing for uncertainty.
The 57 countries that fall below this threshold and which fail to attain the 80%
coverage level are defined as having a critical shortage. Thirty-six of them are in
sub-Saharan Africa (Figure 1.5). For all these countries to reach the target levels
of health worker availability would require an additional 2.4 million professionals
globally (Table 1.3). (Based on the upper and lower limits of the threshold, the upper
and lower limits of the estimated critical shortage are 3million and 1.7million,
respectively.) This requirement represents only three types of health service provider.
Multiplying the 2.4 million shortage by 1.8, which is the average ratio of total health
workers to doctors, nurses and midwives observed in all WHO regions (except
Europe, where there are no critical shortages based on these criteria), the global
shortage approaches 4.3 million health workers.
In absolute terms, the greatest shortage occurs in South-East Asia, dominated
by the needs of Bangladesh, India and Indonesia. The largest relative need exists
in sub-Saharan Africa, where an increase of almost 140% is necessary to meet the
threshold.
These estimates highlight the critical need for more health workers in order to
achieve even modest coverage for essential health interventions in the countries most
in need. They are not a substitute for specific country assessments of sufficiency,
nor do they detract from the fact that the effect of increasing the number of health
 health workers: a global profile 13

Table 1.3 Estimated critical shortages of doctors, nurses and midwives, by WHO region
Number of countries In countries with shortages
Percentage
Estimated increase
WHO region Total With shortages Total stock shortage required
Africa 46 36 590 198 817 992 139
Americas 35 5 93 603 37 886 40
South-East Asia 11 6 2 332 054 1 164 001 50
Europe 52 0 NA NA NA
Eastern
21 7 312 613 306 031 98
Mediterranean
Western Pacific 27 3 27 260 32 560 119
World 192 57 3 355 728 2 358 470 70
NA, not applicable.
Data source: (3).

workers depends crucially on other determinants such as levels of income and

education in the community (21, 25). Furthermore, economic factors also enter
the equation: shortfalls based on need can co-exist with unemployment of health
workers due to local market conditions (see Chapter 6 for further discussion).

Addressing the shortage

how much will it cost?
Making up the shortfall through training requires a significant investment. Assum-
ing very rapid scaling up in which all the training is completed by 2015, the annual
training costs range from a low of US$ 1.6 million per country per year to almost
US$ 2 billion in a large country like India. The average cost per country of
US$ 136 million per year is of the same order of magnitude as the estimated cost
of Malawis Emergency Human Resources Programme (see Chapter 2). Financing it
would require health expenditures to increase by US$ 2.80 per person annually in
the average country (the range is from US$ 0.40 to just over US$ 11) an increase
of approximately 11% on 2004 levels (34).
The estimate is limited to doctors, nurses and midwives, the occupations for
which data on workforce numbers and training costs are most complete, so can be
considered a lower limit. In the calculations, the target number of health workers has
been adjusted upwards to account for population growth between 2005 and 2015,
and student intakes have also been adjusted upwards to account for attrition during
and after training. Region-specific training costs that include a capital component
have been used where possible, though data are limited and the results should be
interpreted with caution.
These estimates also assume that present trends and patterns of training will
continue. Other ways of helping to tackle the observed shortages, including those
aimed at increasing the productivity and motivation of the current workforce, or
changing the skill mix of health workers, are described in subsequent chapters.
The additional annual cost of employing the new doctors, nurses and midwives
once training has been fully scaled up is just over US$ 311 million per country
in 2004 prices. By 2015, to pay the salaries of the scaled-up workforce would
14 The World Health Report 2006

require a minimum increase of US$ 7.50 per person per year in the average country.
This can be taken to be a lower limit cost because some level of salary increase is
likely to be necessary to retain the additional health workers in the health sector
and in the country. The extent of the required increase is difficult to determine,
partly because salaries in the deficit countries can be up to 15 times lower than
those in countries that are popular destinations for migrants (32). The Millennium
Project assumed salaries would need to double if the Millennium Development Goals
were to be achieved (33), which would increase the current annual salary cost by
US$ 53 billion in the 57 countries. To put this figure into perspective, this represents
an increase in the annual global wage bill for health workers of less than 5%. It would
also require an increase in annual health spending by 2015 of US$ 20 per person in
the average country an increase of over 75% on 2004 levels.
These figures need to be interpreted with caution, particularly because labour
markets for health workers are evolving rapidly as globalization increases. It is very
likely, for example, that salaries in some of the countries where shortages were not
identified would have to be increased as well, to ensure that their workers did not
migrate to some of the deficit countries. This type of effect is difficult to predict, but
the numbers reported here clearly show the need for the international community
actively to support the process of strengthening human resources for health.

AFP 2005

Dr John Awoonor-Williams is the only doctor at Nkwanta District Hospital, Ghana, serving a population of 187 000 in a remote,
vast area in the northern part of the Volta Region.
 health workers: a global profile 15

Conclusion
The global profile presented here shows that there are more than 59 million health
workers in the world, distributed unequally between and within countries. They are
found predominantly in richer areas where health needs are less severe. Their num-
bers remain woefully insufficient to meet health needs, with the total shortage being
in the order of 4.3 million workers.
The profile also shows how much is not known. Information on skill mix, age
profiles, sources of income, geographical location, and other characteristics that
are important for policy development is far from complete. One reason for this is
the variation between countries in the definitions used to categorize health workers,
which makes it difficult to ensure that the same people are being included as part
of the health workforce in different settings. WHO is confronting this issue by
developing a standard classification of health workers in collaboration with countries
and other partners.
The other reason is simply the lack of data. In some countries, information on
the total size of the health workforce is not routinely collected, while little is known
about certain categories of health workers even in countries with extensive data
reporting systems. The lack of reliable, up-to-date information greatly restricts the
ability of policy-makers at national and international levels to develop evidence-
based strategies to resolve the health workforce crisis, or to develop health systems
to serve the needs of disadvantaged populations. Relatively small investments by the
global community in this area could well have substantial returns. Chapter 7 returns
to this issue and suggests some possible solutions. Meanwhile, Chapter 2 discusses
some of the most important challenges that face the global health workforce today.

References
1. Dal Poz MR, Kinfu Y, Drger S, Kunjumen T, Diallo K. Counting health workers:
definitions, data, methods and global results. Geneva, World Health Organization, 2006
(background paper for The world health report 2006; available at: http://www.who.
int/hrh/documents/en/).
2. International Labour Organization. International Standard Classification of Occupations
(http://www.ilo.org/public/english/bureau/stat/class/isco.htm, accessed 19 January
2006).
3. World Health Organization. Global Atlas of the Health Workforce (http://www.who.
int/globalatlas/default.asp, accessed 19 January 2006).
4. Zurn P, Vujicic M, Diallo K, Pantoja A, Dal Poz MR, Adams O. Planning for human
resources for health: human resources for health and the projection of health outcomes/
outputs. Cahiers de Sociologie et de Dmographie mdicales, 2005, 45:107133.
5. Buchan J, Dal Poz MR. Skill mix in the health care workforce: reviewing the evidence.
Bulletin of the World Health Organization, 2002, 80:575580.
6. Dussault G, Adams O, Fournier P. Planejamento de recursos humanos em sade [Human
resources for health planning]. Translation, adaptation and preface by Dal Poz MR. Rio
de Janeiro, Universidade do Estado do Rio de Janeiro, Instituto de Medicina Social, 1997
(Srie Estudos em Sade Coletiva, 158).
7. Lave JR, Lave LB, Leinhardt S. Medical manpower models: need, demand and supply.
Inquiry, 1975, XII:97126.
8. Reinhardt UE.Projecting long-term trends in health manpower: methodological
problems. In: Culyer AJ, ed. The economics of health. Great Yarmouth, Edward Elgar
Publishing Ltd, 1991:26283.
16 The World Health Report 2006

9. McPake B , Asiimwe D, Mwesigye F, Ofumbi M, StreeflandP, Turinde A. Coping

strategies of health workers in Uganda. In: Ferrinho P, Van Lerberghe W, eds. Providing
health care under adverse conditions: health personnel performance and individual coping
strategies. Antwerp, ITG Press, 2000 (Studies in Health Services Organisation and
Policy, 16:131150).
10. Macq J, Van Lerberghe W. Managing health services in developing countries:
moonlighting to serve the public? In: Ferrinho P, Van Lerberghe W, eds. Providing health
care under adverse conditions: health personnel performance and individual coping
strategies. Antwerp, ITG Press, 2000 (Studies in Health Services Organisation and
Policy, 16:177186).
11. Ferrinho P, Van Lerberghe W, Fronteira I, Hipolito F, Biscaia A. Dual practice in the health
sector: review of the evidence. Human Resources for Health, 2004, 2:14.
12. Gupta N, Diallo K, Zurn P, Dal Poz MR. Assessing human resources for health: what can
be learned from labour force surveys? Human Resources for Health, 2003, 1:5.
13. Dubois C-A, McKee M, Nolte E. Analysing trends, opportunities and challenges. In:
Dubois C-A, McKee M, Nolte E, eds. Human resources for health in Europe. Brussels,
European Observatory on Health Systems and Policies, 2005.
14. Buerhaus P, Staiger D, Auerbach D. Implications of a rapidly aging registered nurse
workforce. JAMA, 2000, 283:29482954.
15. Buchan J. The greying of the United Kingdom nursing workforce: implications for
employment policy and practice. Journal of Advanced Nursing, 1999, 30:818826.
16. Hernandez P, Drger S, Evans DB, Tan-Torres T, Dal Poz MR. Measuring expenditure
for the health workforce: challenges and evidence. Geneva, World Health Organization,
2006 (background paper for The world health report 2006; available at: http://www.who.
int/hrh/documents/en/).
17. Speybroeck N, Ebener S, Sousa A, Paraje G, Evans DB, Prasad A. Inequality in access to
human resources for health: measurement issues. Geneva, World Health Organization,
2006 (background paper for The world health report 2006; available at: http://www.who.
int/hrh/documents/en/).
18. The world health report 2005 Make every mother and child count. Geneva, World
Health Organization, 2005:200203.
19. World Health Organization. Burden of Disease Statistics (http://www.who.int/healthinfo/
bod/en/, accessed 19 January 2006).
20. Prasad A, Tandon A, Sousa A, Ebener S, Evans DB. Measuring the efficiency of human
resources for health in attaining health outcomes across provinces in Viet Nam. Geneva,
World Health Organization, 2006 (background paper for The world health report 2006;
available at: http://www.who.int/hrh/documents/en/).
21. Sousa A, Tandon A, Prasad A, Dal Poz MR, Evans DB. Measuring the efficiency of health
workers in attaining health outcomes across sub national units in Brazil. Geneva, World
Health Organization, 2006 (background paper for The world health report 2006; available
at: http://www.who.int/hrh/documents/en/).
 health workers: a global profile 17

22. Population Reference Bureau. Search Population and Health Data (http://www.prb.
org/datafind/datafinder6.htm, accessed 20 January 2006).
23. Anand S, Brnighausen T. Human resources for health and vaccination coverage in
developing countries (unpublished document, Oxford University, 2005).
24. Chen L, Evans T, Anand S, Boufford JI, Brown H, Chowdhury M et al. Human resources
for health: overcoming the crisis. Lancet, 2004, 364:19841990.
25. Speybroeck N, Kinfu Y, Dal Poz MR, Evans DB. Reassessing the relationship between
human resources for health, intervention coverage and health outcomes. Geneva, World
Health Organization, 2006 (background paper for The world health report 2006; available
at: http://www.who.int/hrh/documents/en/).
26. Tandon A, Sousa A, Prasad A, Evans DB. Human resources and health outcomes in Brazil
and Viet Nam: a comparative perspective. Geneva, World Health Organization, 2006
(background paper for The world health report 2006; available at: http://www.who.
int/hrh/documents/en/).
27. Anand S, Brnighausen T. Human resources and health outcomes: cross-country
econometric study. Lancet, 2004, 364:16031609.
28. Smith O. Human resource requirements for scaling up antiretroviral therapy in low-
resource countries (Appendix E). In: Curran J, Debas H, Arya M, Kelley P, Knobler S, Pray
L, eds. Scaling up treatment for the global AIDS pandemic: challenges and opportunities.
Washington, DC, National Academies Press (Board of Global Health), 2004.
29. Kurowski C, Wyss K, Abdulla S, Ymadji N, Mills A. Human resources for health:
requirements and availability in the context of scaling-up priority interventions in low-
income countries. Case studies from Tanzania and Chad. London, London School of
Hygiene and Tropical Medicine, 2003 (Working Paper 01/04).
3 0. UNICEF. WHO UNICEF Estimates on Immunization Coverage 19802004 (http://www.
childinfo.org/areas/immunization/database.php, accessed 20 January 2006).
31. United Nations Population Division. World Population Prospects: the 2004 Revision.
Population Database (http://esa.un.org/unpp, accessed 20 January 2006).
32. Vujicic M, Zurn P, Diallo K, Adams O, Dal Poz MR. The role of wages in the migration of health
care professionals from developing countries. Human Resources for Health, 2004, 2:3.
3 3. Millennium Development Goals needs assessments: country case studies of Bangladesh,
Cambodia, Ghana, Tanzania and Uganda. New York, NY, Millennium Project, 2004
(http://www.unmillenniumproject.org/documents/mp_ccspaper_jan1704.pdf, accessed
23 February 2004).
34. Verboom P, Tan-Torres Edejer T, Evans DB. The costs of eliminating critical shortages
in human resources for health. Geneva, World Health Organization, 2006 (background
paper for The world health report 2006; available at: http://www.who.int/hrh/
documents/en/).