Articles

How much does quality of child care vary between health

workers with differing durations of training? An observational
multicountry study
Luis Huicho, Robert W Scherpbier, A Mwansa Nkowane, Cesar G Victora, and the Multi-Country Evaluation of IMCI Study Group*

Lancet 2008; 372: 910–16
See Editorial page 863
See Comment page 870
*Members listed at end of paper
Universidad Nacional Mayor de
San Marcos, Universidad
Peruana Cayetano Heredia and
Instituto de Salud del Niño,
Lima, Peru (Prof L Huicho MD);
Department of Child and
Adolescent Health and
Development
(R W Scherpbier MD) and
Human Resources for Health
Department
(A M Nkowane MA), WHO,
Geneva, Switzerland; and
Universidade Federal de
Pelotas, Pelotas, Brazil
(Prof C G Victora MD)
Correspondence to:
Dr Luis Huicho, Universidad
Nacional Mayor de San Marcos,
Universidad Peruana Cayetano
Heredia and Instituto de Salud
del Niño, Av Brasil 600, LI05
Lima, Peru
lhuicho@gmail.com
Summary
Background Countries with high rates of child mortality tend to have shortages of qualified health workers. Little
rigorous evidence has been done to assess how much the quality of care varies between types of health workers. We
compared the performance of different categories of health workers who are trained in Integrated Management of
Childhood Illness (IMCI).
Methods We analysed data obtained from first-level health facility surveys in Bangladesh (2003), Brazil (2000), Uganda
(2002), and Tanzania (2000). We compared the clinical performance of health workers with longer duration of
preservice training (those with >4 years of post-secondary education in Brazil or >3 years in the other three countries)
and shorter duration (all other health workers providing clinical care). We calculated quality of care with indicators of
assessment, classification, and management of sick children according to IMCI guidelines. Every child was examined
twice, by the IMCI-trained health worker being assessed and by a gold-standard supervisor.
Findings 272 children were included in Bangladesh, 147 in Brazil, 231 in Tanzania, and 612 in Uganda. The proportions
of children correctly managed by health workers with longer duration of preservice training in Brazil were 57·8%
(n=43) versus 83·7% (n=61) for those with shorter duration of training (p=0·008), and 23·1% (n=47) versus 32·6%
(n=134) (p=0·03) in Uganda. In Tanzania, those with longer duration of training did better than did those with shorter
duration in integrated assessment of sick children (mean index of integrated assessment 0·94 [SD 0·15] vs 0·88
[0·13]; p=0·004). In Bangladesh, both categories of health worker did much the same in all clinical tasks. We recorded
no significant difference in clinical performance in all the other clinical tasks in the four countries.
Interpretation IMCI training is associated with much the same quality of child care across different health worker
categories, irrespective of the duration and level of preservice training. Strategies for scaling up IMCI and other
child-survival interventions might rely on health workers with shorter duration of preservice training being deployed
in underserved areas.

Funding Bill & Melinda Gates Foundation and the US Agency for International Development.

Introduction effective interventions. Task shifting, or the devolution of

About 10 million children die every year, with most deaths
occurring in low-income countries—mainly sub-Saharan
Africa and south Asia.1 However, two-thirds of these
deaths could be avoided if all children had access to
cost-effective preventive and curative interventions that
have been available for several years.2
Maternal, infant, and child survival increase with the
ratio of health workers to the population.3 On the basis of
empirical data, WHO concluded that countries with
fewer than 2·3 health-care professionals per 1000 people
are less likely to achieve 80% coverage rates for deliveries
by a skilled attendant than are those with a higher ratio of
health-care professionals.4 The coverage of measles
immunisation has also been associated with health-worker
ratio.5 Many high-mortality countries have serious
health-worker shortages—eg, the African region accounts
for about half of all deaths in children younger than
5 years and has only 3% of the global health workforce.4,6
An insufficient number of motivated, well-trained
health personnel is a major limitation for scaling up
selected clinical responsibilities to health workers with
shorter duration of preservice training (ie, training
undertaken before obtaining a degree from school), is
increasingly seen as an option to address the shortages of
personnel.7
The Integrated Management of Childhood Illness
(IMCI) is a global strategy that has been adopted by more
than 100 countries. IMCI has three components: im-
proving performance of health workers, strengthening
health systems, and improving family and community
practices that are relevant to child health. Its clinical
guidelines describe how to assess, classify, and manage
children younger than 5 years who have common
illnesses. The guidelines are intended for use by all types
of health workers (doctors, medical assistants, nurses, or
literate paramedical workers) who, after being trained in
IMCI, provide care in first-level outpatient health
facilities.8 Emphasis is given to the use of clinic visits by
sick children, irrespective of the type of illness, as an
opportunity for the delivery of preventive interventions.

910 www.thelancet.com Vol 372 September 13, 2008


The Multi-Country Evaluation (MCE) of IMCI assessed
the effectiveness and cost of the three components of
IMCI in five countries (Bangladesh, Brazil, Peru,
Tanzania, and Uganda). Apart from Peru, surveys
administered in health facilities gathered data for
health-worker performance allowing the assessment of
the training component of IMCI. These surveys are done
in countries with the highest number of deaths in
children younger than 5 years worldwide.1 Doctors and
nurses were the only health personnel trained in IMCI in
Brazil, whereas paramedical workers were also trained in
the other three countries. IMCI training was successfully
scaled up in Tanzania and Bangladesh. In Brazil, achieved
training coverage was low (<10%), and in Uganda the fast
scaling up of training had a negative effect on training.9
Precise estimates of the proportions of first-level facility
workers who were trained in IMCI are not available since
none of the four countries that we studied routinely
collected comprehensive training statistics, indicating
inadequacies in their monitoring systems.
In addition to high and equitable intervention coverage,
good quality of care is a criterion for improving child
health.10 Little rigorous evidence has been done to assess
the comparative clinical performance of physicians and
nurses, especially in primary health care. A recent
systematic review showed that doctor–nurse substitution
in primary care resulted in much the same quality of care,
although the study had several methodological problems.11
A report from the Brazil IMCI assessment suggested that
nurses did as well as doctors, if not better, in several
clinical tasks.12 Task shifting of clinical responsibilities is
not restricted to doctors and nurses. There is increasing
advocacy for a greater involvement of non-physician
clinicians, especially in sub-Saharan Africa,13 where such
personnel could help to scale up interventions for child
survival. However, task shifting takes place within a
context and needs to be accompanied by other measures
such as standardised training, supervision, certification
and assessment, and adaptations of incentives for it to be
successful.7 Rigorously designed studies on quality of care
that is provided by these health workers are urgently
needed for task shifting to be promoted as an
evidence-based strategy for improvement of child health.
On the basis of data from four MCE countries
(Bangladesh, Brazil, Tanzania, and Uganda), we aimed to
assess how different categories of IMCI-trained health
workers vary in terms of quality of child care.
Methods
Study design and setting
Our analyses are restricted to the first component of
IMCI—namely, improvement of health-worker per-
formance. We analysed data obtained as part of the MCE
through first-level health facility surveys in Bangladesh
(2003), Brazil (2000), Uganda (2002), and Tanzania (2000)
to compare the performance of clinical tasks by health
workers. We assessed health workers who had received
www.thelancet.com Vol 372 September 13, 2008
Articles
Panel: Index of integrated child assessment14
This index assesses the quality and completeness of the assessment provided to sick
children. Each of the 14 items below is coded as 0 (not assessed) or 1 (assessed). Three
additional items on feeding assessment are included for children younger than 2 years.
For all children
1 Check for ability to drink or breastfeed
2 Check whether the child vomits everything
3 Check whether the child has had convulsion
4 Check for cough or difficult breathing
5 Check for diarrhoea
6 Check for fever
7 Child weighed on the same day as the survey visit
8 Child’s weight checked against a recommended growth chart
9 Check for palmar pallor
10 Check for visible severe wasting
11 Check for oedema of both feet
12 Child’s vaccination status checked
13 Child’s temperature checked
14 Check for other problems
For children younger than 2 years of age
15 Ask about breastfeeding
16 Ask if the child takes any other foods or fluids
17 Ask whether feeding has changed during the illness
in-service IMCI training. The surveys were undertaken
in each country by trained surveyors who assessed the
quality of care that was provided to sick children. We
included either all health government facilities in the
study areas (14 facilities in Bangladesh) or a probability
sample of these facilities (88 facilities in Brazil, 39 in
Tanzania, and 124 in Uganda).
The study was approved by the ethics review boards in
the four countries. Health workers and patients provided
verbal consent to have the consultations observed by a
third party and to the re-examination procedure.
Data collection and procedures
Data collection entailed observation of case management
by IMCI-trained health-care workers and re-examination
of children by a trained gold-standard surveyor—an
IMCI-trained physician supervisor who was unaware of
the original diagnosis and treatment by the facility
worker. The surveys measure the performance of health
workers in clinical tasks—eg, how well they assess,
classify, and manage sick children and counsel their
caretakers. The generic data-collection tools are available
from the MCE website. For the Multi-Country
Country-level data were revised and checked for Evaluation website see http://
www.who.int/imci-mce
consistency. We calculated indicators of health-worker
performance, including the index of integrated
assessment and indicators of correct classification and
management for IMCI priority and non-priority
illnesses.14,15
We used standard definitions for calculation of the
performance indicators. The index of integrated child
911
 Articles

assessment measures the completeness of the assess-

Level of preservice Duration of Number of Study classification
training preservice children ment of sick children, and was derived as the mean of
training* examined 14 tasks that have to be routinely undertaken on all sick
Bangladesh children according to the IMCI guidelines (panel).14 This
Medical assistants Post-secondary school 4 years 84 Longer duration training index was validated and showed high reliability.14 Correct
Subassistant community Post-secondary school 4 years 117 Longer duration training classification is defined as the proportion of children
medical officer whose diagnostic classifications according to the health
Family welfare visitors Post-secondary school 18 months 71 Shorter duration training worker match all those that were given by the gold-
Brazil standard surveyor. This indicator was presented
Medical doctors University 6 years 74 Longer duration training separately for IMCI priority and non-priority diseases.14
Nurses University 4 years 73 Shorter duration training Correct management is defined as provision of the
Tanzania correct drug, in the correct formulation (dose, times per
Clinical officers Post-secondary school 2–4 years 57 Longer duration training day, number of days), and for which the health worker
or in-service upgrade explained correctly to the patient or carer how to
Assistant clinical officers Post-secondary school 2–4 years 110 Longer duration training administer the drug at home.14
or in-service upgrade We aimed to assess the performance of health workers
Maternal and child Post-primary school 2 years 12 Shorter duration training providing care to children. In Brazil, only doctors and
health aides or in-service upgrade
nurses provided this type of care, but several other
Nurse assistants Post-primary school 1 year 52 Shorter duration training
or in-service upgrade
categories of health workers were involved in the other
three countries. The assessment of educational levels of
Uganda
the health workforce is essential because the knowledge
Physicians University 5 years 11 Longer duration training
and skills that are acquired during preservice training (ie,
Clinical officers Post-secondary school 36 months 191 Longer duration training
before degree from school is obtained) affect health
Nurses Post-secondary school 18–36 months 136 Shorter duration training
workers’ performance.16,17 In our analyses, IMCI-trained
Midwives Post-secondary school 18–36 months 79 Shorter duration training
health workers were categorised into two groups—those
Nurse assistants Post-primary school 3 months 181 Shorter duration training
with longer duration of training and those with shorter
Others Variable Variable† 14 Shorter duration training
duration of training—on the basis of their professional
Data from MCE health facility surveys. *Values are provided by key informants at the Ministry of Health in each country category, level, and duration of preservice training
who described the typical duration of preservice training for each category. †This category in Uganda includes workers (table 1). Because of variability in the duration of
with variable qualifications who were providing child care in the facilities when the survey took place. Some of these
workers were janitors or clerical staff with no former preservice training, and thus we cannot provide more detailed
preservice training, different cutoff points were used in
information about their duration of training. every country. We categorised health workers with more
than 4 years of post-secondary education in Brazil, or
Table 1: Description of the types of IMCI-trained health workers, by country
with 3 or more years on average in the other three
countries, as the group with longer duration of training.
All other health workers providing clinical child care
Longer duration Shorter duration p value
training training
were included in the group with shorter duration of
training (table 1). Throughout this Article, we refer to
Index of integrated assessment of children*
duration and level of preservice training to describe the
Bangladesh (n=272) 0·73 (0·15) 0·72 (0·12) 0·62
comparisons being made.
Brazil (n=147) 0·48 (0·18) 0·53 (0·18) 0·05
Tanzania (n=231) 0·94 (0·13) 0·88 (0·13) 0·004
Statistical analysis
Uganda (n=612) 0·59 (0·25) 0·60 (0·23) 0·56
The units of analyses were consultations with children
Children correctly classified younger than 5 years. Because data collection took place
Bangladesh (n=272) 145 (71·9%) 47 (66·7%) 0·41 on one day in every facility, all children were seen only
Brazil (n=147) 45 (61·4%) 53 (73·2%) 0·14 once. 272 children were included in Bangladesh, 147 in
Tanzania (n=231) 126 (75·6%) 51 (79·7%) 0·53 Brazil, 231 in Tanzania, and 612 in Uganda. All surveys
Uganda (n=612) 91 (45·0%) 161 (39·3%) 0·25 were reported separately.12,15,18,19 We used t tests from
Children correctly managed individual samples to compare the index of integrated
Bangladesh (n=272) 126 (62·7%) 48 (67·6%) 0·47 management between groups with longer and shorter
Brazil (n=147) 43 (57·8%) 61 (83·7%) 0·008 duration of training, whereas we compared the
Tanzania (n=231) 107 (64·4%) 40 (62·7%) 0·82 performance in correct classification and correct
Uganda (n=612) 47 (23·1%) 134 (32·6%) 0·03 management between the same groups with χ² tests.
Control for confounding variables was done with the
Data are mean (SD) or number (%) of children. All values are unadjusted. *Panel
provides an explanation of the index of integrated assessment of children. general linear model20 when the dependent variable was
continuous (index of integrated assessment). In this
Table 2: Integrated assessment, classification, and management of children
analysis, type of health worker (recoded into two
by IMCI-trained health workers, stratified by length of preservice training
categories: shorter and longer duration of preservice

912 www.thelancet.com Vol 372 September 13, 2008

 Articles

Results
Adjusted mean difference (95%CI) p value
between health workers with shorter
Table 1 shows the different categories of health workers
and longer preservice training providing care in every country, the duration of their
Bangladesh* –0·01 (–0·05 to 0·02) 0·46
preservice training, and the number of children that they
Brazil† 0·19 (0·08 to 0·30) 0·001
saw. In Bangladesh, children were seen by medical
Tanzania‡ –0·06 (–0·09 to –0·02) 0·004
assistants and subassistant community medical officers,
Uganda§ 0·01 (–0·03 to 0·06) 0·51
both of whom received 4 years of post-secondary training,
and by family welfare visitors, who received 18 months of
Adjusted mean difference through general linear model. *Adjusted for child age, post-secondary training. Physicians did not manage sick
and child sex. †Adjusted for child age, child sex, municipality, and facility type.
children in the first-level facilities who were enrolled in
‡Adjusted for child age, child sex, district, and facility type. §Adjusted for child
age, child sex, and district category. the Bangladesh study site. In Brazil, IMCI was provided
by doctors and nurses who received 6 and 4 years of
Table 3: Index of integrated assessment of children by IMCI-trained
university training, respectively. In Tanzania, children
health workers
were managed by clinical officers and assistant clinical
officers (who received 2–4 years of post-secondary
Crude odds p value Adjusted odds p value training), as well as by maternal and child health aides
ratios (95% CI) ratios (95% CI)
(2 years of post-primary school training) and nurse
Correct classification assistants (1 year of post-primary school training). In
Bangladesh 1·28 (0·71–2·31) 0·41 1·28 (0·70–2·31) 0·42 Uganda, clinical officers (3 years of post-secondary
Brazil 0·58 (0·29–1·19) 0·14 0·39 (0·09–1·66) 0·20 training), nurses and midwives (18–36 months of
Tanzania 0·79 (0·38–1·64) 0·53 0·98 (0·39–1·82) 0·98 post-secondary training), and nurse assistants (3 months
Uganda 1·26 (0·85–1·86) 0·25 1·28 (0·87–1·90) 0·22 of post-primary training) accounted for almost all IMCI
Correct management care provision, which was complemented by physicians
Bangladesh 0·8 (0·45–1·45) 0·47 0·81 (0·45–1·47) 0·49 (5 years of university training) and other IMCI-trained
Brazil 0·27 (0·09–0·73) 0·008 0·11 (0·02–0·73) 0·02 health workers.
Tanzania 1·02 (0·86–1·21) 0·82 1·56 (0·70–3·49) 0·28 Table 2 shows the clinical performance of IMCI-trained
Uganda 0·62 (0·41–0·95) 0·03 0·59 (0·38–0·90) 0·01 health workers by type, grouped by duration and level of
preservice training. In Brazil, nurses did better than
Adjustment for confounding factors was made through logistic regression.
doctors did in the assessment and management of sick
Table 4: Odds ratios for correct classification and management by IMCI- children, whereas they both had much the same
trained health workers with longer duration of training compared with performance in the correct classification indicator
those with shorter duration of training
(table 2). In Uganda, health workers with shorter duration
of training did better in correct management than did
training) was entered in the model as the random factor, those with longer duration of training, but both categories
district or municipality as a fixed factor, and the covariates showed levels of performance that were well below those
were type of health facility, age of child, and sex of child. reported from the other countries (table 2). In Bangladesh,
We used logistic regression for dichotomous outcomes health workers with shorter duration of training did as
(correct classification and management). In Bangladesh,
information about type of health facility was not available. 100 Rural physicians
In Uganda, most information about facility type was Rural nurses
missing and could not be included in the adjusted model. 90

Because most health workers examined more than one 80

Rural physicians/nurses (%)

child, the analyses should consider such clustering. We 70

were able to do this analysis for Uganda and Bangladesh,
60
but the Brazil and Tanzania datasets did not include an
identification variable that allowed these analyses. The 50
analyses that we present are not clustered to keep 40
consistency across the four countries. The statistical
30
analyses were done with SPSS (version 11.0) and Stata
(version 9.0). 20

10
Role of the funding source
0
The sponsor of the study had no role in the study design, Brazil (2000) Honduras (2000) Algeria (2002) The Gambia Tanzania (2002) Uganda (2000)
data collection, data analysis, data interpretation, or (2003)
writing of the report. The corresponding author had full Country (latest year with data)
access to all the data in the study and had final Figure 1: Proportion of total physicians and nurses working in rural areas
responsibility for the decision to submit for publication. The red bars indicate the proportion of the population in rural areas in 2000. Data from reference 29.

www.thelancet.com Vol 372 September 13, 2008 913

 Articles

classification, and management of sick children—was

20
18
consistently better in areas with IMCI than in comparison
Ratio of rural nurses to physicians

16 areas in the four countries studied.12,15,18,19 This study

14 expands that evidence base by showing that the quality
12 of care provided by health workers who received
10 in-service IMCI training did not differ systematically
8 between those with longer and shorter duration of
6 preservice training. Three different outcomes—correct
4 assessment, classification, and management—were
2
studied in four countries. Only one of these
0
Brazil Honduras Algeria The Gambia Tanzania Uganda 12 comparisons showed a significant benefit with
(2000) (2000) (2002) (2003) (2002) (2000) workers who were trained for a longer time (correct
Country (latest year with data) assessment in Tanzania), whereas two showed a benefit
with workers who were trained for a shorter time (correct
Figure 2: Ratio of nurses to physicians in rural areas management in Uganda and Brazil). The other nine
Data from reference 26.
comparisons showed no significant differences after
adjustment for confounders. Even though the training
Mortality Number of Nurses per Midwives Physicians Percentage of
categories are not consistent across countries because of
rate in child deaths 1000 per 1000 per 1000 human resource the differences between health workforces, the overall
children since 2000 population population population threshold finding is consistent with little effect of duration and
<5 years (rank) (year) (year) (year) achieved
level of preservice training on the quality of IMCI care.
(per 1000
livebirths) We can provide some possible explanations for our
results. First, the IMCI training modules could be
Bangladesh 77 343 000 (7) 0·14 (2004) 0·18 (2004) 0·26 (2004) 25%
sufficiently comprehensive and easy to understand for
Tanzania 126 223 000 (9) 0·30 (2002) 0·07 (2002) 0·02 (2002) 17%
health workers, irrespective of their previous duration and
Uganda 138 145 000 (14) 0·55 (2004) 0·16 (2004) 0·08 (2004) 34%
level of training. Second, in previous reports, longer
Brazil 34 127 000 (17) 3·84 (2000) NA 1·15 (2000) 217%
training was associated with a greater caseload and thus
Data are from references 1, 4, and 26. NA=not available. shorter time per visit,21 implying that the amount of time
that health workers with less training spend with children
Table 5: Ratio of physicians and nurses, and ranking of child deaths in the four study countries
younger than 5 years is higher than that of their colleagues
who have been trained for longer. This tenet was confirmed
well as did those with longer duration in all clinical tasks in additional analyses of our datasets from Bangladesh
(table 2). In Tanzania, health workers with longer duration and Tanzania (Adam T, Alliance for Health Policy and
of training did better than did those with shorter duration Systems Research, Geneva, Switzerland, personal
of training in integrated assessment of sick children, but communication). Good quality of care might simply mean
we recorded no difference in clinical performance in all spending sufficient time on case management of
the other clinical tasks that were assessed (table 2). children—an as yet unchallenged hypothesis.22
Tables 3 and 4 show the comparison of performance These results have to be interpreted with some
after adjustment for confounding factors. The statistical caution. Because the study was restricted to first-level
significance of the associations was similar to that in the health facilities, very few children had serious illnesses
unadjusted analysis (data not shown). (the proportion of children needing referral ranged
We analysed data for Uganda and Bangladesh both from <1% in Brazil to 13% in Uganda). Doctors are likely
taking the clustering into effect and ignoring the to do better than are other health personnel when
clustering, and results turned out to be very similar, managing severe cases, although recent evidence shows
except that variance was greater when clustering was that even children with severe pneumonia can be
considered, as expected (data not shown). All significant managed safely at first-level facilities.23 Conversely,
associations remained significant in Uganda (in health workers with shorter duration of training might
Bangladesh no significant associations were reported be more willing to comply with standard clinical
with either approach). guidelines than might those who have longer duration
of training, who might use different assessment and
Discussion classification strategies and yet manage children
Recent interest in the human resources crisis in the correctly. However, that staff with longer and shorter
health sector in low-income countries4,5 motivated this duration of training performed much the same in terms
reanalysis of data collected between 2000 and 2003 in the of the correct management indicator suggests that
multicountry assessment of IMCI. Previous reports in-service IMCI training could have offset eventual
from this study showed that quality of care—defined as differences between types of health workers in the
compliance with the IMCI guidelines for assessment, management at primary facility level.

914 www.thelancet.com Vol 372 September 13, 2008


In Brazil, nurses did better than did doctors in correct
management, including counselling tasks on how to
administer the drug at home. Doctors might leave this
task to clerks who dispense drugs in the facilities, thus
failing to meet one of the criteria for correct
management. Additionally, the Brazilian nurses also
did better than their physician colleagues did in correct
assessment. The Brazilian health system is much more
sophisticated than that of Bangladesh, Tanzania, or
Uganda—Brazilian nurses receive equal or longer
duration of training than do most health workers who
were classified in the group with longer training in the
other three countries. That we recorded no systematic
advantages of longer training, despite these differences
in the composition of categories that we investi-
gated, strengthens the external validity of our overall
conclusions.
One of the limitations of the study is its observational
design. However, randomly allocating trainees into
becoming nurses or doctors, for example, is challenging
both ethically and practically. The finding that the
performance of health workers with shorter training was
similar to that of those with longer training is in contrast
with this potential selection bias. Furthermore,
characteristics of workers that might affect the clinical
performance—such as tenure and age of worker—could
be important, but data for these variables were not
available.
Our results need to be interpreted within a broader
conceptual framework of factors affecting clinical
performance, including the availability of equipment and
supplies, supportive supervision, workforce motivation,
economic incentives, and a positive personal and
professional environment.3,24 In the sites that were
included in the present analyses, equipment and drug
availability were mostly adequate, but supervision tended
to be irregular.9 However, we did not gather information
about motivation, incentives, or quality of the professional
environment.
Although the ratio of physicians and paramedical
workers varies enormously from country to country,5,25
nurses and other health workers tend to consistently
outnumber doctors.26 Within countries, ratios are
generally highest in urban areas, particularly for
physicians.27 In rural areas, where health-care needs are
greatest, the numbers of health workers, particularly
doctors, tend to be much less than population needs in
most countries (figure 1).28,29 At the same time, the nurse
to physician ratio in rural areas varies between 3 and 19,
indicating a greater availability of nurses in rural areas
than in urban areas, even in countries where less than
half of their physician and nurse workforce reside in
rural areas (figure 2). The differential distribution of
doctors and nurses in figure 2 indicates that in
underserved, rural areas, health workers with longer
preservice training (such as doctors) are less available
than are those with shorter training (such as nurses).
www.thelancet.com Vol 372 September 13, 2008
Articles
Table 5 shows the ratio of physicians and nurses, and
ranking of child deaths in the four study countries.
In some countries, existing regulations preclude health
workers with shorter duration of training from under-
taking tasks that they might appropriately do.25 For
example, in Brazil, pressure from medical societies led
to discontinuation of IMCI training for nurses in 2002.
In view of the health workforce crisis affecting
underserved areas, these are regrettable measures that
are further challenged by results from our study.
However, in other countries—eg, Ghana, Malawi,
Tanzania, and Zambia—nurses were trained and allowed
to undertake tasks that were usually restricted to
physicians, thus providing support for task shifting
practices long before the term was accepted.30 These
analyses drew attention to the importance of proper
regulation, supervision, and motivation to allow task
shifting to succeed.30
Achievement of high and effective population
coverage is a key component in the efforts to reach a
measurable effect with child-health interventions.31 As
argued by Friedman32 in the context of caesarean
sections, a substantial increase in coverage with health
workers with shorter training—even if they are less
skilled in operative deliveries than are those with longer
duration of training—can contribute to an important
reduction in maternal mortality. A similar argument
can be made for the reduction of deaths in children
younger than 5 years, in view of the published work
showing that case management of specific infections by
community health workers with short training is safe
and effective.33,34
Our results support the principle of task shifting as an
important component of the strategy for strengthening
health systems in areas that are understaffed. Training
and deployment of motivated health workers with shorter
duration of preservice training in regions with high
mortality in children younger than 5 years will help
ensure that all children who need clinical care are seen
by a qualified provider.
Contributors
LH, RWS, and CGV participated in the article conception, writing, and
discussion. LH analysed the data. AMN took part in the writing and
discussion of the article. All authors saw and approved the final version
of the report.
The Multi-Country Evaluation of IMCI Study Group
Members who contributed to this paper, particularly in constructing
table 1, include: João Amaral (Department of Maternal and Child Health,
Federal University of Ceará, Fortaleza, Brazil), Shams El Arifeen
(International Centre for Diarrhoeal Disease Research, Dhaka,
Bangladesh), Joanna Armstrong Schellenberg (Ifakara Health Research
and Development Centre, Ifakara, Tanzania and Gates Malaria
Partnership, London School of Hygiene and Tropical Medicine, London,
UK), Eleanor Gouws (Epidemiology and Analysis
Division, UNAIDS, Geneva, Switzerland), Theopista John (Department
of Family and Reproductive Health, WHO, Dar es Salam, Tanzania), and
George Pariyo (Department of Health Policy Planning and Management,
Makerere University Institute of Public Health, Kampala, Uganda).
Pedro C Hallal from the Universidade Federal de Pelotas, Pelotas, Brazil
provided assistance with the clustering analyses.
915
 Articles
Conflict of interest statement
Only one of the authors is a nurse (AMN), who should be expected to
promote the interests of her profession. LH, RWS, CGV declare that
they have no conflict of interest.
Acknowledgments
This paper is part of the Multi-Country Evaluation (MCE) of IMCI
Effectiveness, Cost and Impact, which is arranged and coordinated by
the WHO Department of Child and Adolescent Health and
Development. The MCE is financially supported by the Bill & Melinda
Gates Foundation and the US Agency for International Development.
We thank Taghreed Adam (WHO Department of Health Systems
Financing) for her critical and constructive comments on the paper;
Robert Black, David Evans, Jean Pierre Habicht, Patrick Vaughan,
Technical Advisers of the Multi-Country Evaluation of IMCI,
Elizabeth Mason (Director of the WHO Department of Child and
Adolescent Health and Development), and Jean Yan (Coordinator of
Health Professions Networks, Nursing, and Midwifery team, WHO
Human Resources for Health Department) for their continuous
technical support; and Jennifer Bryce, who conceptualised the MCE and
managed it until 2002; Alice Ryan, who continues to assist the MCE, for
her organisational skills; and the numerous colleagues who were
involved in study management and data collection in Bangladesh, Brazil,
Uganda, and Tanzania, without whom we would not have had the rich
data that form the foundation of this paper.
References
1 Black RE, Morris SS, Bryce J. Where and why are 10 million
children dying every year? Lancet 2003; 361: 2226–34.
2 Jones G, Steketee RW, Black RE, Bhutta ZA, Morris SS, The
Bellagio Child Survival Study Group. How many child deaths can
we prevent this year? Lancet 2003; 362: 65–71.
3 Anand S, Bärnighausen T. Human resources and health
outcomes: cross-country econometric study. Lancet 2004;
364: 1603–09.
4 WHO. The world health report 2006: working together for health.
Geneva: World Health Organization, 2006.
5 Chen L, Evans T, Anand S, et al. Human resources for
health: overcoming the crisis. Lancet 2004; 364: 1984–90.
6 Bryce J, Boschi-Pinto C, Shibuya K, Black RE, the WHO Child
Health Epidemiology Reference Group. WHO estimates of the
causes of death in children. Lancet 2005; 365: 1147–52.
7 WHO. Task shifting. Global recommendations and guidelines.
Geneva: World Health Organization, 2008.http://www.who.int/
healthsystems/TTR-TaskShifting.pdf (accessed April 25, 2008).
8 Gove S. Integrated management of childhood illness by outpatient
health workers: technical basis and overview.
Bull World Health Organ 1997; 75 (suppl 1): 7–24.
9 Bryce J, Victora CG, Habicht JP, Black RE, Scherpbier RW;
MCE-IMCI Technical Advisors. Programmatic pathways to child
survival: results of a multi-country evaluation of Integrated
Management of Childhood Illness. Health Policy Plan 2005;
20 (suppl 1): i5–17.
10 Tugwell P, de Savigny D, Hawker G, Robinson V. Applying clinical
epidemiological methods to health equity: the equity effectiveness
loop. BMJ 2006; 332: 358–61.
11 Laurant M, Reeves D, Hermens R, Braspenning J, Grol R,
Sibbald B. Substitution of doctors by nurses in primary care.
Cochrane Database Syst Rev 2005; 18: CD001271.
12 Amaral J, Gouws E, Bryce J, Leite AJ, Cunha AL, Victora CG. Effect
of Integrated Management of Childhood Illness (IMCI) on health
worker performance in Northeast-Brazil. Cad Saude Publica 2004;
20: S209–19.
13 Mullan F, Frehywot S. Non-physician clinicians in 47 sub-Saharan
African countries. Lancet 2007; 370: 2158–63.
916
14 Gouws E, Bryce J, Pariyo G, Armstrong Schellenberg J, Amaral J,
Habicht JP. Measuring the quality of child health care at first-level
facilities. Soc Sci Med 2005; 61: 613–25.
15 Bryce J, Gouws E, Adam T, et al. Improving quality and efficiency of
facility-based child health care through Integrated Management of
Childhood Illness in Tanzania. Health Policy Plan 2005;
20 (suppl S1): i69–76.
16 Gupta N, Diallo K, Zurn P, Dal Poz MR. Assessing human
resources for health: what can be learned from labour force
surveys? Hum Resour Health 2003; 1: 5.
17 Diallo K, Zurn P, Gupta N, Dal Poz M. Monitoring and evaluation
of human resources for health: an international perspective.
Hum Resour Health 2003; 1: 3.
18 Armstrong Schellenberg J, Bryce J, de Savigny D, et al; Tanzania
IMCI Multi-Country Evaluation Health Facility Survey Study Group.
The effect of Integrated Management of Childhood Illness on
observed quality of care of under-fives in rural Tanzania.
Health Policy Plan 2004; 19: 1–10.
19 El Arifeen S, Blum LS, Hoque DME, et al. Integrated Management
of Childhood Illness (IMCI) in Bangladesh: early findings from
a cluster-randomised study. Lancet 2004; 364: 1595–602.
20 Nelder JA, Wedderburn RWM. Generalized linear models.
JR Statist Soc A 1972; 135: 370–84.
21 Khoury SA, Mawajdeh S. Performance of health providers in
primary health care services in Jordan. East Mediterr Health J 2004;
10: 372–81.
22 Tulloch J. Integrated approach to child health in developing
countries. Lancet 1999; 354: SII16–20.
23 Chowdhury EK, El Arifeen S, Rahman M, et al. Care at first-level
facilities for children with severe pneumonia in Bangladesh:
a cohort study. Lancet 2008; published online Aug 19. DOI:10.1016/
S0140-6736(08)61166-6.
24 Dal Poz MR, Quain EE, O’Neil M, McCaffery J, Elzinga G,
Martineau T. Addressing the health workforce crisis: towards
a common approach. Human Resour Health 2006; 4: 21.
25 Hongoro C, McPake B. How to bridge the gap in human resources
for health. Lancet 2004; 364: 1451–56.
26 WHO. World health statistics, 2006. Geneva: World Health
Organization, 2006.
27 Dussault G, Franceschini MC. Not enough there, too many
here: understanding geographical imbalances in the distribution of
the health workforce. Hum Resour Health 2006; 4: 12.
28 WHO. Global health atlas. Geneva: World Health Organization,
2005. http://globalatlas.who.int/globalatlas/DataQuery/default.asp
(accessed Jan 30, 2007).
29 Population Division of the Department of Economic and Social
Affairs of the UN Secretariat, World Urbanization Prospects: the
2005 revision population database. http://esa.un.org/unup
(accessed Dec 8, 2007).
30 Dovlo D. Using mid-level cadres as substitutes for internationally
mobile health professionals in Africa. A desk review.
Hum Resour Health 2004; 2: 7.
31 Victora CG, Hanson K, Bryce J, Vaughan JP. Achieving universal
coverage with health interventions. Lancet 2004; 364: 1541–48.
32 Freedman LP. Shifting visions: “delegation” policies and a building
of a “rights based approach” to maternal mortality.
J Am Med Womens Assoc 2002; 57: 154–58.
33 Sazawal S, Black RE; Pneumonia Case Management Trials Group.
Effect of pneumonia case management on mortality in neonates,
infants, and preschool children: a meta-analysis of
community-based trials. Lancet Infect Dis 2003; 3: 547–56.
34 UNICEF. UNICEF joint health and nutrition strategy for
2006–2015. (http://unicef.org/about/execboard/files/06-8_health_
and_nutrition_final_ODS.pdf (accessed Dec 9, 2007).
www.thelancet.com Vol 372 September 13, 2008