Am. J. Trop. Med. Hyg., 87(1), 2012, pp. 57–63
Copyright © 2012 by The American Society of Tropical Medicine and Hygiene
Community Health Workers Use Malaria Rapid Diagnostic Tests (RDTs) Safely and Accurately:
Results of a Longitudinal Study in Zambia
Helen Counihan, Steven A. Harvey,* Masela Sekeseke-Chinyama, Busiku Hamainza, Rose Banda, Thindo Malambo,
Freddie Masaninga, and David Bell
Malaria Consortium, Maputo, Mozambique; Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland; Malaria Consortium,
Lusaka, Zambia; Zambia National Malaria Control Center, Lusaka, Zambia; Livingstone District Health Management Team, Livingstone,
Zambia; World Health Organization, Lusaka, Zambia; Foundation for Innovative New Diagnostics (FIND), Geneva, Switzerland
low-resource populations. However, reliance upon community health workers (CHWs) remains controversial because
of concerns about blood safety and appropriate use of artemisinin combination therapy. This study assessed CHW
ability to use RDTs safely and accurately up to 12 months post-training. We trained 65 Zambian CHWs, and then
provided RDTs, job-aids, and other necessary supplies for village use. Observers assessed CHW performance at 3, 6,
and 12 months post-training. Critical steps performed correctly increased from 87.5% at 3 months to 100% subse-
quently. However, a few CHWs incorrectly read faint positive or invalid results as negative. Although most indicators
improved or remained stable over time, interpretation of faint positives fell to 76.7% correct at 12 months. We
conclude that appropriately trained and supervised CHWs can use RDTs safely and accurately in community practice
for up to 12 months post-training.
Malaria rapid diagnostic tests (RDTs) could radically improve febrile illness management in remote and
Recent advances in point-of-care (POC) diagnostic tech-
nology offer tremendous potential for community-based
infectious disease management. POC diagnostics could signif-
icantly increase the quality of basic health services for remote
populations in low-resource settings. Currently, many such
services are provided by community health workers (CHWs)
with limited training and supervision.1,2Realizing the poten-
tial of POC diagnostics at the community level will depend
upon demonstrating that CHWs can prepare and interpret
such tests accurately and safely.
POC diagnosis could be particularly useful for managing
malaria. The cost of artemisinin-based combination therapy
(ACT), growing anti-malarial resistance, and declining preva-
lence have all heightened recognition of the need for parasite-
based diagnosis.3The World Health Organization (WHO)
now recommends parasite-based diagnosis for patients of all
ages in all transmission zones.4However, many suspected
malaria cases occur far from microscopy-capable health facil-
Malaria rapid diagnostic tests (RDTs) make parasite-based
diagnosis possible where microscopy is unavailable.7,8RDTs
are becoming a key component of community case manage-
ment of malaria (CCMm), also known as home management
of malaria (HMM).9,10Access to accurate diagnosis and
ACTs close to home should reduce malaria-related deaths,
especially among young children. Parasite-based malaria
diagnosis at the community level could also serve as the first
step toward better management of non-malarial causes of
febrile illness, some equally life-threatening.11,12
However, parasite-based diagnosis at community level pre-
sents challenges, particularly in Africa where almost 90% of
the world’s malaria fatalities occur.3,13RDTs require that
health workers obtain a finger-prick blood sample, follow a
simple but strict preparation procedure, and accurately inter-
pret results. Concerns about blood-borne disease and CHW
competence have made many African health systems reluc-
tant to permit RDT use by CHWs. Previous studies show
CHWs can use RDTs accurately and safely after a brief train-
ing period with support from a well-designed set of pictorial
instructions (a “job-aid”).14,15However, these studies assessed
CHW performance in a health facility immediately after
training. Until now, there has been little evidence about
CHW capability to maintain adequate performance over time
at the village level.
This work reports results from a 12-month longitudinal
study of RDT use by Zambian CHWs. The objective was to
track participant performance over time, assess whether safety
and effectiveness remained adequate up to 1 year post-training,
and thus determine the advisability of incorporating RDTs as
part of HMM.
Study design. The study ran from November 2007 to Decem-
ber 2008. To assess performance over time, the study team
recruited 66 CHWs from Livingstone District, Southern Prov-
ince, Zambia. Each CHW received a job-aid and a half-day
training on RDT use and interpretation based upon a previ-
ously validated curriculum.14,16The CHWs who successfully
completed the training returned home with a copy of the job-
aid, an estimated 3-month supply of RDTs (ICT Malaria Pf
Cassette Test (ML01); ICT Diagnostics, Cape Town, South
Africa) and ACT (Coartem) plus gloves, cotton wool, a sharps
disposal box, and a plastic bucket with lid for non-sharps dis-
posal. All participating CHWs received instructions on what to
do in case of a finger prick or blood contamination accident,
prophylaxis. The district level Ministry of Health provided
CHWs with paracetamol for treating fevers in RDT-negative
patients. The CHWs received home visits within 2 weeks of
the training to ensure they were following safety precautions
and not endangering themselves, their patients, or other com-
munity members. Data collected during this initial visit were
*Address correspondence to Steven A. Harvey, Johns Hopkins
Bloomberg School of Public Health, Department of International
Health, 615 N. Wolfe Street, E5030, Baltimore, MD 21205. E-mail:
not included in the final study analysis. Thereafter, a trained
observer visited each CHW at 3, 6, and 12 months post-training.
Observers used a standardized checklist to assess each CHW’s
performance in preparing the test. Whenever possible, CHWs
were observed testing an actual febrile patient. However,
because of the distances between some villages and the need
to complete each wave of observation in a timely manner,
there were some cases in which no febrile patient was avail-
able for testing when the observer arrived. In such cases,
observers assessed CHW performance using a non-febrile vol-
unteer. Observers also used a photograph of 10 RDTs with
positive, negative, and invalid results to assess CHW ability to
read RDTs accurately. To avoid learning effect, the order of
the photographs was varied at each observation.
The study was conducted in collaboration with the Zambia
National Malaria Control Center (NMCC) and as part of
Zambia’s first HMM pilot. Participating CHWs worked within
the catchment area of eight rural health facilities. Because the
study was designed to assess CHW performance in as close as
possible to real-life conditions, nearby health clinics assumed
responsibility for resupply. The clinic and district health man-
agement team (DHMT) staff was encouraged to continue rou-
tine supervision. At the 3-month follow-up, each CHW
received a poster-sized version of the job aid and a photo-
graphic guide explaining how to interpret RDT results.
Observation. Observers received 2 days of training in obser-
vation techniques including instruction on how to minimize
observer-induced reactivity bias. Throughout most of the 3-
and 6-month observations, study co-investigators accompanied
the observers. The checklist divided test preparation into
19 steps. Observers noted whether the CHW performed each
step correctly, incorrectly, or not at all. Each observer also
noted whether he or she missed observing a step. Steps were
divided into two categories: those critical to safe and accurate
performance and those that would not jeopardize patient,
CHW,or communitysafety, if performed incorrectly or missed.
Table 1 lists each step with critical steps in bold type. We
defined safe use to consist of correct completion of steps 9 and
11 (use a sterile lancet and dispose of the lancet in an approved
sharps container immediately after use). We defined accurate
use as correct completion of steps 12, 13, 15, 16, and 17; all
those necessary to arrive at a correct diagnosis. Four situations
were deemed sufficiently serious to require immediate
observer intervention and notification of study supervisors:
attempting to use a lancet or pipette on more than one person,
failure to adequately dispose of a lancet, incorrectly reading an
actual test result, and failure to treat a positive result. The
CHW was then asked to read the results of 10 RDT photo-
graphs presented on a card. At each observation, the observer
recorded number of RDTs performed since the previous visit,
number of positives, and number of patients < 5 years of age.
Finally, the observer asked CHWs about any challenges or
concerns they were experiencing.
Data analysis. Data were double entered into a Microsoft
Access database (Microsoft, Redman, WA), checked for dis-
cordance, and then analyzed using Stata (College Station,
TX). Percentages of total steps, crucial steps, and non-crucial
steps performed correctly were calculated for each observa-
tion cycle. Generalized estimating equations (GEE) logistic
regression models for panel data were fit to identify factors
associated with correct performance of at least 90% of total
test steps.17The GEE models were also fit to identify factors
associated with correct performance of 100% of critical steps
and at least 90% of non-critical steps. For the reading of the
10 photographic results, performance was assessed for each
observation period by calculating the pooled mean and
median percentage of total tests read correctly and the per-
centage of CHWs who read each individual test correctly.
Correctly interpreting photos of invalid RDT results was a
particular concern: some readers incorrectly interpret the
appearance of a test line but no control line as negative. For
this reason, Stata’s pretest command was used to estimate
equality of proportions between invalid results showing no
line versus invalid results showing only a test and no control
line.18Similarly, some users misread RDTs with a faint posi-
tive line as negative. The t test was used to determine whether
there was a significant difference between correct interpreta-
tion of strong and faint positive test lines.
RDT quality assurance. Before the study, the RDT batches
were lot tested according to WHO protocol at the Research
Institute for Tropical Medicine (RITM)in the Philippines.19At
the end of the study, during the last round of observations, two
RDTs were collected from each of 42 CHWs (all those who
had sufficient RDTs in stock) for further quality control testing
at RITM. Results showed 100% detection at 200 parasites/mL,
indicating no measurable sensitivity loss despite prolonged
storage in homes of participating CHWs.
Ethical approval. This study received ethical approval
from WHO/TDR (RPC 167) and the Tropical Disease
Research Center Ethics Committee, Ndola, Zambia. By the
start of Phase III, Zambia had decided to include RDTs for
confirmed diagnosis of malaria as part of HMM. Thus
community-based RDT use became a routine rather than an
Participant demographics. Sixty-six CHWs participated in
the training in November 2007 using the previously developed
The 19 steps required to correctly and safely prepare a rapid
diagnostic test (RDT) (critical steps noted in boldface type)
1. Assemble new test packet, swab, buffer, pipette, lancet and gloves.
2. Put on new pair of gloves.
3. Check expiry date on package.
4. Check desiccant sachet is still dry (do not include answer in
5. Write patient’s name on cassette.
6. Place cassette on a level surface.
7. Clean finger with antiseptic/alcohol.
8. Allow finger to dry before pricking it.
9. Use a sterile lancet for finger prick.
10. Puncture the side of the ball of the finger.
11. Dispose of lancet in sharps bin immediately after pricking finger.
12. Collect blood with the enclosed pipette making sure to fill close to
the first cross line.
13. Using the pipette, blot blood onto the pad in the smaller well.
14. Dispose of pipette in sharps container immediately.
15. Dispense 5 drops of clearing buffer into the larger well.
16. Wait 15 minutes before reading negative results.*
17. Read test results correctly.
18. Record results in CHW register.
19. Dispose of non-sharps (gloves, wrappers, alcohol swab, and
*Positive results may be read before 15 minutes if control line has also appeared. Results
should not be read after 30 minutes.
COUNIHAN AND OTHERS
manual. One CHW was dropped from the study after failing to
demonstrate sufficient competency during practice sessions.
Thus, 65 were initially enrolled in the 12-month surveillance.
At the 3-month observation period, two of these 65 were
unavailable because of work or travel. At the 6-month obser-
vation one of these two was available, but the second had
dropped out completely and three others were unavailable,
giving a total of 61 CHWs observed. At 12 months, the CHW
who returned at the 6-month observation was missing again,
and four others were unavailable due to illness or bereave-
ment, therefore the final observation included 59 CHWs.
Table 2 presents participant socio-demographic characteristics.
Test preparation. As shown in Figure 1, median CHW per-
formance remained steady or improved over time for critical
steps, non-critical steps, and RDT interpretation. The median
percentage of critical RDT steps performed correctly rose
from 87.5% at 3 months to 100% at 6 and 12 months. From
the perspective of changes in individual score, performance
on critical steps improved over time: 40.3% of CHWs (25 of
62) performed all critical steps correctly at 3 months com-
pared with 61.7% (37 of 60) at 6 months, and 79.7% (47 of
59) at 12 months. If we broaden this group to include CHWs
who made only one error, it includes 79% of participants
(49 of 62) at 3 months compared with 90% (54 of 60) at
6 months and 97% (57 of 59) at 12 months. Table 3 provides
step-by-step performance of both critical and non-critical
steps at 3, 6, and 12 months.
On the basis of GEE logistic regression, CHW age and time
since training (3, 6, or 12 months) were the only factors signif-
icantly associated with correct performance of all critical steps
(Table 4). At 6 months post-training, CHWs were 2.4 times
more likely to perform all critical steps correctly compared
with the 3-month assessment (P = 0.02). At 12 months, they
were 6.4 times more likely to perform all critical steps correctly
compared with the 3-month assessment (P < 0.001). Compared
Socio-demographic characteristics of participating CHWs
Characteristicn (% or range)
Mean age (years)
Prior malaria treatment experience
Median months malaria treatment experience*
Prior malaria RDT experience
*A few community health workers (CHWs) with extensive malaria treatment experience
result in a mean (31.3 months) that is not representative of the group as a whole. Thus,
median is presented here instead of mean.
Figure 1.Percent of rapid diagnostic tests (RDT) steps completed or interpreted correctly at 3, 6, and 12 months post-training.
SAFE ACCURATE MALARIA RDT USE BY ZAMBIAN CHWS
with those 40 and under, CHWs 50 years of age and older were
62% less likely to score 100% on critical steps (P = 0.04).
Safety. In general, CHWs applied appropriate blood safety
measures. All CHWs wore gloves, although at 3- and 6-month
observations, two used a single pair of gloves on more than
one patient, in one case due to insufficient supply. Lancet use
was generally satisfactory. Once during the 3-month observa-
tion, an observer intervened when one CHW appeared to be
about to re-use a lancet on a second patient. In all other cases,
the only errors were setting the unused lancet down on its
wrapper before use or setting the used lancet down on the
table before disposing of it in the sharps box. At the 6-month
observation, one CHW disposed of her lancet with her regular
waste because she had no sharps box. The CHWs were
instructed during training to report needle-stick injuries
immediately to receive post-exposure prophylaxis for HIV,
but no such injuries were observed or reported during the
Errors in RDT use. Median scores on most study measures
improved over the surveillance period. One exception, as
measured by the different test result photographs, was the
ability to correctly interpret faint positive test lines as posi-
tive. This improved from 89.7% at 3 months to 96.7% at
6 months, and then declined to 76.7% at 12 months.
Slightly less than half of the participants collected the cor-
rect amount of blood during the 3-month observation; this
rose to slightly over 60% during the 6- and 12-month obser-
vations. In most cases, participants collected too little blood
rather than too much. As in other studies, CHWs found the
blood collection device included with the RDT—in this case a
disposable plastic pipette—difficult to use. Early on, CHWs
also had difficulty transferring the blood to the test cassette:
at the 3-month observation, eight of the 64 participants (13%)
transferred most of the blood to the well wall rather than the
absorbent pad. However, this error occurred in only three out
of 60 of cases (5%) at the 6-month observation. At 12 months
all participants transferred the blood correctly. Finally, three
CHWs read test results incorrectly at 3 months, compared
with only one at 6 and 12 months—with all but one being
cases of reading faint positives as negative.
Job-aid use. Documenting job-aid use by CHWs was diffi-
cult. In many observations, the job-aid was present (e.g.,
hanging on a wall or tree), but the observer could not reliably
determine whether the CHW referred to it while preparing or
interpreting the test. Many CHWs said they avoided using the
job-aid because they believed the observer wanted to test
their ability to use the RDT without support. The notable
exceptions were CHWs who used the job-aid to explain test
procedures and possible results to patients before beginning
Accuracy at interpreting RDT results. Using the photo-
graph of 10 different RDT results, CHWs correctly identified
96.5% of positive tests at 3 months and 98.3% at 6 months,
but only 90.5% at 12 months. Similarly, CHWs correctly
identified 94.3% of negative results at 3 months, 97.9% at
6 months, and 94.7% at 12 months. Interpretation of invalid
test results improved from 90.2% correct at 3 months to
96.7% at 6 months and 96.5% at 12 months.
Each photograph of 10 tests included two invalid results,
one illustrated by no line (neither test nor control) and one
illustrated by a test line only (no control line). As shown in
Table 5, a significantly higher proportion of participants at
3 months correctly interpreted the RDT with no line than the
RDT with a test line only. This difference remained signifi-
cant, though smaller, at 6 months, but was no longer significant
Step-by-step performance: percent of CHWs who performed each step correctly at each observation
3 months (n = 63)6 months (n = 61) 12 months (n = 59) Most common reasons for errors
Clean finger before pricking
Use sterile lancet
Dispose of lancet safely
Blot blood correctly
Dispense buffer drops accurately
Wait correct amount of time
Read RDT results correctly
Assemble everything before starting
Check expiry date
Write patient’s name on cassette
Place RDT on a level surface
Allow finger to dry before pricking
Prick side of finger
Collect blood correctly
Dispose of pipette correctly
Record result in register
Dispose of non-sharps correctly
Did not change gloves between patients
Tip of lancet touching gloves or work surface before pricking
Set used lancet on table before disposing of it in sharps box
Most blood went on wall of well
Too many or too few drops
Faint positive called negative
Pricked middle of finger instead
Too little blood in most cases
Usually because the subject was not a real patient
*CHWs = community health workers; RDT = rapid diagnostic test.
Adjusted odds ratios for a score of 100% on critical steps (GEE
logistic regression model)*
Adjusted odds ratio (95% CI)P value
40 years and under
50 years and above
*GEE = generalized estimating equations; CI = confidence interval.
COUNIHAN AND OTHERS
at 12 months. The CHWs consistently read photographs of
strong positive test results correctly. Faint positives were
more problematic, most notably at the 12-month observation.
Supplies and supervision. Many participating CHWs had
difficulty obtaining additional RDTs and supplies from associ-
ated health facilities after exhausting initial stocks. In some
cases, this occurred because the health facilities themselves
experienced stock-outs. In other cases, facility managers may
have been resistant to CHW use of RDTs as they were not
informed by their supervisor who attended the introductory
workshop. Occasionally, there were discrepancies in reporting
of stock sufficiency between supervisors and CHWs within their
area. Consequently, some CHWs experienced stock-outs of
RDTs, drugs, and supplies until the study team resupplied
them. To the best of our knowledge, no supervisory visits by
health facility or DHMT staff occurred during the study period.
The study demonstrated consistently high performance over
12 months by CHWs using malaria RDTs after receiving a
half-day competency-based training and a field-tested job-aid.
The study also revealed a few instances of poor practice, with
some participants repeating the same or similar errors through-
out the 12-month period. In particular, difficulty reading faint
positive test lines could lead to false-negative results. Previous
studies have noted both this difficulty and the difficulty manip-
ulating blood transfer devices.14,20–23
The study did not attempt to assess the frequency with which
supervision visits or retraining should occur, but a supervisor
should visit to address any high-risk behaviors soon after train-
ing a new cadre of CHWs. Regular visits thereafter seem
warranted, particularly for CHWs demonstrating inadequate
performance. Resupply provides an opportunity for skill
reinforcement—in the CHW’s normal environment if a super-
visor brings supplies to the village, or at a health facility if the
CHW visits regularly to collect supplies and submit data. Other
studies on RDT use have also noted the need for ongoing
training and supervision.23
The drop in correct interpretation of faint test lines at
12 months suggests that supervisory visits should stress this,
and should occur at least semi-annually. Training programs
should reinforce the distinction between test and control
lines, ensure that CHWs make that distinction accurately,
and emphasize that even faint lines indicate a positive result.
CHWs using the same pair of gloves on more than one
patient were providing direct support to a health facility
where this was considered normal practice if the gloves had
no blood on them. More extensive training or supervision of
health facility staff, and better supply of such commodities,
appear to be needed in that case.
Improvement in the design of blood collection devices might
address some of the problems of RDT use encountered in this
study, in particular the reliability of volume transferred.20–22In
addition, initial training and supervision should emphasize the
importance of collecting the correct volume of blood. This
should include ensuring that trainees practice with whatever
blood collection device will be packaged with their RDT(s)
until they are able to collect the required volume of blood
correctly every time. Retractable lancets would also reduce
the significance of the (rare) lancet mishandling observed.
Although CHWs may prepare and interpret RDTs safely
and accurately, acting on the result is fundamental to good
outcomes. This study did not specifically assess CHW adher-
ence to RDT results, but a retrospective assessment of drug
consumption indicates strong compliance with prescription of
ACTs (Chinyama et al., unpublished data). During follow-up
visits, prescription of anti-malarials to RDT negative cases was
observed by two CHWs at 3 months, but was no longer an
issue during later visits. It is worth noting that the prevalence
of malaria remained low throughout the study period, which
resulted in some CHWs expressing concern that the RDTs
may not work correctly. Despite this, they continued to use
RDTs and only treat positive cases, as trained.
The observed errors led to minor modifications to the train-
ing manual, with increased emphasis on aspects of blood safety
and RDT reading. Generic versions of the resulting materials
are available on the WHO malaria RDT website.16A library
of product-specific versions is available from FIND.23
Limitations. Some limitations may have influenced study
results. First, Livingstone district is economically better off
than many areas of Zambia. Livingstone was chosen as a
study site because it was the first district to implement
HMM. Consistent with Livingstone’s relatively higher socio-
economic status, 75% of participating CHWs had completed
some secondary education. This could have produced better
results than might occur with less educated CHWs.
The repeated visits by observers may also have influenced
results. Initially, some CHWs were uncomfortable with pre-
paring the RDT in front of an observer, which may have
affected the quality of their performance. Over time CHWs
became more comfortable with the visit procedure. To mea-
sure performance as accurately as possible, observers were
instructed not to interfere with or correct a CHW unless that
CHW was about to do something that presented an immedi-
ate danger to him- or herself, the patient, or the community.
At times the observers tended to err on the side of inter-
vention in ambiguous situations, avoiding harm to partici-
pants rather than preserving data integrity. Thus, although
CHWs did not receive any retraining during the 12-month
surveillance, they did receive some correction from observers
from time to time.
Correct interpretation of rapid diagnostic test (RDT) results (percent read correctly in each category by time post-training)
Ambiguous tests: no line vs. test line onlyPositive test results: strong vs. faint positive line
Time post-training No lineTest line onlyDifferenceP value*
Strong positive Faint positiveDifferenceP value*
*t test for equality of proportions.
Over the 12-month surveillance period, there were four consistently poor performers. Two of these incorrectly interpreted 4 tests out of 30 total (13.3%) incorrectly, two incorrectly interpreted
5 out of 30 (16.7%) incorrectly, and one incorrectly interpreted 7 (23.3%) incorrectly.
SAFE ACCURATE MALARIA RDT USE BY ZAMBIAN CHWS
Use of volunteers rather than febrile patients for some
observations could have biased results. This was sometimes
necessary because each observation cycle had to be com-
pleted within a short period, and febrile patients were not
always available. However, the testing procedure is the same
whether performed on a patient or a volunteer. Furthermore,
because many CHWs lived in remote areas, the study team
notified them one day before each observer visit to ensure
they would be home for the observation. Having observers
arrive unannounced would have been preferable from a
methodological standpoint, but would have greatly reduced
the probability of finding the CHW at home and available.
With a well-designed job-aid and half-day training, CHWs
can diagnose malaria safely and accurately using RDTs. CHWs
participating in the study retained these skills in community
practice up to 12 months post-training. However, occasional
errors occur, therefore supervisors should conduct periodic
performance appraisals and make necessary corrections to
ensure patient, CHW, and community safety. The results offer
good evidence for the appropriateness of CHWs taking finger-
prick blood samples and using POC diagnostic tests in the
community, provided they receive adequate training, job-aids,
supplies, and follow-up supervision.
Acknowledgments: We acknowledge the close support and dedica-
tion of the volunteer health workers and clinic staff of Livingstone
District, Zambia, and the Livingstone District Health Management
Team. We appreciate Mary Shepherd’s able assistance performing
the statistical analysis.
Financial support: The project was financed jointly by the Foundation
for Innovative New Diagnostics (FIND) through funding from the
Bill and Melinda Gates Foundation and the UK Department for
International Development (DFID), and by the Accessible Quality
Assured Diagnostics Programme of the Special Programme for
Research on Tropical Diseases (TDR).
Disclaimer: The authors declare that they have no conflict of interest
in relation to this study.
Rua Joseph Ki-Zerbo, Maputo, Mozambique, E-mail: h.counihan@
malariaconsortium.org. Steven A. Harvey, Johns Hopkins Bloomberg
MD, E-mail: email@example.com. Masela Sekeseke-Chinyama, formerly
with Malaria Consortium Lusaka, Zambia, E-mail: msekeseke@gmail
.com. Busiku Hamainza, Zambia National Malaria Control Center,
Chainama Hospital College Grounds, Lusaka, Zambia, E-mail:
firstname.lastname@example.org. Rose Banda and Thindo Malambo, Living-
stone District Health Management Team, Livingstone, Zambia,
E-mails: email@example.com and firstname.lastname@example.org.
Freddie Masaninga, World Health Organization, UN Annex, Lusaka,
Zambia, E-mail: email@example.com. David Bell, Foundation
for Innovative New Diagnostics (FIND), Geneva, Switzerland, E-mail:
1. Ajayi IO, Browne EN, Bateganya F, Yar D, Happi C, Falade CO,
Gbotosho GO, Yusuf B, Boateng S, Mugittu K, Cousens S,
Nanyunja M, Pagnoni F, 2008. Effectiveness of artemisinin-
based combination therapy usedin the context of home manage-
ment of malaria: a report from three study sites in sub-Saharan
Africa. Malar J 7: 190.
2. Pagnoni F, 2009. Malaria treatment: no place like home. Trends
Parasitol 25: 115–119.
3. World Health Organization, 2010. World Malaria Report 2010.
Geneva: WHO Global Malaria Programme.
4. World Health Organization, 2010. Guidelines for the Treatment of
Malaria. Second edition. Geneva: World Health Organization.
5. Bell D, Wongsrichanalai C, Barnwell JW, 2006. Ensuring quality
and access for malaria diagnosis: how can it be achieved? Nat
Rev Microbiol 4: S7–S20.
6. Murray CK, Bell D, Gasser RA, Wongsrichanalai C, 2003.
Rapid diagnostic testing for malaria. Trop Med Int Health 8:
7. de Oliveira AM, Skarbinski J, Ouma PO, Kariuki S, Barnwell
JW, Otieno K, Onyona P, Causer LM, Laserson KF, Akhwale
WS, Slutsker L, Hamel M, 2009. Performance of malaria rapid
diagnostic tests as part of routine malaria case management in
Kenya. Am J Trop Med Hyg 80: 470–474.
8. Kyabayinze DJ, Asiimwe C, Nakanjako D, Nabakooza J,
Counihan H, Tibenderana JK, 2010. Use of RDTs to improve
malaria diagnosis and fever case management at primary
health care facilities in Uganda. Malar J 9: 200.
9. Phommanivong V, Thongkham K, Deyer G, Rene JP, Barennes H,
2010. An assessment of early diagnosis and treatment of malaria
by village health volunteers in the Lao PDR. Malar J 9: 347.
10. Ajayi IO, Browne EN, Bateganya F, Yar D, Happi C, Falade
CO, Gbotosho GO, Yusuf B, Boateng S, Mugittu K, Cousens
S, Nanyunja M, Pagnoni F, 2008. Effectiveness of artemisinin-
based combination therapy used in the context of home
management of malaria: a report from three study sites in
sub-Saharan Africa. Malar J 7: 190.
11. Yeboah-Antwi K, Pilingana P, Macleod WB, Semrau K, Siazeele
K, Kalesha P, Hamainza B, Seidenberg P, Mazimba A, Sabin
L, Kamholz K, Thea DM, Hamer DH, 2010. Community case
management of fever due to malaria and pneumonia in chil-
dren under five in Zambia: a cluster randomized controlled
trial. PLoS Med 7: e1000340.
12. Black RE, Cousens S, Johnson HL, Lawn JE, Rudan I, Bassani
DG, Jha P, Campbell H, Walker CF, Cibulskis R, Eisele T, Liu
L, Mathers C, 2010. Global, regional, and national causes of
child mortality in 2008: a systematic analysis. Lancet 375: 9730.
13. Bisoffi Z, Sirima B, Angheben A, Lodesani C, Gobbi F, Tinto H,
Van de Ende J, 2009. Rapid malaria diagnostic tests versus
clinical management of malaria in rural Burkina Faso: safety
and effect on clinical decisions. A randomized trial. Trop Med
Int Health 14: 491–498.
14. Harvey SA, Jennings L, Chinyama M, Masaninga F, Mulholland
K, Bell DR, 2008. Improving community health worker use of
malaria rapid diagnostic tests in Zambia: package instructions,
job aid and job aid-plus-training. Malar J 7: 160.
15. Rennie W, Phetsouvanh R, Lupisan S, Vanisaveth V,
Hongvanthong B, PhompidaS, Alday P, Fulache M, Lumagui R,
Jorgensen P, Bell D, Harvey S, 2007. Minimizing human error in
malaria rapid diagnosis: clarity of written instructions and health
worker performance. TransRSoc Trop MedHyg101: 9–18.
16. World Health Organization, 2011. RDT Instructions and Train-
ing. Available at: http://www.wpro.who.int/sites/rdt/using_rdts/
training/. Accessed March 2011.
17. StataCorp., 2007. Stata Longitudinal/Panel-Data Reference Man-
ual: Release 10. College Station, TX: StataCorp LP.
18. StataCorp., 2007. Stata Statistical Software: Release 10. College
Station, TX: StataCorp LP.
19. WHO, 2008. Methods Manual for Laboratory Quality Control Test-
ing of Malaria Rapid Diagnostic Tests. Web Version 5a. Ermita,
Manila: WHO - Regional Office for the Western Pacific.
Available at: http://www2.wpro.who.int/NR/rdonlyres/A31142C7-
MethManuV5_a_WEBVERSIONAug2008.pdf. Downloaded 5
20. Luchavez J, Lintag ME, Coll-Black M, Baik F, Bell D, 2007. An
assessment of various blood collection and transfer methods
used for malaria rapid diagnostic tests. Malar J 6: 149.
21. Hopkins H, Oyibo W, Luchavez J, Mationg ML, Asiimwe C,
Albertini A, Gonzalez IJ, Gatton ML, Bell D, 2011. Blood
COUNIHAN AND OTHERS
transfer devices for malaria rapid diagnostic tests: evaluation Download full-text
of accuracy, safety and ease of use. Malar J 10: 30.
22. Seidahmed OM, Mohamedein MM, Elsir AA, Ali FT, Malik el
FM, Ahmed ES, 2008. End-user errors in applying two malaria
rapid diagnostic tests in a remote area of Sudan. Trop Med Int
Health 13: 406–409.
23. Montoya PJ, Lukehart SA, Brentlinger PE, Blanco AJ,
Floriano F, Sairosse J, Gloyd S, 2006. Comparison of the
diagnostic accuracy of a rapid immunochromatographic
test and the rapid plasma reagin test for antenatal syphi-
lis screening in Mozambique. Bull World Health Organ
SAFE ACCURATE MALARIA RDT USE BY ZAMBIAN CHWS