O’Connell et al.

Malaria Journal 2011, 10:326

http://www.malariajournal.com/content/10/1/326

RESEARCH Open Access

Got ACTs? Availability, price, market share and

provider knowledge of anti-malarial medicines in
public and private sector outlets in six malaria-
endemic countries
Kathryn A O’Connell1*, Hellen Gatakaa1, Stephen Poyer1, Julius Njogu1, Illah Evance1, Erik Munroe1,
Tsione Solomon1, Catherine Goodman2, Kara Hanson2, Cyprien Zinsou3, Louis Akulayi4, Jacky Raharinjatovo5,
Ekundayo Arogundade6, Peter Buyungo7, Felton Mpasela8, Chérifatou Bello Adjibabi9, Jean Angbalu Agbango10,
Benjamin Fanomezana Ramarosandratana11, Babajide Coker12, Denis Rubahika13, Busiku Hamainza14,
Steven Chapman15, Tanya Shewchuk1 and Desmond Chavasse1

Abstract
Background: Artemisinin-based combination therapy (ACT) is the first-line malaria treatment throughout most of
the malaria-endemic world. Data on ACT availability, price and market share are needed to provide a firm evidence
base from which to assess the current situation concerning quality-assured ACT supply. This paper presents supply
side data from ACTwatch outlet surveys in Benin, the Democratic Republic of Congo (DRC), Madagascar, Nigeria,
Uganda and Zambia.
Methods: Between March 2009 and June 2010, nationally representative surveys of outlets providing anti-malarials
to consumers were conducted. A census of all outlets with the potential to provide anti-malarials was conducted
in clusters sampled randomly.
Results: 28,263 outlets were censused, 51,158 anti-malarials were audited, and 9,118 providers interviewed. The
proportion of public health facilities with at least one first-line quality-assured ACT in stock ranged between 43%
and 85%. Among private sector outlets stocking at least one anti-malarial, non-artemisinin therapies, such as
chloroquine and sulphadoxine-pyrimethamine, were widely available (> 95% of outlets) as compared to first-line
quality-assured ACT (< 25%). In the public/not-for-profit sector, first-line quality-assured ACT was available for free
in all countries except Benin and the DRC (US$1.29 [Inter Quartile Range (IQR): $1.29-$1.29] and $0.52[IQR: $0.00-
$1.29] per adult equivalent dose respectively). In the private sector, first-line quality-assured ACT was 5-24 times
more expensive than non-artemisinin therapies. The exception was Madagascar where, due to national social
marketing of subsidized ACT, the price of first-line quality-assured ACT ($0.14 [IQR: $0.10, $0.57]) was significantly
lower than the most popular treatment (chloroquine, $0.36 [IQR: $0.36, $0.36]). Quality-assured ACT accounted for
less than 25% of total anti-malarial volumes; private-sector quality-assured ACT volumes represented less than 6%
of the total market share. Most anti-malarials were distributed through the private sector, but often comprised non-
artemisinin therapies, and in the DRC and Nigeria, oral artemisinin monotherapies. Provider knowledge of the first-
line treatment was significantly lower in the private sector than in the public/not-for-profit sector.
Conclusions: These standardized, nationally representative results demonstrate the typically low availability, low
market share and high prices of ACT, in the private sector where most anti-malarials are accessed, with some
exceptions. The results confirm that there is substantial room to improve availability and affordability of ACT

* Correspondence: koconnell@psi.org
1
Population Services International, Malaria & Child Survival Department, P.O.
Box 43640, Nairobi, Kenya
Full list of author information is available at the end of the article

© 2011 O’Connell et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative
Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and
reproduction in any medium, provided the original work is properly cited.
O’Connell et al. Malaria Journal 2011, 10:326 Page 2 of 14
http://www.malariajournal.com/content/10/1/326

treatment in the surveyed countries. The data will also be useful for monitoring the impact of interventions such
as the Affordable Medicines Facility for malaria.

Background Standardized information on availability, price and mar-

In 2000, African Heads of State agreed in Abuja that by ket share of ACT and other anti-malarials in the public/
2010 60% of people with malaria would be able to not-for-profit and private sectors will inform such inter-
access “affordable and appropriate treatment within 24 ventions and help gauge their relative success, guiding
hours of the onset of symptoms“ [1]. This target was policy makers and programmers at the national and
increased to 80% in 2005 [2]. Artemisinin-based Combi- international levels.
nation Therapy (ACT) is the first-line malaria treatment To help address some of these gaps, Population Ser-
throughout most of the malaria-endemic world. vices International (PSI) in partnership with the London
Although public sector procurement of ACT has School of Hygiene and Tropical Medicine (LSHTM)
increased sharply in recent years, it was estimated that launched a five-year multi-country research project in
in 2009 less than 15% of children under five years of age 2008 called ACTwatch to monitor anti-malarial supply
received ACT when they had fever in 11 of 13 African and demand in seven malaria endemic countries in
countries, and only 32% received anti-malarials of any Africa and Asia. This involves multiple surveys of
kind [3]. Furthermore, the availability and use of oral households and anti-malarial outlets, and a systematic
artemisinin monotherapy [4,5] is worrying given the risk analysis of the anti-malarial supply chain in Benin, Cam-
of development of artemisinin resistance [6] and WHO’s bodia, the Democratic Republic of Congo (DRC), Mada-
recommendation to countries to ban the importation gascar, Nigeria, Uganda and Zambia [22].
and manufacturing of these drugs. This paper presents the results from ACTwatch outlet
A variety of country-specific studies have raised surveys carried out between March 2009 and June 2010
important concerns about ACT supply that may explain in the six sub-Saharan countries. An ACT was the first-
these low levels of ACT use. Studies have generally line anti-malarial in all these countries at the time of
found low availability of ACT and higher availability of the study, with artemether-lumefantrine and/or artesu-
monotherapies in both the public and private sector nate-amodiaquine the recommended combination. All
[7-9]. While ACT is typically free-of-charge or subsi- countries recommended quinine for treatment of severe
dized in the public and not-for-profit sector, in the pri- malaria, and SP for intermittent preventive treatment in
vate sector ACT is between ten and twenty times more pregnancy.
expensive than non-artemisnin therapies, such as chlor-
oquine and sulphadoxine-pyrimethamine (SP) [8,10-12]. Methods
Provider knowledge that ACT is the first-line treatment Design & sampling
for malaria is also low, particularly in the private sector A nationally representative sample of outlets providing
[13-16]. anti-malarials to consumers was selected. Each country
Research addressing questions on the anti-malarial was first stratified using information on country size,
market is typically limited in geographical scope and uti- malaria endemicity, geopolitical areas, or urbanization.
lizes a variety of different methods to ascertain price and Benin was designated as a single stratum; Madagascar,
availability. Data are typically collected in a few purpo- Uganda, and Zambia were divided into two strata; the
sively selected geographical areas within one country, DRC into four and Nigeria into six [22]. Stratification
and at times include only certain outlet types or sectors. was agreed in consultation with national authorities
Studies are not typically designed with sufficient power including the national malaria control programme and
to detect differences between sectors. With a few excep- the national drug regulatory authority.
tions [14,17-20], there is little evidence on the market Following stratification, a one-stage probability-pro-
share of anti-malarial medicines sold through the private portional-to-size (PPS) cluster design was used to select
retail sector, where most malaria treatment is sought. clusters within each stratum, with cluster population
The available data, therefore, fail to provide a firm evi- serving as the measure of size. The primary sampling
dence base from which to assess the current situation unit, or cluster, selected was an administrative unit with
concerning ACT supply across countries, or to help 10,000 to 15,000 inhabitants. The proportion of outlets
evaluate the impact of important initiatives to improve with any ACT in stock was used as a key indicator for
this, such as the Affordable Medicines Facility for sample size calculation, conservatively estimated at 40%
Malaria (AMFm), which provides heavily subsidized, for each country. A minimum of 290 outlets with anti-
quality assured, ACT to pilot countries [21]. malarials in stock was needed to detect a 20% increase
O’Connell et al. Malaria Journal 2011, 10:326 Page 3 of 14
http://www.malariajournal.com/content/10/1/326

in this indicator, at 80% power, setting the level of sig- and (b) anti-malarials were available on the day of the
nificance at 5% and adjusting for an estimated design survey. All providers answering “yes” to at least one of
effect of 3. It was estimated that the selection of 19 clus- these questions were deemed eligible for inclusion in
ters per stratum would provide at least this number of the survey and the questionnaire was administered to
outlets stocking anti-malarials. these outlets. An interview with the staff member who
Within each selected cluster all outlet types with the was most likely to sell or prescribe medications was
potential to provide anti-malarials to consumers were conducted.
sampled, with the identification of these outlet types The interview was carried out in the appropriate local
based on available literature and local knowledge. The language. Questionnaires were pre-tested and validated
main types of outlets sampled included public and not- in each country before data collection.
for profit health facilities, private health facilities, phar- The questionnaire consisted of two parts. A drug audit
macies, drug stores and grocery stores/kiosks. In addi- collected information on all anti-malarials found at the
tion, a range of other outlet types were considered time of the interview, including information on brand
relevant in specific countries, for example, market stalls and generic names, dose, manufacturer, amount sold in
in Benin, gargotes in Madagascar, itinerant vendors in the last seven days, and retail price. The provider ques-
Nigeria and Benin, and community health workers in tionnaire collected information on provider demo-
Nigeria, Uganda and Madagascar. graphics, anti-malarial treatment knowledge, anti-
As public health facilities and pharmacies are impor- malarial storage conditions, stock-outs, and licensing.
tant providers of anti-malarials but are relatively uncom- Data collection lasted from one to five months in each
mon, over-sampling was conducted for these outlet country and took place between March 2009 and June
types. This booster sample was obtained by including 2010 (Benin, April 28 th - May 13 th 2009; the DRC,
public health facilities and pharmacies in the larger August 10th - October 27th 2009; Madagascar, April 27th
administrative area from which a given cluster was - June 21st 2010; Nigeria August 4th - September 16th
selected. For example, if the cluster was defined as the 2009; Uganda, March 16 th - April 7 th 2009; Zambia,
sub-district, the booster sample covered all public health April 14th - July 3rd 2009).
facilities and pharmacies in the whole district within During data collection, approximately 80% of all ques-
which the sub-district was located. tionnaires were reviewed by the team supervisor and 15-
20% of all outlets were revisited by a supervisor or/and
Training and fieldwork quality controller for quality control checks.
The outlet surveys were conducted during the months
of peak malaria transmission in each country. The sur- Data analysis
vey data were collected by teams of between 21 and 76 Double data entry was conducted using Microsoft
enumerators with at least 12 years of education who Access (Microsoft Corporation, Redmond, Washington,
were fluent in the local language. Interviewer training USA) with built-in range and consistency checks.
was provided over six days, focusing on outlet identifica- Entered data were triangulated with questionnaires, field
tion, informed consent procedures, and the procedures supervision notes and daily activity logs filled by each
for completing the questionnaire. interviewer in the field. Data were analysed using Stata
Field workers were provided with a list of selected 11.0 (StataCorp College Station, TX).
clusters and maps that illustrated their administrative Stata survey settings were used to account for the
boundaries, in addition to lists of public health facilities stratified and clustered sampling strategy. Results were
and pharmacies obtained from relevant authorities. adjusted by sampling weights. Sampling weights were
Snowball sampling was also used by field workers to based on the inverse of the probability of selection to
identify facilities that were not on official lists. In each allow for the differences in strata and cluster sizes and
selected cluster, fieldworkers conducted a census of all the oversampling of public health facilities and pharma-
outlets that had the potential to provide anti-malarials. cies. Distribution of the outlets was assumed to be pro-
For each outlet that was identified during the census, portional to population size. Data analysis included
the outlet type and location were noted, along with its descriptive summaries; between-group proportions were
longitude and latitude coordinates (obtained via hand- compared using chi-square analysis and median prices
held global positioning units). A fieldworker then compared using the Wilcoxon rank sum test.
approached the outlet’s main seller or provider and For the presentation of results, outlets were grouped
invited him or her to participate in the study. Providers into two categories: 1) the public and not-for-profit sec-
who agreed to participate were asked two screening tor, and 2) the private sector. The former included pub-
questions to determine whether (a) the outlet had lic health facilities, facilities supported by NGOs,
stocked anti-malarials within the previous three months, mission hospitals and/or community health workers.
O’Connell et al. Malaria Journal 2011, 10:326 Page 4 of 14
http://www.malariajournal.com/content/10/1/326

The private sector included all for-profit outlets. For the Volumes and price
purpose of analysis, anti-malarials were classified into The volume sold/distributed and retail price of the anti-
ACT, non-artemisinin therapies (such as chloroquine malarials recorded in the drug audit were standardized
and SP) and artemisinin monotherapies. ACT was clas- using the adult equivalent treatment dose (AETD) to
sified as two mutually exclusive categories: quality- allow meaningful comparisons between anti-malarials
assured (appearing on the WHO list of approved formu- with different treatment courses. For the five countries
lations of ACT or the UNICEF procurement records) that conducted surveys in 2009, prices were converted
and non-quality assured. Quality-assured ACT was to US dollars using their average annual exchange rate
further divided into 1) quality-assured first-line ACT [23]. For Madagascar (where data were collected in
and 2) quality-assured non first-line ACT (see Table 1). 2010), local currency prices were adjusted to 2009 prices
Artemisinin monotherapies were further divided into using the Madagascar consumer price index [24], and
oral and non-oral artemisinin monotherapies. Data were then converted to US dollars using the 2009 average
not collected on drugs intended solely for malaria che- exchange rate.
moprophylaxis as the focus was on drugs for treatment. Prices per AETD are presented for the three anti-
The antibiotic cotrimoxazole was also excluded as, malarial categories believed to be most pertinent to pol-
although it has anti-malarial properties, it is very rarely icy level decisions; first-line quality-assured ACT, the
used as a malaria treatment. most popular anti-malarial based on volume (defined by
generic name), and oral artemisinin monotherapy.
Availability Price measures include tablet anti-malarials only as
Availability was calculated using two methods. For each this is the most common formulation. The price of non-
country, availability of anti-malarials was measured as tablet formulations, such as powders for reconstitution,
the proportion of outlets with at least one anti-malarial suspensions, suppositories and syrups, was excluded.
in stock, among all censused outlets. Within each coun- The price distributions for these non-tablet formulations
try, the availability of specific anti-malarial categories, tend to be different from those for tablets, with much
restricted to those outlets that had anti-malarials in higher medians, implying that it could be misleading to
stock, was also calculated. Thus, for example, the avail- use one measure of central tendency for all formula-
ability of ACT was measured as the proportion of out- tions. However, measures of volume include all dosage
lets stocking ACT, among all outlets with at least one forms to provide a complete assessment of anti-malarial
anti-malarial in stock. market shares.
In a supplementary analysis, the availability in public
health facilities was also measured as the proportion of Ethical approval
facilities with different categories of anti-malarials in The study was approved by the local ethical review
stock, among all censused public health facilities. This boards: Benin Ministry of Health, Authorization for
operationalized the expectation that public health facil- Data Collection Number 3989/MS/DC/SGM/DRS/SCI/
ities should stock anti-malarials. SA, 14 th July 2008; Cambodia, Ministry of Health,

Table 1 Description of Quality-Assured ACT classifications

First-line quality-assured ACTs: Non first-line quality-assured ACTs:
Definitions Government recommended first-line treatments (regardless of ACTs that are not the government’s recommended first-line
strength) for uncomplicated malaria that appear on the WHO list treatment for uncomplicated malaria, but which appear on the
of approved ACTs or the UNICEF procurement records. WHO list of approved ACTs or the UNICEF procurement records.
Brands on these lists and audited in each country were: Brands on these lists and audited in each country were:
Benin Artefan 20/120; Coartem; Lumartem 20/120; Lumet 20/120 Coarsucam; Winthrop
DRC Arsuamoon; Artesunate & Amodiaquine from Ipca; Artesunate & Artefan 20/120; Coartem 20/120; Lumartem 20/120
Amodiaquine from Cipla; Falcimon Kit; Serenadose; Coarsucam;
Winthrop
Madagascar Arsuamoon; Artesunate & Amodiaquine from Ipca; Actipal 50/153; Artefan 20/120; Coartem 20/120;
Winthrop; Coarsucam; Falcimon kit; Larimal Coartem D 20/120;
Lumartem 20/120
Nigeria Coartem 20/120; Coartem D 20/120; Lumerax 20/120 *Arsuamoon; Artecospe
*Coarsucam; *Larimal
Uganda Coartem; Artefan 20/120; Lumartem 20/120 Larimal, Falcimon kit
Zambia Artefan 20/120; Coartem 20/120; Lumartem 20/120; Lumerax; Arsuamoon; Artecospe
Lumet 20/120
*Alternate first-line treatment in Nigeria
O’Connell et al. Malaria Journal 2011, 10:326 Page 5 of 14
http://www.malariajournal.com/content/10/1/326

National Ethics Review Committee, Reference Number facilities, pharmacies and drug stores were more likely
109 NECNR, 7th November 2008; Democratic Republic to stock anti-malarials than grocery stores and other
of Congo, Ministerie de L’Enarignement Superior at outlets, where stocking rates were generally less than
Universitaire, School of Public Health, Reference Num- 5%. Exceptions were grocery stores in Benin and Mada-
ber ESP/CF/020/2008, 11th June 2008; Madagascar, Min- gascar and market stalls in Benin, with around one in
istry of Health du Planning Familial et de la Protection three of these outlets stocking anti-malarials.
Sociale, Reference Number 206SANPFPS, 16 th June Figure 1 shows the relative distribution of all outlets
2008; Nigeria; Federal Department of Ministry of Health, that had at least one anti-malarial in stock, by country.
Department of Public Health, Reference Number MH/ Results show the diversity of distribution of outlet types
1158/5/137, 20th October 2008; Uganda, Makerere Uni- stocking anti-malarials. For example, public health facil-
versity, Faculty of Medicine, Reference Number 2008- ities varied from 4% of all outlets stocking anti-malarials
057, 1 st September 2008; and Zambia, University of in Nigeria to 31% in Zambia. In Madagascar, grocery
Zambia, Biomedical Ethical Research Committee, Refer- stores constituted 73% of all outlets stocking anti-malar-
ence Number 014-08-08, 2nd October 2008. ials, as compared to 2% in Uganda. Drug stores were
the most common type of outlet stocking anti-malarials
Results in Nigeria, Uganda and the DRC; groceries and stalls
A total of 28,263 outlets across six countries were were most common in Madagascar and Benin respec-
approached to participate in the surveys. In total, 348 tively; and public health facilities were the most com-
outlets were closed down permanently, 665 outlets were mon outlets stocking anti-malarials in Zambia.
not open, 299 eligible providers were not available for Among outlets stocking at least one anti-malarial
interview, 395 providers refused to participate and 301 (Table 3), non-artemisinin therapies were generally the
outlets were excluded for other reasons. As a result, most commonly stocked category of anti-malarial in both
26,255 outlets were screened for stocking anti-malarials. the public/not-for-profit sector and private sector. A sig-
Of these, 9,118 outlets met the screening criteria and nificant exception to this was found in Madagascar’s pub-
were administered the questionnaire (8,383 outlets were lic/not-for-profit sector, which was more likely to stock
found to have at least one anti-malarial in stock and first-line quality-assured ACT (92%) than non-artemisi-
735 outlets had no anti-malarials in stock at the time nin therapies (36%). In the private sector, stocking rates
but reported having stocked anti-malarials in the past of first-line quality-assured ACT were less than 25%
three months). across all countries, and for most countries significantly
Among outlets stocking anti-malarials on the day of lower than the public/not-for-profit sector (p < .05)
survey, 51,158 individual anti-malarial products were (Table 3). Stocking rates of oral artemisinin monothera-
audited (66% tablets). The proportion of tablet forma- pies were generally low, though exceptions included pri-
tions in the audit varied by country (Benin, 73%, n=5, vate sector outlets in the DRC (41%) and Nigeria (45%).
232; the DRC, 58%, n=11, 939; Madagascar, 87%, Availability of different categories of anti-malarials in
n=5,579; Nigeria, n=61%; n=20,841; Uganda, 63%, public health facilities was also calculated among all
n=5,784; and Zambia, 82%, n=1,783). censused public health facilities (see Additional File 1
Table 1). Among all public health facilities (excluding
Availability community health workers), stockage rates of first-line
Table 2 shows the availability of anti-malarials as the quality-assured ACT at the time of survey were as fol-
proportion of outlets with at least one anti-malarial in lows: Benin, 81%; the DRC, 82%; Madagascar, 80%;
stock, among all censused outlets. The proportion of Nigeria 43%; Uganda, 72% and Zambia, 85%. The stock-
outlets with any anti-malarials in stock at the time of ing rate of SP among all outlets in the public sector,
the interview varied by outlet type. Overall, stocking used in intermittent preventive treatment of pregnant
rates in the public/not-for-profit sector were generally women (IPTp), also varied (Benin, 50%; the DRC, 70%;
high (~90%) with the exception of Madagascar (40%) Madagascar, 46%; Nigeria 63%; Uganda, 62% and Zam-
and Uganda (70%). Both these countries had large num- bia, 59%). Quinine injection, the government-recom-
bers of CHWs included in the census, with less than mended treatment for severe malaria for persons who
half of these providers having anti-malarials available cannot take oral medication, ranged in public sector
(27% and 40% respectively). In public health facilities, availability as well: Benin, 61%; the DRC, 66%; Madagas-
stocking rates were greater than 90% across all car, 43%; Nigeria 16%; Uganda, 73% and Zambia, 53%.
countries.
In the private sector, stocking of anti-malarials was Price
lower and varied by outlet type and country, ranging First-line quality-assured ACT was free at public sector
from 6% in Zambia to 36% in Benin. Private health outlets in four of the six countries. In Benin and the
 http://www.malariajournal.com/content/10/1/326
O’Connell et al. Malaria Journal 2011, 10:326
(N)
Table 2 Proportion of outlets censused with at least one anti-malarial in stock on the day of interview, by sector, outlet type and country¹
Public Sector/Not-for-Profit Private Sector Total
Public Community Private not-for-profit Total Private Pharmacy Drug Grocery Other Outlet Types Total All
Health Health Worker Health Facility Public/Not- Health Store Store Private
Facility for-Profit Facility
Shop/Kiosk/Bar/ Itinerant
Market Stall Provider
(182) (47) (229) (118) (118) (433) (691) (81) (1, 441) (1, 670)
Benin 95.4 - 91.2 94.0 84.2 96.7 - 30.5 34.4 42.7 36.3 39.0
(111) (33) (144) (204) (33) (1, 089) (2, 245) (3, 571) (3, 715)
DRC 96.8 - 97.2 96.9 75.7 100.0 96.5 - 1.7 - 24.9 28.4
(531)
Madagascar 96.8 26.8 (226) 80.6 (7)
40.4 (764)
87.6 (87)
99.6 (69) 97.4 (263)
33.1 (5, 056)
1.1 (530)
- 33.9 (6, 005)
35.0 (6, 769)

(255) (19) (11) (285) (405) (409) (1, 031) (2, 141) (1, 164) (21) (5, 171) (5, 456)
Nigeria 91.8 80.0 98.7 89.2 91.4 99.5 95.6 3.8 2.5 70.2 25.7 26.6
(525) (90) (11) (626) (208) (97) (398) (3, 747) (191) (4.641) (5, 267)
Uganda 95.4 39.8 88.6 69.6 96.2 99.3 96.4 0.4 0.0 - 13.9 17.0
(165) (16) (181) (34) (50) (165) (1, 946) (997) (5) (3, 197) (3, 378)
Zambia 97.4 - 100.0 97.8 92.3 100.0 76.5 3.7 0.4 14.7 6.3 9.5
¹In Benin the drug store category was deemed redundant as, strictly, there were no unregulated private-sector medicine vendors operating from formal structures (such as permanent buildings). In Benin, such
medicine selling outlets should be registered and appear on the list of pharmacies. Unregulated vendors do operate however, often in markets, and were thus captured by the market stall and market shop
classifications. In the DRC, grocery stores, as defined for all other countries, were very rare. In the DRC and Zambia, community health workers were not targeted for inclusion due to difficulties in ascertaining their
presence in the clusters. In Benin, community health workers were not formally included in the Ministry of Health structure at the time of data collection [25]. It is expected that community health workers will be
included in later survey rounds, as their numbers increase and as the Ministry of Health makes moves to include them explicitly in national policy [26].

Page 6 of 14
O’Connell et al. Malaria Journal 2011, 10:326 Page 7 of 14
http://www.malariajournal.com/content/10/1/326

" # $ % $ / 0 1

& ' ( ) ) & ' 2 3 4 5 6

-   .

    

       

       

 

!


      

      

          

     
          

    * 

      

        

     

      

        

      


  

  


+ 
,   


    

*  

  

  
 

< = > = ? = @ A = B & % ? # B % =

    8    

& ' C 3 ) 2 ) & ' C 3 2 2 4

     .

          

    

        ;


:     

       

-   .

      


       7

     

  
 
 

    8    


    

* 

!


        

          

      

     

    9 

        

      


+ 
,     


     9  

D ? = $ > = G = H I % =

& ' 2 3 C C 6 & ' ) 4 6

:      E   F  


    

       

-   .

 

    

* 


    

  

7

       

  
 

      

    8    

!


          

     

      


+ 
,   

*  

          

   

* 

        

    


+ 
,   

    

  

  

        

      

Figure 1 Outlets stocking anti-malarials, by outlet type.

DRC, where public-sector patients pay for anti-malarials, Private sector first-line quality-assured ACT prices
the median prices were $1.29 (Inter Quartile Range varied considerably between countries (Table 4). Their
[IQR]: $1.29, $1.29) and $0.52 (IQR: $0.00, $1.29) median price per AETD was highest in Zambia ($9.63;
respectively for an adult equivalent treatment dose (see IQR: $3.01, $11.04), over $6 in Nigeria (IQR: $5.05,
Additional File 2 Table 2). $6.74), $4.50 in Uganda (IQR: $2.49, $5.97), around $3
O’Connell et al. Malaria Journal 2011, 10:326 Page 8 of 14
http://www.malariajournal.com/content/10/1/326

Table 3 Outlets with specific anti-malarial categories as a percentage of outlets with at least one anti-malarial in
stock, by sector¹
Benin DRC Madagascar Nigeria Uganda Zambia
Public/ Private Public/ Private Public/ Private Public/ Private Public/ Private Public/ Private
Not- for- Not- for- Not- for- Not- for- Not- for- Not- for-
Profit Profit Profit Profit Profit Profit
N=212 N=632 N=134 N=1, N=560 N=1, N=249 N=1, N=534 N=691 N=176 N=259
240 854 864
a b a b a b a a a b a b
Any ACT 71.4 6.4 84.4 52.7 92.2 8.4 60.6 36.7 81.5 20.1 82.2 19.3
a b a b a b a b a b a b
First-line 66.4 5.7 75.8 24.5 91.9 7.9 49.2 25.3 73.4 7.8 81.5 16.3
quality-assured
ACT
a b a b a b a a a b a b
Quality- 66.4 5.9 78.7 29.0 91.9 8.1 49.2 26.1 73.4 8.1 81.5 16.3
assured ACT
a b a b a a a a a a a b
Non-Quality- 5.1 0.5 5.6 23.6 0.3 0.3 11.5 10.5 9.3 15.2 0.8 2.9
assured ACT
a b a a a b a b a b a a
Any non- 96.7 99.7 93.8 97.8 36.3 99.3 81.7 98.2 68.6 99.6 98.4 98.8
artemisinin therapy
a b a b a b a b a b a b
Chloroquine 9.3 82.1 0.0 4.0 1.6 95.5 57.7 91.9 15.5 61.0 0.0 21.7
a b a a a a a a a a a a
SP 51.2 15.8 66.9 57.9 22.0 9.2 62.7 78.3 49.9 68.2 65.5 80.2
a b a a a a a a a b a b
Quinine 92.5 24.0 85.7 84.3 23.3 10.7 20.0 9.2 59.0 82.9 89.1 16.8
a b a b a a a b a a a b
Quinine 60.0 8.0 65.6 40.7 20.9 9.5 19.0 0.6 56.4 32.6 51.2 8.7
injection
a a a b a a a b a b a a
Oral artemisinin 3.8 0.7 10.2 40.5 0.0 < 0.1 4.1 45.1 1.3 13.3 1.1 2.5
monotherapy
¹Statistical comparisons conducted across outlet types, within countries. Statistical difference is labelled with a superscript, a or b. Proportions with different
letters in their superscripts differ significantly from one another within countries (p < 0.05 for all tests).

in Benin (IQR: $1.94, $5.77), under $2 in the DRC (IQR: The median private sector AETD price of the most
$1.03, $3.61), and lowest in Madagascar ($0.14; IQR: popular anti-malarial in tablet form, either SP or chloro-
$0.10, $0.57), where subsidized ACT was distributed quine in each country, ranged from $0.36 (IQR: $0.36,
through a social marketing campaign. $0.36) to $0.65 (IQR: $0.43, $1.08). This was

Table 4 Median price in US dollar (inter-quartile range) for an adult-equivalent treatment dose in the private sector
(tablet formulation only), by anti-malarial type
Median $ (IQR)
Most First-line Oral Artemisinin Monotherapy
Popular¹ quality-assured ACT
Benin2 0.65 a
(0.43, 1.08) N = 462 3.24 b
(1.94, 5.77) N = 216 8.10 (8.07, 10.45) N = 56
a b
DRC 0.39 (0.26, 0.52) N = 1, 258 1.86 (1.03, 3.61) N = 252 3.23 (2.45, 4.13) N = 956
Madagascar 0.36 a (0.36, 0.36) N = 1, 847 0.14 b
(0.10, 0.57) N = 302 (0 and 7.33) N = 2
Nigeria3 0.54 a (0.40, 0.81) N = 4, 061 6.40 b
(5.05, 6.74) N = 372 3.24 (2.70, 3.77) N = 1, 438
4 a b
Uganda 0.50 (0.30, 0.75) N = 653 4.48 (2.49, 5.97) N = 81 9.55 (7.96, 11.94) N = 229
a b
Zambia 0.40 (0.30, 0.61) N = 261 9.63 (3.01, 11.04) N = 83 6.74 (5.72, 6.74) N = 16
¹ The most popular ACT was chloroquine in Madagascar, and SP in all other countries. The most popular anti-malarial is based on volumes of anti-malarials sold
or distributed in the last week, within each country.
2
Statistical comparisons were conducted between the most popular treatment and the first-line quality-assured ACT. Statistical difference is labelled with
superscripts ‘a’ or ‘b’. Estimates with different letters in their superscripts differ significantly from one another within countries (p < 0.01 for all tests).
3
In Nigeria, price is presented for the first-line quality-assured ACT (AL), rather than the alternate first-line quality-assured ACT (ASAQ). The median price for the
alternate first-line treatment in Nigeria is $3.23 (1.89, 4.04) N = 622.
4
In Uganda the sampled clusters included two areas located in districts that were undertaking a pilot of subsidized ACT in the retail sector [27]. Due to the
presence of this pilot, the percent of private sector outlets stocking ACT was somewhat greater in these clusters than in the rest of the sample (50% and 19%
respectively). These areas comprised 1.4% of the total sample of private outlets, and while accounting for only 5.4% of the private sector ACT products audited
(25 out of 459 ACTs) they accounted for 17% of ACT products once sampling weights are taken into account (these clusters have high weights because they had
a relatively low chance of selection under PPS). As in the pilot districts private sector ACT had a much lower price than elsewhere in the country, inclusion of the
two pilot clusters can give a distorted picture of the average price available across the country as a whole. Uganda findings are therefore calculated both with
and without the subsidized product observations from the 2 clusters for the first-line quality assured treatment. Data in the price table are presented excluding
the subsidized product piloted in the clusters. The inclusion of the subsidized products provides a lower price for the first line quality assured treatment of $0.38
(0.38, 4.73) N = 104.
O’Connell et al. Malaria Journal 2011, 10:326 Page 9 of 14
http://www.malariajournal.com/content/10/1/326

significantly lower (p < 0.01) than the median price of was most pronounced in Nigeria, where 98% of anti-
the first-line quality-assured ACT in each country, with malarial volumes were delivered through the private sector
Madagascar as an exception, where the first-line quality- and only 2% of these were quality-assured ACT.
assured ACT was significantly less expensive than the For most countries, non-artemisinin therapies domi-
most popular anti-malarial, chloroquine. nated the overall market, followed by ACT and then
Overall, the median price of chloroquine tablets ran- oral artemisinin monotherapies. Exceptions to this find-
ged from $0.24 (IQR: $0.24, $0.48) to $0.48 (IQR: $0.48, ing were noted in Nigeria and the DRC where more
$0.48) and SP from $0.38 (IQR: $0.29, $ 0.48) to $0.65 oral artemisinin monotherapies were sold than ACT in
(IQR: $0.43, $1.08) in the private sector (see Additional the private sector. Across all countries quality-assured
File 3 Table 3). ACT accounted for less than 25% of total anti-malarial
Median prices of oral artemisinin monotherapies in volumes; private-sector quality-assured ACT volumes
the private sector ranged from $3.23 in the DRC (IQR: represented less than 6% of the total market share. In
$2.45, $4.13) to $9.55 in Uganda (IQR: $7.96, $11.94). the public/not-for-profit sector, SP made up the largest
market share of non-artemisinin therapies (Benin,
Market share 11.6%; the DRC, 16.2%; Madagascar, 9.5%; Nigeria 1.4%;
Figure 2 shows the market share of different categories of Uganda, 21.6%; Zambia, 34.1%).
anti-malarials sold or distributed in the seven days prior to
the survey. With the exception of Zambia, the private sec- Provider knowledge
tor played a larger role than the public/not-for-profit sec- Knowledge of the first-line treatment was significantly
tor in the distribution of anti-malarials. This difference higher in the public/not-for-profit sector than the

100

90

80

70

60

50


40

30

20

10

0

        
        
        
        
        
        

                 

                         

Oral artemisinin monotherapy Non-oral artemisinin monotherapy Non-artemisinin therapy

Other ACT Quality-assured ACT

Figure 2 Relative volumes of anti-malarials sold/distributed in the last seven days, by public/not-for-profit and private sectors and
anti-malarial category¹. ¹For each country, the public/not for profit and private columns sum to 100%. The figure shows 1) the contribution of
the pubic/not for profit and private market share of each sector and 2) the breakdown of each sector’s sales by antimalarial type.
O’Connell et al. Malaria Journal 2011, 10:326 Page 10 of 14
http://www.malariajournal.com/content/10/1/326

private sector across all countries: 44% to 93% of provi- knowledge, and address the policy implications of these
ders in the public/not-for-profit sector could correctly results.
state the first-line treatment, compared with 12% to 60%
in the private sector (Table 5). Similar patterns were Strengths and limitations
found for provider knowledge of the dosing regimens The ACTwatch survey methodology allows for the mea-
for the first-line treatment in children and adults. surement of anti-malarial availability and prices in a
Knowledge of any dosing regimens was particularly low standardized way, on a nationally representative sample
for the private sector in Benin and Madagascar, where of outlets, with multiple steps to ensure data quality.
less than 10% of providers were able to describe any When combined with the ACTwatch household survey
child or adult regimen correctly. [29], and the supply chain study on the structure and
operation of the anti-malarial distribution chain, ACT-
Diagnostics watch data provide a full picture of the anti-malarial
Each outlet was assessed for the availability of rapid market, with multiple uses for policy making.
diagnostic tests (RDTs) (see Additional File 4 Table 4). Despite the study’s strengths, the outlet survey data
In the public/not-for-profit sector, availability varied presented here have several limitations. 1) Providers
considerably by country (Benin, 72%; the DRC, 36%; may have had incentives to mis-report certain informa-
Madagascar, 43%; Nigeria, 7%; Uganda, 22% and Zam- tion, such as whether they stocked any anti-malarials, or
bia, 86%). In the private sector, availability was less than specific anti-malarial categories, especially if this was
20% and significantly lower across all countries than in not permitted by government regulations. Although
the public/not-for-profit sector. interviewers stressed that they were not connected with
any regulatory body, anti-malarial availability and stocks
Discussion may have been under-reported. 2) Volumes were based
This paper presents standardized, nationally representa- purely on provider recall. Given the lack of sales records
tive data on the anti-malarial market in six sub-Saharan in most private outlets, this was felt to be the most
African countries between 2009 and 2010. The data logistically feasible method for estimating volumes, but
confirm the low market share of ACT, reflecting a com- it is likely that some recall bias was present. 3) Adult
bination of low ACT availability in both the public/not- equivalent treatment dose calculations were used when
for-profit and private sectors, high ACT prices in the calculating prices and volumes. This has the advantage
private sector and low provider ACT knowledge, parti- of allowing direct comparability between drugs. How-
cularly in the private sector. Other studies have high- ever, it should be noted that in practice many customers
lighted similar challenges [8,10,11,17,28]. The following will be obtaining drugs for children and/or purchasing
sub-sections describe the limitations of the ACTwatch incomplete doses [30]. As a result, the price actually
outlet surveys, summarize the main findings according paid per customer may be considerably lower than the
to availability, price, market share and provider median price per AETD. Similarly, the actual number of

Table 5 Provider knowledge of first-line treatment and dosing regimens, by sector¹

Benin DRC Madagascar Nigeria Uganda Zambia
Public/ Private Public/ Private Public/ Private Public/ Private Public/ Private Public/ Private
Not-for- Not- Not- Not- Not- Not-
Profit for- for- for- for- for-
Profit Profit Profit Profit Profit
N=218 N=735 N=135 N=1, N=575 N=2, N=258 N=1, N=563 N=699 N=178 N=283
242 016 839
a b a b a b a b a b a b
Correctly state the 73.0 17.7 76.4 42.4 71.7 12.4 43.7 25.5 92.8 59.8 92.7 59.3
recommended first-line
treatment for uncomplicated
malaria
Correctly state the dosing 70.7 a
8.4 b
70.4 a
28.5 b
42.4 a
5.1 a
–2 – 86.6 a
53.2 a
92.7 a
48.2 b

regimen of the first-line

treatment for an adult
Correctly state the dosing 70.2 a
7.8 b
69.6 a
28.4 b
68.8 a
8.5 b
– – 89.5 a
49.3 b
92.7 a
47.3 b

regimen of the first-line

treatment for a two year old
¹ Statistical comparisons conducted across outlet types, within countries. Statistical difference is labelled with a superscript, a or b. Proportions with different
letters in their superscripts differ significantly from one another within countries (p < 0.01 for all tests).
2
Information on Nigeria’s dosing regimen for their first-line alternative ACT, ASAQ, was not collected.
O’Connell et al. Malaria Journal 2011, 10:326 Page 11 of 14
http://www.malariajournal.com/content/10/1/326

customers to whom each drug was sold will be consid- distributing anti-malarials, the higher cost of first-line
erably higher than the volume of AETDs dispensed. It quality-assured ACT is a concern. Although ACT is free
should also be noted that AETD prices presented in this or at a low price in the public sector, this is not where
paper do not include additional fees that may have been the majority of people seek treatment [29].
incurred for customers, such as consultation fees or ser- In Madagascar, the price of quality-assured ACT in
vice charges. 4) While price indicators were standar- the private sector is exceptionally low at $0.14. This is
dized according to the 2009 US dollar, it is due to the presence of a socially-marketed ACT, which
acknowledged that there are different costs of living is sold nationally at a highly subsidized price through
within countries and varying average or minimum pharmacies and drug stores. This mirrors findings from
wages, which make like-with-like comparisons across a pilot subsidy programme in Tanzania which demon-
countries challenging. 5) Data were only collected at strated that after the introduction of an ACT subsidy,
one point in the year, generally around the peak malaria ACT prices were as low as $0.58 in the private sector
transmission season. Seasonal variation may be and did not differ significantly from the price paid for
expected, meaning in particular that anti-malarial avail- SP, the most common alternative [32].
ability and sales volumes may have been higher during
the surveys than their annual averages, especially in Market share
countries with marked seasonal transmission, such as Market share data highlight the dominant role of the
Madagascar and Zambia. 6) While a methodological private sector across all countries except Zambia, where
strength of the survey is the full census of outlets in 60% of all anti-malarials pass through the public/not-
selected clusters, there are a number of practical imple- for-profit sector. Of the other countries, the private sec-
mentation challenges with this method. Conducting a tor was the most important in Nigeria accounting for
full census requires interviewers in the clusters to screen 98% of anti-malarial sales volumes, and elsewhere ran-
every outlet. It is acknowledged that some outlets with ged from 58%-82%. Unfortunately, most of the anti-
the potential to sell anti-malarials may have been malarials sold through the private sector are non- arte-
missed, thus distorting the true figure of anti-malarial misinin therapies, as well as concerning amounts of oral
availability across countries. artemisinin monotherapies in the DRC and Nigeria (7%-
8% of the total market). ACT comprised less than 25%
Availability of the total volume of anti-malarials distributed across
Overall availability of ACT was low, particularly in the all sectors. Quality-assured ACT fared even worse in the
private sector. In the private sector, non-artemisinin private sector, where it generally accounted for less than
therapies, typically chloroquine or SP, were by far the 6% of the total market share.
most abundant anti-malarial available while first-line Of note, in Nigeria and the DRC, the market share of
quality-assured ACT was available in less than one- ACT was particularly low and more oral artemisinin
quarter of the anti-malarial stocking outlets. Oral arte- monotherapies were sold than ACT in the private sec-
misinin monotherapies were available in more than 40% tor. In Nigeria, oral artemisinin monotherapy is less
of private sector anti-malarial-stocking outlets in the expensive than first-line quality-assured ACT treatment,
DRC and Nigeria. Within the public/not-for-profit sec- perhaps explaining why higher volumes of oral artemisi-
tor, quality-assured ACT stockage rates ranged between nin monotherapies are found.
49%-92%.
When observing stocking rates for all public health Provider knowledge
facilities, availability of ACT hovered around 80%, but in Provider knowledge was generally lower in the private
Nigeria less than half of facilities had a quality assured sector than in the public/not- for-profit sector. Other
ACT in stock. This suggests that maintaining a consis- studies have shown that medicine seller knowledge of
tent supply to public health facilities may be difficult, a drugs and doses, particularly in the private sector, is
finding supported by other facility based research, which often poor [13,15,16]. This is likely to be exacerbated by
also highlights stock-outs as a common problem in pub- the relatively recent introduction of ACT in sub-Saharan
lic health facilities [31]. Africa because these have new dosage requirements and
more than one product may be available with different
Price dosages [13].
With one notable exception in Madagascar, first-line
quality-assured ACT treatment in the private sector was Policy implications
significantly more expensive, between 5-23 times so, Poor ACT availability in both public/not-for-profit and
than the most popular anti-malarial (chloroquine or SP). private sectors, as well as the high price observed in the
Given the importance of the private sector in private sector, and poor provider knowledge, serve as
O’Connell et al. Malaria Journal 2011, 10:326 Page 12 of 14
http://www.malariajournal.com/content/10/1/326

barriers to obtaining effective treatment. Of grave con- Prominent among private sector initiatives is the
cern, quality-assured ACT availability in the public sec- Affordable Medicines Facility for Malaria (AMFm),
tor ranges between 43-85%. This suggests that there are which began operations in pilot countries during 2010,
issues within the public-sector supply chain that require and provides a mechanism for heavily subsidized, qual-
urgent attention, especially in Nigeria and Uganda ity-assured ACT to be sold at a target retail price of
where stocking rates were lowest. $0.50 [21]. However, there continues to be controversy
In the private sector, non-artemisinin therapies are about such initiatives, with some arguing that the sub-
more widely available and more commonly sold than sidy will be captured by middle-men in the distribution
other classes of anti-malarials; these medicines are also chain; that market power in the private sector will lead
usually sold at a much lower price than ACT. Given the to continued problems of affordability and inaccessibility
relative affordability and accessibility of non-artemisinin by the poor; and that widespread availability of ACT
therapies, consumers have strong incentives to choose without improving access to diagnosis will result in mis-
the ‘wrong’, ineffective, anti-malarial when seeking use of ACT and fuel the development of resistance
treatment. [35,38].
Oral artemisinin monotherapy, discouraged for its Of the countries covered in this paper, Madagascar
potential to speed the spread of artemisinin resistance, provides an interesting example of a setting where an
was commonly available and sold at private outlets in ACT subsidy has been in operation since 2008. The sub-
the DRC and Nigeria. These disappointing findings sidized ACT was open to all wholesalers registered or
should act as a rallying call given the Millennium Devel- unregistered in the country and it was sold nationally
opment Goal targets and that together these two nations through community health workers, pharmacies and
account for nearly half of the total global burden of rural drug stores. The ACT did not have over-the-coun-
malaria disease [33]. While all the ACTwatch countries ter status and had a recommended retail price of $0.05.
have taken regulatory measures to withdraw the market- In contrast to the other countries, the price of quality-
ing authorization of oral artemisinin monotherapies, the assured ACT was lower than non-artemisinin therapies.
manufacturing and marketing of these products is still However, accessibility remained low, as ACTwatch data
ongoing (and many of these companies are located in show that less than 10% of private sector outlets stock-
Nigeria, though renewal licenses will not be granted to ing this type of ACT. Provider knowledge of the first-
existing companies currently marketing oral artemisinin line treatment was also low (12%). As such, substantial
monotherapies [34]). This calls into question the effec- gains in market share of ACT were not apparent, and
tiveness of current regulatory practices in countries rates of ACT use in febrile children remained less than
where this drug is banned. Interventions such as tighter 5% [29]. Despite this, smaller scale pilots of ACT subsi-
regulation and building awareness of issues with oral dies have shown impressive results [27,32,39]. Promising
artemisinin monotherapies should be considered so that findings have also been demonstrated in Cambodia,
this national policy of public health significance is where an ACT social marketing programme has been in
indeed followed. For other countries, availability of oral place for the past 10 years [40].
artemisinin monotherapy was low, suggesting a positive The data suggest a need for interventions to address a
outcome from the 2006 WHO regulatory ban on oral full range of supply side issues that go above and
artemisinin monotherapy, and heightened government beyond a price subsidy (i.e. targeting provider knowl-
response to the threat of artemisinin resistence. edge, increasing supply and product promotion). Inter-
As the majority of anti-malarials are delivered through ventions should also focus on demand creation through
the private sector, there is a clear need to consider how activities that include behaviour change communication,
to handle the role of the private sector when developing to encourage the stocking and uptake of ACT, particu-
anti-malarial drug policies. Some argue that emphasis larly in the private sector. It remains to be seen how far
should remain on the public sector, to attract more peo- the “supporting interventions, “ (e.g. provider trainings
ple to outlets where care is regulated and provided by and behaviour change communication, all to promote
qualified staff [35]. Others argue that, given the current ACT utilization), being implemented as part of the
state of developing country health systems, while the AMFm and other initiatives will go in addressing these
public sector cannot be ignored, high private sector use concerns.
is inevitable in many countries, and therefore strategies
should be adopted to improve its quality [36]. The Conclusions
extent to which the emphasis should be placed on the ACTwatch provides standardized, nationally representa-
private sector continues to be debated [37] and data tive data on anti-malarial supply from multiple countries
presented here provide evidence to help facilitate further collected within 16 months. Data presented in this paper
discussion. report on key, policy relevant indicators across
O’Connell et al. Malaria Journal 2011, 10:326 Page 13 of 14
http://www.malariajournal.com/content/10/1/326

countries. The results typically demonstrate the low Androhibe Antananarivo, 101, Madagascar. 12National Malaria Control
Programme, Abia House, Near Adamawa Plaza, First Avenue, Central District
availability, low market share and high price of ACT, at Area, Postal Code 900001, Federal Capital Territory, Abuja, Nigeria. 13Ministry
least in the private sector where most anti-malarials are of Health, Malaria Control Programme, Lourdel Rd Wandegeya, Kampala,
being accessed, with some exceptions. Data from ACT- Uganda. 14National Malaria Control Center, P.O Box 32509, Lusaka, Zambia.
15
Population Services International, 1120 19th Street N.W., 20036, Washington
watch outlet surveys confirm that there is substantial D.C., USA.
room to improve availability and affordability of ACT
treatment in six countries across sub-Saharan Africa, Authors’ contributions
KOC, SC, DC devised the study design and objectives. CBA, JAA, BFR, BC,
and in both the public/not-for-profit and private sectors. DR, BH, CZ, LA, JR, EA, PM, FM, TS contributed to the country specific
The data will also be useful for monitoring the impact study design adaptations and interpretations. CZ, LA, JR, EA, PM, FM, TS,
of interventions such as the Affordable Medicines Facil- EM, JN contributed to field implementation and data collection. HG, JN, IE,
EM, SP, KOC contributed to the analysis. KOC wrote the first draft of the
ity for malaria. More detailed, country specific findings manuscript. DC, TS, CG, KH, HG, EM, SP contributed to interpretation and
(for example, outlet anti-malarial availability by country subsequent drafts of the manuscript. All authors read and approved the
strata) can be found in country reports on the ACT- final manuscript.
watch website [41]. Competing interests
The authors declare that they have no competing interests.
Additional material
Received: 29 July 2011 Accepted: 31 October 2011
Published: 31 October 2011
Additional file 1: Proportion of all public health facilities stocking at
least one anti-malarial on the day of survey. This table shows the
References
availability of different categories of anti-malarials in public health 1. World Health Organization: The African Summit on Roll Back Malaria, Abuja,
facilities as the proportion of facilities with anti-malarials in stock, among Nigeria. Geneva 2000.
all censused public health facilities. 2. RBM: Global Strategic Plan: Roll Back Malaria 2005-2015. Geneva 2005.
Additional file 2: Median price in US dollar (inter-quartile range) of 3. World Health Organization: World Malaria Report 2009. Geneva 2009 [http://
adult-equivalent anti-malarial treatment doses in the public sector www.who.int/malaria/world_malaria_report_2009/en/index.html].
(tablet formulation). This table shows the median price of the first-line 4. Kachur SP, Black C, Abdulla S, Goodman C: Putting the genie back in the
quality assured ACT anti-malarial treatment doses in the public sector bottle? Availability and presentation of oral artemisinin compounds at
across all countries. retail pharmacies in urban Dar-es-Salaam. Malar J 2006, 5:25.
Additional file 3: Median price in US dollar (inter-quartile range) of 5. Yeung S, Damme WV, Socheat D, White NJ, Mills A: Cost of increasing
adult-equivalent anti-malarial treatment doses in the private sector access to artemisinin combination therapy: the Cambodian experience.
(tablet formulation), by antimalarial type. this table shows the Malar J 2008, 7:84.
median price of SP and CQ anti-malarial treatment doses in the private 6. Eastman RT, Fidock DA: Artemisinin-based combination therapies: a vital
sector across all countries. tool in efforts to eliminate malaria. Nat Rev Microbiol 2009, 7:864-74.
7. Buabeng KO, Duwiejua M, Matowe LK, Smith F, Enlund H: Availability and
Additional file 4: Percent of outlets stocking RDTs, by outlet choice of antimalarials at medicine outlets in Ghana: the question of
category. This table shows the availability of RDTs in public/not for access to effective medicines for malaria control. Clin Pharmacol Ther
profit and private sectors across all countries.
2008, 84:613-619.
8. Medicines for Malaria Venture: Understanding the Antimalarials Market:
Uganda 2007-An overview of the supply side. Geneva 2008 [http://www.mmv.
org/newsroom/publications/understanding-antimalarials-market-uganda-
Acknowledgements 2007-overview-supply-side].
The authors are grateful to country teams who undertook the surveys. The 9. Kangwana BB, Njogu J, Wasunna B, Kedenge SV, Memusi DN, Goodman CA,
authors are also grateful to Amin Abdinasir, Kelly Safreed-Harmon, Susan Zurovac D, Snow RW: Malaria drug shortages in Kenya: a major failure to
Duvall, Meg Kays, and Angus Spiers for comments on an earlier draft of the provide access to effective treatment. Am J Trop Med Hyg 2009,
manuscript. We are indebted to the study participants and the enumerators 80:737-738.
for data collection. 10. Amin AA, Snow RW: Brands, costs and registration status of anti-malarial
This study received financial support from the Bill and Melinda Gates drugs in the Kenyan retail sector. Malar J 2005, 4:36.
Foundation (#058992) and from the Global Fund to Fight AIDS, Tuberculosis 11. Amuasi JH, Diap G, Blay-Nguah S, Boakye I, Karikari PE, Dismas B, Karenzo J,
and Malaria (PO 20013306). Nsabiyumva L, Louie KS, Kiechel JR: Access to artesunate-amodiaquine,
quinine and other anti-malarials: policy and markets in Burundi. Malar J
Author details 2011, 10:34.
1
Population Services International, Malaria & Child Survival Department, P.O. 12. Larson BA, Amin AA, Noor AM, Zurovac D, Snow RW: The cost of
Box 43640, Nairobi, Kenya. 2Department of Global Health and Development, uncomplicated childhood fevers to Kenyan households: implications for
London School of Hygiene and Tropical Medicine, 15-17 Tavistock Place, reaching international access targets. BMC Public Health 2006, 6-314.
London WC1H 9SH, UK. 3Association Béninoise pour le Marketing Social/PSI, 13. Wafula FN, Goodman CA: Are interventions for improving the quality of
B.P. 08-0876, Tri Postal, Cotonou, Benin. 4Association de Santé Familiale, 4630 services provided by specialized drug shops effective in sub-Saharan
Avenue de la Science, Immeuble USTC, Bloc C, Gombe, Kinshasa, Democratic Africa? A systematic review of the literature. Int J Qual Health Care 2010,
Republic of Congo. 5PSI/Madagascar, Immeuble- FIARO, Rue Jules Ranaivo, 22:316-323.
Escalier-D, 2eme Etage, BP 7748, Antananarivo 101, Madagascar. 6Society for 14. Diap G, Amuasi J, Boakye I, Sevcsik AM, Pecoul B: Anti-malarial market and
Family Health, 8 Port Harcourt Crescent, Area 11 Garki Abuja, Nigeria. 7PACE, policy surveys in sub-Saharan Africa. Malar J 2010, 9:S1.
Plot 2 Ibis Vale, P.O. Box 27659, Kololo, Kampala, Uganda. 8Society for Family 15. Hetzel MW, Obrist B, Lengeler C, Msechu JJ, Nathan R, Dillip A,
Health, Plot No. 549, Ridgeway, P.O. Box 50770, Lusaka, Zambia. 9Ministère Makemba AM, Mshana C, Schulze A, Mshinda H: Obstacles to prompt and
de la Santé, Programme National de Lutte contre le Paludisme (PNLP), effective malaria treatment lead to low community-coverage in two
Akpakpa, Cotonou, Bénin. 10National Malaria Control Programme, 1, Avenue rural districts of Tanzania. BMC Public Health 2008, 16:317.
du Tourisme, Ngaliema, Kinshasa, Democratic Republic of Congo. 11Ministre 16. Minzi OM, Haule AF: Poor knowledge on new malaria treatment
de la Santé Publique, Centre National de Lutte contre le Paludisme (CNLP), guidelines among drug dispensers in private pharmacies in Tanzania:
O’Connell et al. Malaria Journal 2011, 10:326 Page 14 of 14
http://www.malariajournal.com/content/10/1/326

the need for involving the private sector in policy preparations and 39. Kangwana B, Kedenge SV, Noor AM, Alegana VA, Nyandigisi AJ, Pandit J,
implementation. East African J of Pub Health 2008, 5:117. Fegan GW, Todd J, Brooker S, Snow RW, Goodman C: The impact of retail
17. Goodman C, Kachurc SP, Abdullad S, Blolande P, Mills A: Concentration sector delivery of artemether-lumefantrine on malaria treatment of
and drug prices in the retail marketfor malaria treatment in rural children under five in Kenya-a cluster randomized controlled trial. PLoS
Tanzania. Health Econ 2009, 18:727-742. Med 2011, 8:e1000437.
18. Alba S, Hetzel MW, Goodman C, Dillip A, Jafari L, Mshinda H, Lengeler C: 40. Littrell M, Gatakaa H, Phok S, Allen H, Yeung S, Chuor CM, Dysoley L,
Improvements in access to malaria treatment in Tanzania after switch to Socheat D, Spiers A, White C, Shewchuck T, Chavasse D, O’Connell KA: Case
artemisinin combination therapy and the introduction of accredited management of malaria fever in Cambodia: results from national anti-
drug dispensing outlets-a provider perspective. Malar J 2010, 9:164. malarial outlet and household surveys..
19. Kindermans JM, Vandenbergh D, Vreeke E, Olliaro P, D’Altilia JP: Estimating 41. The ACTwatch Group. [http://www.actwatch.info].
anti-malarial drugs consumption in Africa before the switch to
artemisinin-based combination therapies (ACTs). Malar J 2007, 6:91. doi:10.1186/1475-2875-10-326
20. Sabot O, Yeung S, Pagnoni F, Gordon M, Petty N, Schmits K, Talisuna A: Cite this article as: O’Connell et al.: Got ACTs? Availability, price, market
Distribution of Artemisinin- Based Combination Therapies through Private- share and provider knowledge of anti-malarial medicines in public and
Sector Channels Lessons from Four Country Case Studies. Washington, DC private sector outlets in six malaria-endemic countries. Malaria Journal
2008. 2011 10:326.
21. Adeyi O, Atun R: Universal access to malaria medicines: innovation in
financing and delivery. Lancet 2010, 376:1869-1871.
22. Shewchuk S, O’Connell KA, Goodman C, Hanson K, Chapman D,
Chavasse D: ACTwatch: a project to study antimalarial markets in seven
countries..
23. OANDA [https://fxtrade.oanda.com/your_account/fxtrade/register/forex-
demo-account/signup/why-oanda?xid=2227&cr_id=3309420362].
24. International Monetary Fund: World Economic Database [http://www.imf.
org/external/pubs/ft/weo/2011/01/weodata/index.aspx].
25. National Malaria Control Programme: National Strategic Plan, Ministry of
Health. Benin: 2006-2010 .
26. President’s Malaria Initiative: Country Profiles, Benin [http://www.
fightingmalaria.gov/countries/profiles/benin.html].
27. Talisuna A, Grewal P, Rwakimari JB, Mukasa S, Jagoe G, Banerji J: Cost is
killing patients: subsidising effective antimalarials. Lancet 2009,
374:1224-6.
28. Bosman A, Mendis KN: A major transition in malaria treatment: the
adoption and deployment of artemisinin based combination therapies.
Am J Trop Med Hyg 2007, 77:193-197.
29. Littrell M, Gatakaa H, Evance I, Poyer S, Njogu J, Solomon T, Munroe E,
Chapman S, Goodman C, Hanson K, Zinsou C, Akulayi L, Raharinjatovo J,
Arogundade E, Buyungo P, Mpasela F, Adjibabi CB, Agbango JA,
Ramarosandratana BF, Coker B, Rubahika D, Hamainza B, Shewchuk T,
Chavasse D, O’Connell KA: Monitoring fever treatment behavior and equitable
access to effective medicines in the context of initiatives to improve ACT
access: baseline results and implications for programming in six African
countries .
30. Kagashe GA, Minzi O, Matowe L: An assessment of dispensing practices in
private pharmacies in Dar-es-Salaam, Tanzania. Int J Pharm Pract 2011,
19:30-5.
31. Lufesi NN, Aursnes AM: Deficient supplies of drugs for life threatening
diseases in an African community. BMC Health Services Research 2007, 7:86.
32. Sabot OJ, Mwita A, Cohen JM, Ipuge Y, Gordon M, Bishop D, Odhiambo M,
Ward L, Goodman C: Piloting the global subsidy: the impact of
subsidized artemisinin-based combination therapies distributed through
private drug shops in rural Tanzania. PLoS One 2009, 4:e6857.
33. Breman JG, Mills A, Snow RW, Mulligan JA, Lengeler C, Mendis K, Sharp B,
Morel C, Marchesini P, White NJ, Steketee RW, Doumbo OK: “Conquering
malaria”. 2006. In Disease Control Priorities in Developing Countries.. 2
edition. Edited by: Jamison DT, Breman JG, Measham AR, Alleyne G. New
York: Oxford University Press; 2006:413-432.
34. World Health Organization: Marketing of oral artemisinin-based
monotherapy medicines. [http://www.who.int/malaria/ Submit your next manuscript to BioMed Central
monotherapy_manufacturers.pdf]. and take full advantage of:
35. Oxfam: Blind Optimism London 2008 [http://www.oxfam.org.uk/resources/
policy/health/downloads/bp125_blind_optimism_private_health_care.pdf].
• Convenient online submission
36. World Bank: Healthy Development: The World Bank Strategy for Health,
Nutrition, and Population Results Washington DC 2007. • Thorough peer review
37. Hanson K, Gilson L, Goodman C, Mills A, Smith R, Feachem R, Feachem NS, • No space constraints or color figure charges
Koehlmoos TP, Kinlaw H: Is private health care the answer to the health
problems of the world’s poor? PLoS Med 2008, 5:e233. • Immediate publication on acceptance
38. Moon S, Pérez Casas C, Kindermans JM, de Smet M, von Schoen-Angerer T: • Inclusion in PubMed, CAS, Scopus and Google Scholar
Focusing on quality patient care in the new global subsidy for malaria
• Research which is freely available for redistribution
medicines. PLoS Med 2009, 6:e1000106.

Submit your manuscript at

www.biomedcentral.com/submit