Forecasting global procurement of malaria rapid diagnostic tests: estimates and uncertainties

World Health Organization

Western Pacific Region

Forecasting global procurement of malaria rapid diagnostic tests: estimates and uncertainties
While malaria rapid diagnostic tests (RDT) may bring overall benefits to malaria case management through improved diagnostic accuracy and reduction in unnecessary drug use, cost and quality of RDT remain significant variables affecting procurement. Over the past few years, RDT procurement orders have increased in number (compiled data, The Global Fund to Fight AIDS, Tuberculosis and Malaria) but often with short delivery times requested by procurers. This places burdens on manufacturers of RDT and their components to ramp up production at short notice, reducing potential economies of scale and increasing pressure on quality assurance processes during production. A reliable forecast of RDT procurement can guide manufacturers in anticipating orders, particularly components manufacturers, allowing a reduction in price and time to ensure maintenance of quality. Thus, this report examines past procurement and diagnostic practices to estimate RDT future procurement. METHOD OF ESTIMATION
The methods used to estimate RDT procurement included estimates of RDT need, the capacity of health services to utilize RDT and recent components and procurement figures from a limited number
PAGE 1 OF 4

FIGURE 1: Global RDT Procurement Forecasts, assuming (1) a continuation of recent exponential increase, (2) a linear increase, and (3) a linear increase adjusted due to limitation of the public health ceiling. Figure is based on 2005 estimates; ceiling likely to rise as health sector funding improves (see text).
100 90 80 70
Global public health ceiling Continuing exponential increase
tren d li ne

RDTs in millions

60 50 40 30 20 10 0 2000 2002 2004 2006 2008

RDT production 28 250 000
L

ar ine

Adjusted forecast

2010

2012

2014

2015

of sources detailed in Subdocument A. There are inherent inaccuracies in the figures due to limited available data, and results should be viewed as best estimates. RDT procurement was projected to 2015 using a trend line based on RDT production data from 2000 to 2005. Data for production as opposed to procurement was used for the forecast on the basis of its relative reliability. Assuming that procurement will not exceed the global RDT market ceiling calculated above, a trend line, approaching but limited by the ceiling, was produced according to the equation y = (5x106)x - (5x106). The upper curve is
IN SUMMARY

Potential Public Sector RDT Market Global RDT Market Ceiling

= =

(number of fever cases recorded in the public health sector) Potential Public Sector RDT Market

(number of microscopy diagnoses)

+ UNICEF/NGO market
www.wpro.who.int/sites/rdt

(Assumes the private sector market is small).

World Health Organization

Western Pacific Region

Forecasting global procurement of malaria rapid diagnostic tests: estimates and uncertainties

projected from the current trend, ignoring the ceiling and using Microsoft Excel software. It is recognized that the global RDT market ceiling may shift due to changes in health service capacity, fluctuations in fever cases and unpredictable nongovernmental and private sector procurement. Further, it is unclear whether or not nongovernmental procurement figures are included in national reporting to our referenced data sources. These uncertainties will correspondingly impact the ceiling of maximum capacity for use.

data (Table 1). As such, the forecast analysis in this paper relies predominately on estimated production data.

RDT procurement forecast

According to our forecast, RDT procurement will increase steadily over the next decade (Figure 1). This prediction is supported by several observable trends. First, an increasing number of countries are adopting RDT, which will consequently increase gross RDT procurement. Moreover, the vast majority of malaria proposals from The Global Fund to Fight AIDS, Tuberculosis and Malaria are increasing budgets for RDT use, often introducing RDT into countries that have previously not used them. In the next ten years, the estimates predict over 460 million RDT will be procured, and in the next three years, over 100 million RDT will be procured. Following the simple trend-line, full market saturation (to the current public sector ceiling) would be achieved within a few years. If it is assumed the ceiling (67 million) remains constant, but that the annual rate of expansion of RDT use reduces in proportion to the remaining available capacity to absorb them, annual RDT procurement would reach 45 million by 2010 and grow to almost 60 million by 2015. Using 2005 production and rate of increase as a baseline, RDT procurement will increase 52% by 2010 and 96% by 2015. By contrast, the simple trend-line projection with no ceiling, predicts a 67% and 150% increase in the same respective years.

ESTIMATES
RDT procurement and production
RDT procurement reported by the available sources was 12 million units in 2005. Since 2000, RDT procurement is observed to have increased rapidly, at nearly a doubling rate from 2000 to 2004. Moreover, since 2000, an increasing number of countries are adopting the use of RDT and have budgeted RDT as part of anti-malaria activities (Table 1). Global RDT production, extrapolated from known HRP2 production, is estimated to be 28 250 000 in 2005. Except for the period between 20022003, RDT production has increased at nearly a doubling rate since 2000 (Table 1). There is therefore a gap between reported use and estimated production of 16 million RDT (57% of estimated production). Ideally, production and procurement numbers should agree, given that RDT are generally only produced when procurement orders are received. However, consistent with a lack of private sector data and incomplete data reporting by, and collection from, procuring bodies, a large discrepancy exists between data on RDT procurement and production

DIScUSSION AND INTERPRETATION

There is clearly a large element of uncertainty in the future volume of RDT use, and this depends heavily on the existence and level of the procurement ceiling in the public health sector (Figure 1). The reducing effect of the RDT market ceiling on the procurement forecast can be reasoned TABLE 1: Country Procurement and Manufacturer Production Data (source: Global by several factors. The need to recruit Fund, antibody supplier data) and train health workers, improve access 2000 2001 2002 2003 2004 2005 to remote areas, and strengthen quality Global RDT control will delay increased procurement of 16 700 98 500 1 110 133 5 068 269 9 330 000 11 588 848 procurement RDT. Further, for countries newly adopting Global RDT RDT, pilot studies used to assess feasibility 2 875 000 4 500 000 7 625 000 8 875 000 18 500 000 28 250 000 production and efficacy are inevitably time consuming number of and will most likely abate increased 1 1 6 18 21 32 countries using RDTs procurement of RDT. While it is probable that procurement growth will be limited in this manner, uncertainties that variably affect
PAGE 2 OF 4

www.wpro.who.int/sites/rdt

Forecasting global procurement of malaria rapid diagnostic tests: estimates and uncertainties

World Health Organization

Western Pacific Region

both the forecast curve and the RDT market ceiling exist and are discussed below. The first uncertainty relates to the availability of microscopy services. The availability of microscopy diminishes the need for RDT. While microscopy services are readily available in parts of Northern Africa, South and central America and South-East Asia, many other endemic regions have room to expand microscopy services. Nevertheless, quality microscopy services are difficult to maintain in remote areas (Durrheim et al. 1997; Kachur et al. 1998; Kain et al. 1998; Kilian et al. 2000; coleman et al. 2002b; OMeara et al. 2005), so it seems likely that the global level of microscopy services will remain relatively stable and impact the RDT market little in the near future. A second major uncertainty deals with acceptance of RDT. Manufacturers have yet to penetrate the entire global market, and RDT remain virtually unused in many endemic countries. This may result in part from concerns over diagnostic accuracy, shipping and handling requirements, and the need to train health workers. In sub-Saharan African nations in particular, concern over blood handling by health workers and HIV risk may further reduce demand at the village level. A third major uncertainty concerns the level of the public sector RDT market ceiling, and future movement of it. clinical malaria incidence is estimated at 350500 million annually (WHO 2005), and the incidence of malaria-like fevers will be far higher (Redd et al. 1992; Peters et al. 2004; Reyburn et al. 2004; carcillo 2005; WHO 2005). The ceiling in Figure 1 is therefore based on an assumed inadequacy of current health services to address the need. Should the ceiling rise, through expansion of health sector capacity or through RDT procurement driven by funding rather than capacity to implement, the public health ceiling (number of fever cases potentially diagnosed with an RDT by the public health system) will rise. A simple projection of current trends (Figure 1) suggests that this ceiling may be breached (is set too low). Finally, the rate of RDT use in the private sector remains relatively obscure, but this sector could have an important impact on our forecast. The private sector is generally represented by private health care facilities, informal and formal dispensaries, and to a lesser extent travellers in malaria-endemic regions. Its
PAGE 3 OF 4

impact will vary widely between countries, and RDT use may not play a major role for some time as costincentives for the provider and the client are not clear. Ultimately, to continue to expand the market, it will be necessary to (1) clearly demonstrate test accuracy by strengthening quality assurance, (2) intensify efforts in training health workers, and (3) improve access and delivery systems. This could be aided by a coordinated procurement and staggered delivery scheme Subdocument B. To achieve further advantages, a reduction in the number of products on the market, currently well over 40, may be necessary (WHO 2006a). It is hoped that a planned WHO product testing programme will allow this to occur through an orderly process based on quality (WHO 2006b).

cONclUSION
While great uncertainty remains in both current RDT procurement and future trends (Figure 1), it seems likely that the present rate of increase will slow as capacity to absorb their use approaches saturation, though this capacity itself is hard to quantify and likely to increase. Much will depend on price and quality. A more orderly system of procurement by countries that allows better prediction of current and future needs, maximizing economies of scale and allowing for improved planning and quality assurance within the manufacturing sector, would appear to offer considerable benefits to malaria management.
ACKNOWLEDGEMENTS The WHO thanks various manufacturers for their assistance, Martine Guillerme and Derryck Klarkowski of Medicines Sans Frontieres, and Thuy Huong Ha of UNICEF. Authors Fred Baik University of California, San Diego School of Medicine David Bell World Health Organization Regional Office for the Western Pacific Manila, Philippines. REFERENcES Belizario, V. Y., C. J. Pasay, et al. (2005). Field evaluation of malaria rapid diagnostic tests for the diagnosis of P. falciparum and non-P. falciparum infections. Southeast Asian J Trop Med Public Health 36(3): 552-61. carcillo, J. A. (2005). Reducing the global burden of sepsis in infants and children: a clinical practice research agenda. Pediatr Crit Care Med 6(3 Suppl): S157-64. coleman, R. E., N. Maneechai, et al. (2002a). Short report: Failure of the OptiMAl rapid malaria test as a tool for the detection of asymptomatic malaria in an area of Thailand endemic for Plasmodium falciparum and P. vivax. Am J Trop Med Hyg 67(6): 563-5. coleman, R. E., N. Maneechai, et al. (2002b). comparison of field and expert laboratory microscopy for active surveillance for asymptomatic Plasmodium

www.wpro.who.int/sites/rdt

World Health Organization

Western Pacific Region

Forecasting global procurement of malaria rapid diagnostic tests: estimates and uncertainties

falciparum and Plasmodium vivax in western Thailand. Am J Trop Med Hyg 67(2): 141-4. Durrheim, D. N., P. J. Becker, et al. (1997). Diagnostic disagreement--the lessons learnt from malaria diagnosis in Mpumalanga. S Afr Med J 87(8): 1016. Fryauff, D. J., Purnomo, et al. (2000). Performance of the OptiMAl assay for detection and identification of malaria infections in asymptomatic residents of Irian Jaya, Indonesia. Am J Trop Med Hyg 63(3-4): 139-45. Huong, N. M., T. M. Davis, et al. (2002). comparison of three antigen detection methods for diagnosis and therapeutic monitoring of malaria: a field study from southern Vietnam. Trop Med Int Health 7(4): 304-8. Iqbal, J., N. Khalid, et al. (2002). comparison of two commercial assays with expert microscopy for confirmation of symptomatically diagnosed malaria. J Clin Microbiol 40(12): 4675-8. Iqbal, J., A. Sher, et al. (2000). Plasmodium falciparum histidine-rich protein 2-based immunocapture diagnostic assay for malaria: cross-reactivity with rheumatoid factors. J Clin Microbiol 38(3): 1184-6. Jorgensen, P., l. chanthap, et al. (2006). Malaria rapid diagnostic tests in tropical climates: The need for a cool chain. American Journal of Tropical Medicine and Hygiene 74(5). Kachur, S. P., E. Nicolas, et al. (1998). Prevalence of malaria parasitemia and accuracy of microscopic diagnosis in Haiti, October 1995. Rev Panam Salud Publica 3(1): 35-9. Kain, K. c., M. A. Harrington, et al. (1998). Imported malaria: prospective analysis of problems in diagnosis and management. Clin Infect Dis 27(1): 142-9. Kilian, A. H., W. G. Metzger, et al. (2000). Reliability of malaria microscopy in epidemiological studies: results of quality control. Trop Med Int Health 5(1): 3-8. Kolaczinski, J., N. Mohammed, et al. (2004). comparison of the OptiMAl rapid antigen test with field microscopy for the detection of Plasmodium vivax and P. falciparum: considerations for the application of the rapid test in Afghanistan. Ann Trop Med Parasitol 98(1): 15-20. Mason, D. P., F. Kawamoto, et al. (2002). A comparison of two rapid field immunochromatographic tests to expert microscopy in the diagnosis of malaria. Acta Trop 82(1): 51-9. Mishra, B., J. c. Samantaray, et al. (1999). Study of false positivity of two rapid antigen detection tests for diagnosis of Plasmodium falciparum malaria. J Clin

Microbiol 37(4): 1233. Moody, A. H. and P. l. chiodini (2002). Non-microscopic method for malaria diagnosis using OptiMAl IT, a second-generation dipstick for malaria plDH antigen detection. Br J Biomed Sci 59(4): 228-31. OMeara, W. P., F. E. McKenzie, et al. (2005). Sources of variability in determining malaria parasite density by microscopy. Am J Trop Med Hyg 73(3): 593-8. Peter l chiodini, Katherine Bowers, et al. The heat stability of plasmodium lactate dehtdrogenase-based and jistidine-rich protein 2-based malaria rapid diagnsotic tests. Trans R Soc Trop Med & Hyg In Press. Peters, R. P., E. E. Zijlstra, et al. (2004). A prospective study of bloodstream infections as cause of fever in Malawi: clinical predictors and implications for management. Trop Med Int Health 9(8): 928-34. Redd, S. c., P. B. Bloland, et al. (1992). Usefulness of clinical case-definitions in guiding therapy for African children with malaria or pneumonia. Lancet 340(8828): 1140-3. Reyburn, H., R. Mbatia, et al. (2004). Overdiagnosis of malaria in patients with severe febrile illness in Tanzania: a prospective study. Bmj 329(7476): 1212. Ricci, l., I. Viani, et al. (2000). Evaluation of OptiMAl Assay test to detect imported malaria in Italy. New Microbiol 23(4): 391-8. Richter, J., K. Gobels, et al. (2004). co-reactivity of plasmodial histidine-rich protein 2 and aldolase on a combined immuno-chromographic-malaria dipstick (IcT) as a potential semi-quantitative marker of high Plasmodium falciparum parasitaemia. Parasitol Res 94(5): 384-5. WHO (2004). The Use of Malaria Rapid Diagnostic Tests. Manila, World Health Organization - Regional Office for the Western Pacific. WHO (2005). World malaria report 2005. Geneva, World Health Organization and UNIcEF. WHO (2006a). Malaria Rapid Diagnostic Tests: Making Rapid Diagnosis Work, World Health Organization - Regional Office for the Western Pacific. 2006. WHO (2006b). Towards quality testing of malaria rapid diagnostic tests: Evidence and methods. Manila, World Health Organization. WHO (2006c). World Health Report 2006: Working together for health. Geneva, World Health Organization.

www.wpro.who.int/sites/rdt

PAGE 4 OF 4

Forecasting global procurement of malaria rapid diagnostic tests: estimates and uncertainties

World Health Organization

Western Pacific Region

< Return to main document

Subdocument A

Methods
Estimating global RDt pRocuREmEnt
Data on RDt procurement was extracted from individual country proposals and performance reports to the global Fund to fight aiDs, tuberculosis and malaria; the WHo World malaria Report (WHo 2005); procurement information provided by unicEF and mdecins sans Frontires (msF); and the WHo global atlas Data Query (http://www.who.int/globalatlas). global procurement was aggregated from these sources and includes data from 40 countries; not all countries reported data. to clarify, procurement is defined as RDt ordered, purchased, and obtained by each country from a manufacturer, or purchased by a nongovernmental organization. in contrast, production will refer to RDt produced by manufacturers. health services. thus, the public sector RDt market was estimated by subtracting total cases diagnosed by microscopy from total fever cases, considered as probable malaria in case reporting. Estimates of total malaria-like fever cases were based on data reported by public health services (WHo 2005). this figure underestimates the true incidence of fever. it is estimated that 350500 global malaria cases occur annually, but the incidence of fever cases, many of which are not malaria, will be far greater. However, the vast majority of such cases are not seen by public health services and, consequently, are not likely candidates for parasite-based diagnosis. therefore, for the purpose of these estimates, the number of fever cases impacting RDt use is restricted to those reported by public major health services. as RDt use requires considerable logistics and training, it can be assumed that where case reporting of fever fails, the ability to introduce and maintain RDt use is also absent. as no global data for the number of fever cases exists, the prevalence of fever seen by organized public sector services was inferred from data in the World malaria Report (WHo, 2005) on the number of probable malaria cases, reported malaria cases, or slides taken, whichever was greater. total cases diagnosed by microscopy was determined by the number of slides taken or microscopy-confirmed cases in the World malaria Report, whichever was greater. RDt procurement outside public health services was estimated using procurement data obtained from the two organizations considered the major procurers in this sector; mdicines sans Frontires (msF) and unicEF. procurement data from 2003 was used, noting that the most recent data reported in the World malaria Report came from that year. it is assumed that total numbers used by other nongovernmental organizations and the private sector market is relatively small.
REFEREncE WHo (2005). World malaria report 2005. geneva, World Health organization and unicEF.

Estimating RDt pRoDuction

Data on production of RDt was based on antibody supplies for the histidine-rich protein-2 (HRp2) RDt, which constitute the vast majority of RDt currently produced. annual HRp2 RDt production was calculated by first obtaining the total amount of anti-HRp2 antibody supplied to manufacturers (data from a supplier to manufacturers of most currently available HRp2 RDt products. this number was then divided by 0.7 (micrograms), the average amount of antibody used per RDt (information provided by manufacturers). total annual RDt production (HRp2 plus plDH RDts) was calculated assuming that HRp2 RDt represent 80% of the RDt market, estimated from procurement patterns known to WHo.

Estimating total RDt maRkEt

the potential global RDt market consists predominantly of potential procurement by: (1) national public health services and (2) nongovernmental organizations. the public sector RDt market was estimated by the number of fever cases, or potential/probable malaria cases, which were reported but not diagnosed by microscopy in public
PAGE 1 OF 1

www.wpro.who.int/sites/rdt

Forecasting global procurement of malaria rapid diagnostic tests: estimates and uncertainties

World Health Organization

Western Pacific Region

< Return to main document

Subdocument B

Coordination of Malaria Rapid Diagnostic Test Purchasing

Background
Procurement of large volumes of malaria rapid diagnostic test (rdt) by countries and nongovernmental organizations are likely to increase over the next few years. requests for procurement tend to have short time lines for delivery, requiring manufacturers to produce large batches of rdt at short notice. rdt may then be stored for prolonged periods before the whole procured batch is utilized, increasing the risk of deterioration due to heat and humidity. although over 40 rdt products are listed on the market, many are produced at single sites under re-labelling and licensing agreements. some component producers, in turn, produce for a number of manufacturing sites. this improves the potential for combining procurement. some components (e.g. monoclonal antibodies) are produced on demand only, requiring production at short notice when orders come in from rdt manufacturers. Prices of some monoclonal antibodies produced under such circumstances are highly dependent on the quantity procured, and 40% to 50% cost reductions can apply to larger purchases. the requirement for production and delivery of monoclonal antibodies and rapid diagnostic tests at short notice leads to higher costs by preventing manufacturers from planning production runs efficiently. this reduces the time available for components and rdt manufacturers to maximize test quality for each production run and to perform, and react to the results of, quality control testing during manufacture. Longer lead times for delivery of RDT would allow more time for manufacturers to ensure product quality. Staggered delivery of individual purchases would reduce the required storage time between production and product use, minimizing the risk of deterioration. Staggered delivery and planned delivery times would allow manufacturers to better forecast production requirements, resulting in potential cost savings, and may produce a better environment for manufacturers to invest in product research and development. Disadvantages Staggered delivery may increase transport costs. After placing an order, countries would have to wait a longer to receive RDT.

Potential mechanisms to achieve cost savings and imProve quality

Fixed purchase times orders for procurement, made through a coordinating body, could be placed with the manufacturer with fixed delivery dates of, for example, four times per year (Figure 1). orders received in the interim are accumulated during this time, allowing manufacturers to give greater lead time to component producers who, along with rdt producers, could plan production runs with adequate warning of the capacity required. in the example in Figure 1, purchase orders of rdts due for delivery in december would be finalized by the start of october, with orders submitted during the previous three months (July to september) having accumulated during that time. this leaves two months available for final production and assembly of rdt. as delivery is staggered (e.g. a 12 month requirement of rdts is delivered at two 6 month intervals), manufacturers would have up to 8 months lead time to plan for a large proportion of this rdt order.
www.wpro.who.int/sites/rdt

coordinated Procurement and staggered delivery

Advantages Coordinated purchasing of malaria RDT would allow larger batches of components to be produced, reducing component costs.
PAGE 1 OF 2

World Health Organization

Western Pacific Region

Forecasting global procurement of malaria rapid diagnostic tests: estimates and uncertainties

< Return to main document

Bulk procurement facility a coordinating body could bulk-purchase accumulated orders from several countries, accumulated over 23 months, in a similar method to (1) above. Limiting available products. current Who policy to limit malaria rdt procurement to manufacturers able to demonstrate it is assumed that total numbers used by other nongovernmental organizations and the private sector market is relatively small. evidence of good manufacturing practice, future product testing and pre-qualification of rdt, is likely to facilitate either of the above schemes by reducing the number of products involved.

FIGURE 1: Example of quarterly joint procurement through a coordinating body, with six monthly staggered delivery of product

Jan
Submit Final order combined placed with order* manufacturer Scheduled delivery Country 1 Country 2 Country 3 Country 4 Country 5

Feb

Mar

Apr
Final order placed with manufacturer

May

Jun

Jul
Final order placed with manufacturer

Aug

Sep

Oct
Final order placed with manufacturer

Nov

Dec

Jan

1
Request to Coordinator Request to Coordinator Request to Coordinator

2
Delivery of half of order Delivery of half Request 2 to Coordinator

4
Delivery of half x 2 Delivery of half

Delivery of half Request to Coordinator Request to Coordinator Delivery of half Delivery of half

This example allows a minimum two month lead time for manufacturers to submit orders to a components manufacturer, receive monoclonal antibodies (MAbs), manufacture and complete quality assurance. A MAb manufacturer can predict most orders due that month, will have increased lead time for manufacture, and thus can provide MAbs at lower cost to the end-product manufacturer. Orders are staggered, allowing good prediction of total order for the next 6 to 12 months.

www.wpro.who.int/sites/rdt

PAGE 2 OF 2