Recognising the tremendous burden of bacterial sexually transmitted infections (STIs) and the lack of access to laboratory diagnostic services in the developing world, in 1993 the Rockefeller Foundation posted a prize of US$1 million for the development of an inexpensive rapid simple test to detect genital chlamydial or gonococcal infection. The prize was never claimed and was withdrawn in 1998. In stark contrast, ultra sensitive and specific assays such as nucleic acid amplified tests were developed and became widely and effectively used in the developed world to screen for these infections and facilitate treatment. Not surprisingly, the prevalences of curable STIs remain very high in much of the developing world, where women and children continue to bear a disproportionate share of the disease burden. An estimated one million new cases of curable bacterial STIs occur each day worldwide¹ and in sub-Saharan Africa alone, half a million babies die each year from congenital syphilis.²

Diagnostic testing is particularly critical for STIs, including HIV, as most infected individuals are often asymptomatic but, nonetheless, infectious. Sensitive diagnostic tests are required for early detection to guide treatment, to prevent the development of reproductive sequelae, and to interrupt onward transmission.

Reasons for the failure to prevent and control STIs in the developing world are complex, but a major barrier to effective health care in Africa is the lack of access to reliable, affordable, accurate laboratory services.^3,4 The 2004 World Development Report cited lack of access and unaffordability as two major reasons why services fail.⁵ Table 1 shows the mean distance to the nearest medical facility in selected countries.⁶ Even if pregnant women manage to walk the long distance to the clinic, they are often unable to return for their test results and treatment.

The Sexually Transmitted Diseases Diagnostics Initiative (SDI) conducted an expert consultation in 2001 to determine the next steps for improving STI diagnostics for the developing world; priority was given to the evaluation of simple, affordable, commercially available point-of-care tests for the major curable bacterial STIs: syphilis, chlamydia and gonorrhoea. The tests should meet the ASSURED criteria⁶ (table 2).

For syphilis, the evaluation of diagnostics that would increase access to prenatal screening was selected. For genital chlamydial and gonococcal infection, priority was given to the evaluation of diagnostics that can be used for screening of high risk populations as well as for increasing the specificity of syndromic management.

This review provides the background for the work on STI diagnostics featured in this supplement, and proposes 10 specific strategies or pathways for the way forward.

LACK OF ACCESS TO PRENATAL SYPHILIS SCREENING

In a survey conducted in 2000, Gloyd reported that in sub-Saharan Africa, antenatal screening was a national policy in 17 of 22 respondent countries.⁷ Of the estimated 19.9 million pregnant women per year in these countries, 14.5 million (73%) received antenatal care but only 5.5 million (38%) had access to syphilis screening. If this survey is representative of sub-Saharan Africa, with an estimated mean syphilis seroprevalence of 8.3%, an estimated 2.3 million pregnant women each year in this region would harbour Treponema pallidum and potentially could transmit the infection to their babies, and 1.64 million of these would either not be diagnosed or not treated. As a result, in sub-Saharan Africa alone, half a million babies die of syphilis each year.^2,7

The consequences of syphilis in pregnancy are so severe that universal screening of pregnant women for syphilis is national policy in most countries. Prenatal syphilis screening remains cost beneficial at prevalences as low as 5 per 100 000.^8,9,10 Antenatal screening and treatment for syphilis are among the most cost-effective health interventions available (table 3).

Simple rapid tests such as the rapid plasma reagin (RPR) or venereal diseases research laboratory (VDRL) tests are used for prenatal screening.¹³ Unfortunately, these are often not available at primary health care settings in developing countries because they require a source of electricity to refrigerate the RPR or VDRL reagent, operate a centrifuge to separate sera from whole blood and power a rotator to mix the reaction. The requirement for cold storage also limits the volume of reagents and tests that can be procured by a central facility at any one time, leading to frequent stock-outs of the tests at district level hospitals and health care centres. Women must return to receive test results as the testing is batched to save time and reagents, or specimens may be sent to a central laboratory for processing. In a consultation conducted by the SDI on strategies to increase access to prenatal screening in Tanzania, a number of barriers to RPR testing were identified (table 4)

Despite recent successes in increasing coverage for antenatal services in developing countries, most pregnant women fail to get screened for syphilis.¹⁴ In Bolivia in 1996, 76% of women had access to prenatal care but only 16% were screened for syphilis.¹⁵ In the 1990s, demonstration projects showed that significant reductions in the prevalence of maternal and congenital syphilis can be achieved with decentralisation of antenatal screening.^16–19 In Haiti, the introduction of solar-powered generators to operate centrifuges and rotators, propane powered refrigerators to store the RPR reagent, and the training of field workers to perform RPR tests in rural health centres, resulted in a 75% decline in the incidence of congenital syphilis over two years.²⁰ However, these demonstration projects on decentralisation of screening services have not led to sustainable expanded coverage of antenatal screening for syphilis.

In this supplement, nine rapid syphilis tests that approach the ASSURED criteria in terms of operational characteristics were evaluated with well-characterised archived specimens from laboratories in diverse geographical settings.²¹ Tests with reasonable performance characteristics which can be performed on whole blood were evaluated in field settings in Bolivia, Brazil, China, Haiti, Peru and Tanzania.^22–24 Mathematical models developed with data from the field site in Mwanza, Tanzania showed that the use of these rapid tests is cost-effective, and has the potential to make a substantial impact in averting congenital syphilis in primary health care settings.²⁵

LACK OF DIAGNOSTIC TESTS FOR THE DETECTION OF GENITAL CHLAMYDIAL AND GONOCOCCAL INFECTIONS

The World Health Organization (WHO) has promoted syndromic management of STIs in settings where access to laboratory services or resources is limited.²⁶ Patients with syndromes such as urethral or vaginal discharge, genital ulcers or lower abdominal pain are treated for all the likely causes of that syndrome.

The syndromic approach works particularly well for men with urethral discharge and genital ulcers, but evaluations showed that the algorithm for vaginal discharge lacks both sensitivity and specificity for the identification of women with Chlamydia trachomatis and Neisseria gonorrhoeae infection.^27–29 Most women (usually over 90%) with this syndrome do not have these cervical infections, but many suffer from vaginal infections (especially bacterial vaginosis or trichomoniasis).³⁰

Since facilities for culture, antigen detection or nucleic acid amplification tests are not available outside of major urban centres, it is not possible to diagnose gonorrhoea or C trachomatis infection in women at peripheral health services in most developing countries. Where microscopy is available, Gram stained smear of an endocervical swab can be used to diagnose gonorrhoea, but this technique is at best only about 50% sensitive in women and is also non-specific. Hence the syndromic approach is widely promoted and used. Although simple to use and inexpensive, the poor performance of the syndromic approach to the management of cervical infection is well documented and often can lead to problems in unnecessary treatment and difficulties with partner notification.^27–29 Unnecessary treatment results in increased drug costs, possible side effect of multiple drugs, changes in endogenous flora and the potential for increased drug resistance. Women may be at risk of stigma and violence from their sexual partners if they are asked to refer their partners for STI treatment when they do not have an STI.

There is a vast need for simple, affordable point-of-care tests for these infections in women. Articles in this supplement describe the results of evaluations of two rapid tests for gonorrhoea and one for chlamydia.^31–33 Although these tests have reasonable specificity, their sensitivity could be improved.

LACK OF REGULATORY CONTROL OF THE QUALITY OF STI DIAGNOSTICS

Clinical care for STIs in the developing world is compromised if regulatory control of diagnostics in both the public and private sector is not done effectively. Many countries do not have sufficient infrastructure to determine the quality of diagnostic tests. A CE mark on a product only addresses manufacturing quality and does not imply acceptable test performance. WHO/TDR conducted a global survey of diagnostics regulatory practices in 2001. A questionnaire sent to 191 WHO member states inquired about regulation of in vitro diagnostics (other than those used for blood banking), and if clinical trials were required for regulatory approval.³⁴ Of the 85 countries that responded, 48% regulate in vitro diagnostics for infectious diseases. Of the 24 countries that provided data on which diagnostics were regulated, 83% regulate diagnostics for HIV, 92% for hepatitis, 42% for STIs and 13% each for tuberculosis and malaria.

Of the countries that regulate diagnostics, 68% require the submission of clinical trial data. However, no standard exists for diagnostic evaluation or approval in developing countries. An industry survey conducted by WHO/TDR in 2003 showed that company spending on evaluation of a product varied from US$2000 to more than $1 million. Diagnostic trials had been conducted in as few as 15 patients (unpublished TDR data). A recent dossier received at the WHO SDI on a syphilis rapid test contained a claim of 100% sensitivity and 100% specificity. The evaluation was conducted in just over 100 patients, of whom only three were positive for syphilis!

Physicians often select tests based on information provided in the product insert or from published data. Information on the performance of tests sold in developing countries is often biased because of flaws in the design and conduct of evaluations.³⁵ Hence diagnostics are often used without proof of their effectiveness. The consequences of low quality diagnostics for STIs include wasted precious resources and mismanagement of patients, potentially leading to serious sequelae in the infected individual and failure to interrupt the chain of transmission in the population.

THE WAY FORWARD

Rapid tests for HIV infection have shown the potential for such tests to be rapidly and widely introduced in both developed and developing countries. Having been evaluated by WHO and shown to be both sensitive and specific, they became available at affordable prices through the WHO bulk procurement programme, and were approved for use in the USA by the Food and Drug Administration. The demand for an immediate result, and the advantages in terms of patient management, are so great that rapid tests are now routinely used in voluntary counselling and testing centres and antenatal clinics worldwide. This will hopefully be similar for rapid STI tests that fulfill the ASSURED criteria. However, there are 10 areas where more work is needed before STI tests can be widely used.

1. Continued search for ASSURED tests

The ASSURED score card based on select STI tests evaluated in this supplement is shown in table 5.^{21–24,31–33}

Mathematical models developed at the Tanzania site for prenatal syphilis screening showed that screening with rapid tests is cost-effective if the price per test does not exceed US$0.63.²⁵ Currently eight rapid syphilis tests evaluated by the SDI are included in the WHO Bulk Procurement Scheme at negotiated pricing ranging from US$0.19–1.00. Improved tests for syphilis that can distinguish between active and past treated infection are still needed. The US Centers for Disease Control and Prevention has been assisting test developers and manufacturers to develop rapid tests that will combine non-treponemal and treponemal test on a solid phase, in either a flow-through or immunochromatographic strip format. These prototypes will soon be ready for evaluation.

Clearly there is a need for more sensitive, simpler and cheaper rapid tests for C trachomatis and N gonorrhoeae.^36,37 Gift et al have shown that a rapid test with a sensitivity of 65% can lead to a greater proportion of infected patients treated compared to a nucleic acid amplified test with a sensitivity of 90%, when the return rate for test results and treatment is low.³⁸ Mathematical models developed using data from the SDI site in Benin showed that rapid tests with sensitivity of 70% and specificity of 98% have the potential to reduce the prevalence of these infections in sex workers and the potential to avert HIV infections in their clients.³⁹ Encouraging data on an improved rapid test for ocular chlamydial infection have recently been published,⁴⁰ and a number of tests for gonorrhoea await evaluation by SDI.

2. Piloting a road map for test introduction

Too often, research findings are not effectively translated into control strategies, policies and programmes.⁴¹ There is a gap between demonstrating the effectiveness of a novel intervention in a demonstration project and its adoption to scale in countries. In preparation for introduction of rapid tests into settings where no previous testing was possible, SDI conducted consultations in several regions of the world to define the key elements required for the successful and sustainable introduction of rapid syphilis tests (table 6). These activities form a basic road map for the introduction of new tools that can be used for diagnostics, vaccines and medicines.

As a result of the SDI regional consultations, 12 countries have initiated the development of a road map for the introduction of syphilis screening in their country. Such a road map can be broadly used for the introduction of other technologies for a variety of infectious diseases.

3. Developing sustainable diagnostic quality assurance programmes

Increasing access to testing in primary care settings means that tests will be stored at temperatures in excess of 30°C and no longer be performed by trained laboratory technicians. It is therefore critical that countries set up external quality assurance (EQA) schemes to ensure that the quality of the tests has not been compromised during storage and that health workers are proficient at performing the tests. These national schemes can be developed in conjunction with quality assurance schemes for other diagnostic testing and should be supported technically by a network of expert laboratories, inside and outside of the country. This requires the development of a business plan for obtaining sufficient funds to cover set up costs and identify means of sustaining the programme.

4. Taking advantage of available programme infrastructure and resources

Capacity for introduction of new tools varies significantly between countries, and between rural and urban areas within countries. The success of increasing access to syphilis screening depends on the level of political commitment, the existence of a robust health infrastructure to facilitate the implementation, and the availability of screening tools. In resource limited settings, health service providers have to prioritise their scarce monetary and human resources to accommodate many competing demands. There are also operational and administrative difficulties such as frequent stock-outs of tests or drugs for treatment, leading to poor motivation in implementing and scaling-up screening services. Instead of setting up a new infrastructure for decentralised STI diagnostic services, every effort should be made to take advantage of new or existing initiatives, such as antenatal care programmes, Prevention of Mother to Child Transmission programmes for HIV, and HIV Voluntary Counselling and Testing programmes.

5. Using the internet more creatively

In recent years, the internet has emerged as a risk environment for STIs and many STI testing services are available from the internet.^42–45 The sensitivity of nucleic acid amplification tests has made it possible to use non-invasive specimens, such as urine or vaginal swabs, for the diagnosis of bacterial STIs. As these specimens can be stored and transported at ambient temperatures, there has been a proliferation of internet-based STI testing services. The type of tests and the quality of testing used on these internet services are unknown. Simple rapid STI tests are sold on the internet or “over the counter”. Few countries regulate STI tests sold through these venues. STI control programmes need to exploit the popularity of the internet to provide scientific information to health care professionals and the lay public on how STIs should be diagnosed and treated, and make the public aware of the consequences of misdiagnosis.

6. Ensuring a sustained effort at disease control programmes

The availability of new tools that can be used to increase coverage for prenatal screening has sparked interest in planning for the global elimination of congenital syphilis. While “eradication” or “elimination” are laudable goals, eradication has been achieved for smallpox, and is being pursued for river blindness and polio, with local success in some regions. Another treponematosis, yaws, has been eliminated from some regions, but not globally. Elimination of syphilis is being pursued in the USA, and has been achieved transiently in some areas of the world, such as China and King County, Washington, USA. However, it returned with a vengeance as sexual behaviours and patterns of sexual mixing changed.^46,47 Syphilis outbreaks elsewhere in the USA, eastern Europe and the Russian Federation are also testament to the resilience of the pathogen.^48,49 In general, the concept of elimination of syphilis, given available tools, is unrealistic and diverts attention for the need for continuous and sustained control programmes. The addition of ASSURED rapid diagnostics for syphilis, as one component of a sustained effort, can improve the potential for better control of adverse consequences of syphilis globally regardless of ongoing endemic disease.

7. Defining the role of rapid diagnostics for viral STIs

Recent studies have identified herpes simplex virus 2 (HSV-2) as a major risk factor for the acquisition and transmission of HIV.⁵⁰ A vaccine for human papilloma virus (HPV) has been shown to be effective in preventing the infection by the major oncogenic types.⁵¹ Although the role of diagnostics in the control of many viral STIs remains unclear, more strategic planning needs to be given to the role of diagnostics in the control and prevention of HSV.

8. Assessing the need for rapid tests for vaginal discharge

The syndromic management of vaginal discharge remains problematic. Rapid tests could improve the specificity of syndromic management. The performance, operational characteristics and costs of rapid tests for C trachomatis, N gonorrhoeae, Trichomonas vaginalis and bacterial vaginosis are now available. In the absence of a multiplex test, the cost of using multiple rapid tests to detect all the major causes of this syndrome may be prohibitive, and certainly not cost-effective compared to presumptive treatment. A consensus meeting to review and recommend how these tests should be used in the management of vaginal discharge is warranted.

9. Developing a consensus on how rapid tests for C trachomatis and N gonorrhoeae should be used

The performance and operational characteristics of rapid tests for the screening and case management of individuals with genital chlamydial and gonococcal infections need to be better defined. Given the psychosocial consequences of a misdiagnosis for STIs, tests should be highly specific. From the perspective of STI control programmes, rapid tests can improve control of STIs in high risk populations. In the general population, where the prevalence may be low, even a test with a specificity of 98% may yield more false than true positives. Once rapid tests become commercially available, it would be difficult to prevent their widespread use. Guidelines are needed on how rapid tests should be used and updated as test performance improves.

10. Inclusion of STI diagnostics as part of an essential diagnostics package for improving health in developing countries

The control of infectious diseases of public health importance in the developing world is often hampered by lack of resources. WHO has developed an Essential Medicines package with the aims of ensuring that public health programmes in developing countries have access to medicines that are deemed essential for a sound public health programme. As it is increasingly clear that the lack of affordable appropriate diagnostics is a major barrier to health in the developing world, an essential diagnostics package that includes rapid tests for HIV, tuberculosis, malaria and STIs should be developed.

CONCLUSIONS

At the Millennium Summit in 2000, representatives from 189 countries committed to the attainment of a set of millennium development goals (MDGs). Goal number 4 calls for the reduction of mortality under 5 years of age (U5MR) by 75%; goal number 5 calls for improvements in maternal health; and goal number 6 calls for halting and reversing the spread of AIDS by 2015. Now nearly half way towards the deadline, most countries are still struggling to sustain a course of action that will achieve the desired results.

In the second global project on disease control priorities in developing countries (DCP2), released in 2006 after consultations with over 600 experts worldwide, two concerns are:

• Half of all deaths (including stillbirths) of children under age 5 years occur at ages under 28 days, but little attention has been paid to this age group. Cost-effective interventions exist.

• Treatment of HIV positive mothers, treatment of STIs, free distribution of condoms, and other interventions can cost-effectively interrupt HIV transmission. These preventive interventions continue to receive inadequate attention.

Almost certainly if we are to achieve the MDGs of reducing mortality under the age of 5 years by 75% by 2015, comprehensive clinic based screening and treatment programmes using rapid tests will need to be made routine in prenatal care. It is hoped that increased access to quality-assured rapid STI tests will go a long way towards addressing DCP2 recommendations, and help countries to achieve their MDGs.