## Abstract

Stakeholders agree that supporting high-quality diagnostics is essential if we are to continue to make strides in the fight against human immunodeficiency virus (HIV) and tuberculosis. Despite the need to strengthen existing laboratory infrastructure, which includes expanding and developing new laboratories, there are clear diagnostic needs where conventional laboratory support is insufficient. Regarding HIV, rapid point-of-care (POC) testing for initial HIV diagnosis has been successful, but several needs remain. For tuberculosis, several new diagnostic tests have recently been endorsed by the World Health Organization, but a POC test remains elusive. Human immunodeficiency virus and tuberculosis are coendemic in many high prevalence locations, making parallel diagnosis of these conditions an important consideration. Despite its clear advantages, POC testing has important limitations, and laboratory-based testing will continue to be an important component of future diagnostic networks. Ideally, a strategic deployment plan should be used to define where and how POC technologies can be most efficiently and cost effectively integrated into diagnostic algorithms and existing test networks prior to widespread scale-up. In this fashion, the global community can best harness the tremendous capacity of novel diagnostics in fighting these 2 scourges.

On 28 June 2011, the National Institute of Allergy and Infectious Diseases, together with the Office of the Global AIDS Coordinator, the Centers for Disease Control and Prevention, and the Eunice Kennedy Shriver National Institute of Child Health and Human Development, sponsored a workshop on “TB and HIV Diagnostics in Adult and Pediatric Populations” in Silver Spring, Maryland. One of the 3 concurrent tracks within the workshop concentrated on the “Strategies for Implementing Molecular Point-of-Care Diagnostics for HIV and TB.” It brought diagnostic developers together with in-country representatives, nongovernmental organizations (NGOs), and other partner organizations that may purchase and/or use molecular point-of-care (POC) diagnostic devices. Over the next few years, several groups plan to develop easy-to-use, cost-effective diagnostic assays for human immunodeficiency virus (HIV) and tuberculosis that can detect these pathogens at or near to the POC. Appropriate and accelerated adoption of these technologies requires concerted assay evaluation and discussion of challenges beyond the technical and reagent cost restrictions to identify potential barriers of adoption.

The meeting provided developers and stakeholders an opportunity to discuss key issues with purchasers (representatives from country health ministries, NGOs, and US government agencies) regarding the barriers for implementing diagnostics, regulatory concerns, how purchasers are leveraging resources, the current push for global laboratory strengthening, and linkages with existing treatment programs. Other logistical issues, such as supply chain management, communication, workflow, quality assurance, and integration into the existing infrastructure, were also discussed. This paper, as well as others in this supplement (see Palamountain et al, this issue), outlines and discusses different perspectives, controversies, and challenges and provides lessons learned in implementing HIV and tuberculosis diagnostics.

Because the audience included a broad background and spectrum of expertise, a consensus definition of POC was needed for consistency and to avoid confusion. Borrowing from ISO 15189 and other regulatory documents, a working definition of POC for our purpose was “a diagnostic test that is performed near the patient or treatment facility, has a fast turnaround time, and may lead to a change in patient management.” Moreover, it was felt that the test should not require trained laboratory personnel or clinical laboratory (centrifuge, pipettes, etc) or other infrastructural (power, air conditioning, refrigeration, running water, etc) support. It was accepted, however, that there will need to be some level of training, operator interpretation, maintenance, quality control, and assurance to ensure that the test is accurate and reproducible [1].

An additional point of debate is the definition of “impact” regarding the diagnostic device. Most diagnostic developers equate impact with test accuracy. However, many health officials and purchasers look beyond accuracy to determine if the inclusion of a diagnostic in a particular setting has a positive effect on epidemiological (eg, incidence) or patient-centered (eg, morbidity, mortality) outcomes (see Cobelens et al, this issue). This requires follow-up to identify infections as well as appropriate treatment options that would benefit the population where the test is performed. Although less of a concern for HIV, this is more difficult for tuberculosis diagnostics owing to the need for culture-based, phenotypic drug ssusceptibility testing resulting from the lack of defined molecular markers for certain drugs (see McNerney et al, this issue).

Point-of-care diagnostics are inherently attractive in many resource-limited settings where the healthcare, transportation, and distribution infrastructure is underdeveloped and underfunded (Table 1). The main advantage of a POC diagnostic is the ability to diagnose disease without the support of a laboratory infrastructure; this increases access, removes the need for sample transport, and shortens turnaround times from weeks (or months) to hours. As a result, more patients are effectively diagnosed, enabling faster and more complete treatment. Despite these advantages, there are barriers for implementation of POC tests. We outline these challenges here in hopes of initiating a dialogue between stakeholders that will lead to a clearer path forward through compromise, innovation, and shared experience.

Table 1.

Current Issues and Deficiencies of HIV and Tuberculosis Diagnostics/Laboratory Services in Resource-Poor Countries

 Disease Endorsement Assay Examples Advantages Disadvantages Tuberculosis WHO recommended Microscopy Light, fluorescent LED, sputum filtration, bleach microscopy Ease of use, low cost, infrastructure present in many countries, most widely used test worldwide Low sensitivity especially for children and HIV-infected patients, no information on drug resistance or viability, technician dependent Culture MGIT, MODS, standard solid media, colorimetric, nitrite reductase, TLA Highly sensitive commercial and noncommercial assays are the gold-standard tests Noncommercial assays not standardized, require training and biosafety disposal systems, culture contamination issues, suited to larger laboratories, long time to acquire results Chest x-ray Conventional vs portable devices Able to confirm radiologic findings consistent with tuberculosis infection, portable devices help reach some hard-to-reach populations X-ray film quality affects results, low sensitivity/specificity, lack of availability, operator dependence, cost Nucleic acid LPA, Xpert MTB/RIF High analytical sensitivity, rapid turnaround time, standardized, appears to be substantially better than smear microscopy Expensive, require electricity, continued need for DST beyond INH/RIF Not recommended by WHO Antibody Several are commercially available Ease of use, low cost Low sensitivity/specificity, high FP rate, high variability and heterogeneity between individuals, low sensitivity for HIV-positive patients Recommended for LTBI In vivo IGRA, TST Relatively low cost, TST is easy to perform, IGRA requires laboratory support High FP/FN rate in high-burden settings, unable to differentiate LTBI from active tuberculosis Research use only Antigen Urine LAM May be useful if targeted to high-tuberculosis-burden HIV-infected patients with low CD4 counts, rapid nonsputum-based assay Low sensitivity in most patients with active tuberculosis and in patients with high CD4 counts, limited use, unproven performance requiring further evaluation POC None currently available Diagnose active tuberculosis in all populations, all tuberculosis-infected (including LTBI), treatment monitoring assays, predict LTBI progression to disease Likely will need to compromise in other parameters (cost, speed, accuracy), new biomarkers needed to address clinical questions Other LAMP, VOC, phenotypic tests Potential to develop a low-cost assay is of interest Unproven technology still needs validation HIV May have regulatory approval Antibody EIA, WB, IFA, POC Cheap, easy to use and widely adopted for both screening and confirmatory testing, cost effective and POC permitting community-based testing Misses window phase infections (RNA+Ab–), not useful for EID, WB indeterminant issues, high FP rates in HIV-vaccinated volunteers Antibody/Antigen (4th generation) EIA, POC Reduces window period by capturing individuals who are seronegative but anigenemic, cost effective Misses window phase infections (RNA+Ag–), not useful for EID, high FP rates in HIV-vaccinated volunteers Nucleic acid Commercial kits, in-house assays High sensitivity/specificity, able to monitor treatment High cost, requirement for laboratory facilities and skilled staff Not currently available Nucleic acid POC Useful for acute infection, EID, drug resistance monitoring, and vaccine testing Likely need to compromise in other parameters (cost, speed, accuracy) Ultra-sensitive p24 EIA, POC Best used for EID, potential for identifying most acute infections Likely need to compromise in other parameters (cost, speed, accuracy)
 Disease Endorsement Assay Examples Advantages Disadvantages Tuberculosis WHO recommended Microscopy Light, fluorescent LED, sputum filtration, bleach microscopy Ease of use, low cost, infrastructure present in many countries, most widely used test worldwide Low sensitivity especially for children and HIV-infected patients, no information on drug resistance or viability, technician dependent Culture MGIT, MODS, standard solid media, colorimetric, nitrite reductase, TLA Highly sensitive commercial and noncommercial assays are the gold-standard tests Noncommercial assays not standardized, require training and biosafety disposal systems, culture contamination issues, suited to larger laboratories, long time to acquire results Chest x-ray Conventional vs portable devices Able to confirm radiologic findings consistent with tuberculosis infection, portable devices help reach some hard-to-reach populations X-ray film quality affects results, low sensitivity/specificity, lack of availability, operator dependence, cost Nucleic acid LPA, Xpert MTB/RIF High analytical sensitivity, rapid turnaround time, standardized, appears to be substantially better than smear microscopy Expensive, require electricity, continued need for DST beyond INH/RIF Not recommended by WHO Antibody Several are commercially available Ease of use, low cost Low sensitivity/specificity, high FP rate, high variability and heterogeneity between individuals, low sensitivity for HIV-positive patients Recommended for LTBI In vivo IGRA, TST Relatively low cost, TST is easy to perform, IGRA requires laboratory support High FP/FN rate in high-burden settings, unable to differentiate LTBI from active tuberculosis Research use only Antigen Urine LAM May be useful if targeted to high-tuberculosis-burden HIV-infected patients with low CD4 counts, rapid nonsputum-based assay Low sensitivity in most patients with active tuberculosis and in patients with high CD4 counts, limited use, unproven performance requiring further evaluation POC None currently available Diagnose active tuberculosis in all populations, all tuberculosis-infected (including LTBI), treatment monitoring assays, predict LTBI progression to disease Likely will need to compromise in other parameters (cost, speed, accuracy), new biomarkers needed to address clinical questions Other LAMP, VOC, phenotypic tests Potential to develop a low-cost assay is of interest Unproven technology still needs validation HIV May have regulatory approval Antibody EIA, WB, IFA, POC Cheap, easy to use and widely adopted for both screening and confirmatory testing, cost effective and POC permitting community-based testing Misses window phase infections (RNA+Ab–), not useful for EID, WB indeterminant issues, high FP rates in HIV-vaccinated volunteers Antibody/Antigen (4th generation) EIA, POC Reduces window period by capturing individuals who are seronegative but anigenemic, cost effective Misses window phase infections (RNA+Ag–), not useful for EID, high FP rates in HIV-vaccinated volunteers Nucleic acid Commercial kits, in-house assays High sensitivity/specificity, able to monitor treatment High cost, requirement for laboratory facilities and skilled staff Not currently available Nucleic acid POC Useful for acute infection, EID, drug resistance monitoring, and vaccine testing Likely need to compromise in other parameters (cost, speed, accuracy) Ultra-sensitive p24 EIA, POC Best used for EID, potential for identifying most acute infections Likely need to compromise in other parameters (cost, speed, accuracy)

Abbreviations: DST, drug susceptibility testing; EIA, enzyme-linked immunoassay; EID, early infant diagnosis; FN, false negative; FP, false positive; HIV, human immunodeficiency virus; IFA, immunofluorescence assay; IGRA, interferon-γ release assay; INH/RIF, isoniazid/rifampicin; LAM, lipoarabinomannan; LAMP, loop-mediated amplification; LED, light emitting diode; LPA, line probe assay; LTBI, latent tuberculosis infection; MGIT, mycobacteria growth indicator tube; MODS, microscopic observation drug susceptibility test; MTB/RIF,Mycobacterium tuberculosis/rifampicin; POC, point-of-care; TLA, thin-layer agar test; TST, tuberculin skin test; VOC, volatile organic compound; WB, Western blot; WHO, World Health Organization.

## CURRENT DIAGNOSTIC CHALLENGES FOR HIV

With the adoption and implementation of rapid, cost-effective, POC HIV diagnostic screening assays worldwide, many middle- and low-income countries have developed algorithms for diagnosing HIV infection using strategic orthogonal testing assays without the need for additional confirmatory testing [2]. In addition, the use of alternative samples, such as saliva, have broadened the reach of testing programs [3] despite the recent reports that saliva samples have been associated with higher false-positive rates [4] in some testing sites.

Despite the advances and successes of HIV serological rapid assays, unmet gaps for qualitative POC applications remain (Table 2). First, we need to identify acute HIV infection because recently infected individuals are seronegative but have extremely high viral loads, making them far more infectious compared with chronically infected individuals [5–7]. Early detection can result in early treatment initiation, which limits disease progression [8] and reduces transmission [9–11]. Second, we need to diagnose infants born to HIV-infected mothers, particularly in sub-Saharan Africa [12]. Due to the placental transfer of antibodies during pregnancy, infants may test positive on serological assays yet may not be infected with HIV [13]. Currently, infant diagnosis is primarily performed by collecting dried blood spots and transporting the samples to central testing laboratories for DNA testing [14]. However, many infants go undiagnosed because of long turnaround times or insufficient infrastructure (eg, unreliable sample transportation) [13]. Third, we need appropriate tests for individuals vaccinated with experimental HIV antigens in clinical trials [15]. Vaccine volunteers mount specific immune responses to the vaccine constructs, which react with many serological diagnostic tests, making future HIV diagnosis difficult, potentially unblinding trial staff, and negatively impacting society [16, 17]. Similar assays may also be beneficial in resource-limited settings to monitor antiretroviral therapy (ART) effectiveness and the potential need to switch therapy [18]. In each of these cases, new rapid molecular POC HIV screening tests will fill an important diagnostic gap.

Table 2.

Key Unmeta Needs for Rapid HIV and Tuberculosis Diagnostics

 Disease Gap Current Issues or Rationale HIV Acute infection • Often symptomatic and may represent an important opportunity for early HIV diagnosis. • May account for a disproportionate amount of HIV transmission, which could be averted by early diagnosis and treatment [5] Early infant diagnosis • HIV-exposed infants (in utero, postdelivery, breast milk) may have HIV antibodies but are not HIV infected, reducing the utility of rapid antibody-based tests; long turnaround times associated with laboratory-based testing may result in high patient loss to follow-up Vaccine-induced seroreactivity • As HIV vaccines are tested and brought to market, they may elicit serological cross-reactivity, rendering existing rapid HIV tests unable to distinguish between vaccination and disease. Alternative POC HIV tests are needed to overcome this limitation. Tuberculosis Analytical sensitivity • Because mycobacterial burden is often low, the sensitivity of tests based on bacterial or antigen detection is limited. POC tests are unlikely to surpass the sensitivity of culture, which itself misses an important subset of cases. • Because the humoral immune response to Mycobacteriumtuberculosis is variable, existing serological tests have poor sensitivity and specificity. Infrastructure requirements and cost • Existing high-sensitivity tests (eg, culture) require biosafety precautions, substantial processing time, and technical expertise. Newer rapid tests (eg, Xpert MTB/RIF) are simpler but still require electrical supply. • The cost of many tuberculosis tests is high and volume dependent, necessitating specimen transport networks for economy of scale. Assessment of viability and drug resistance • Rapid tests are needed that can monitor treatment (ie, discriminate between live and dead bacilli) and provide full drug resistance profiles.
 Disease Gap Current Issues or Rationale HIV Acute infection • Often symptomatic and may represent an important opportunity for early HIV diagnosis. • May account for a disproportionate amount of HIV transmission, which could be averted by early diagnosis and treatment [5] Early infant diagnosis • HIV-exposed infants (in utero, postdelivery, breast milk) may have HIV antibodies but are not HIV infected, reducing the utility of rapid antibody-based tests; long turnaround times associated with laboratory-based testing may result in high patient loss to follow-up Vaccine-induced seroreactivity • As HIV vaccines are tested and brought to market, they may elicit serological cross-reactivity, rendering existing rapid HIV tests unable to distinguish between vaccination and disease. Alternative POC HIV tests are needed to overcome this limitation. Tuberculosis Analytical sensitivity • Because mycobacterial burden is often low, the sensitivity of tests based on bacterial or antigen detection is limited. POC tests are unlikely to surpass the sensitivity of culture, which itself misses an important subset of cases. • Because the humoral immune response to Mycobacteriumtuberculosis is variable, existing serological tests have poor sensitivity and specificity. Infrastructure requirements and cost • Existing high-sensitivity tests (eg, culture) require biosafety precautions, substantial processing time, and technical expertise. Newer rapid tests (eg, Xpert MTB/RIF) are simpler but still require electrical supply. • The cost of many tuberculosis tests is high and volume dependent, necessitating specimen transport networks for economy of scale. Assessment of viability and drug resistance • Rapid tests are needed that can monitor treatment (ie, discriminate between live and dead bacilli) and provide full drug resistance profiles.

Abbreviations: HIV, human immunodeficiency virus; MTB, Mycobacterium tuberculosis; POC, point-of-care, RIF, rifampicin.

a

This table lists representative, high-priority unmet needs. Other important needs exist, as described in the text and cited articles.

## CURRENT DIAGNOSTIC CHALLENGES FOR TUBERCULOSIS

Tuberculosis diagnostics usually rely upon microscopy, clinical screening, and chest radiography to determine the need for therapy. However, from a program perspective, each of these tests has performance issues that limit their ability to either reliably diagnose or rule out tuberculosis disease. Clinical screening may be useful for ruling out tuberculosis but is not specific enough to make a firm diagnosis [19, 20]. Laboratory testing for tuberculosis includes sputum smear microscopy, sputum culture, and, more recently, nucleic acid amplification (see Nahid et al, this issue). These tests also have stark limitations that a POC test should attempt to address (Table 2).

The first need is for a test with improved sensitivity. Studies have demonstrated a wide-ranging sensitivity of sputum microscopy as a diagnostic test for tuberculosis, as low as 9% when compared with culture [21]. Even culture is known to miss a substantial number of cases—up to 20% among HIV-coinfected patients [22, 23]. On one hand, a POC test with improved sensitivity could dramatically impact the tuberculosis epidemic by complementing or replacing sputum smear; on the other hand, it is difficult to assess specificity in the absence of a true gold standard.

The second need is for a test with minimal requirements for infrastructure or human resources. For example, microscopy is often of low quality, and sample transport to a local microscopy center might not exist. Culture-based tests require training, biosafety disposal systems, and expensive technology, thus limiting their availability in most high-burden settings. More recently, less expensive microscopic-observation drug-susceptibility (MODS) culture assays have been introduced, but these are not well standardized. Similarly, traditional nucleic acid amplification tests require laboratory infrastructure and are either expensive (commercial platforms) or provide inconsistent results (in-house assays); as such, they are not widely used.

Third, we need a tuberculosis POC test capable of providing rapid information on viability (for following treatment progress) and drug resistance. Sputum smear offers no information in these regards, and culture results can take 6–8 weeks to finalize, with contamination of specimens often causing further delays.

The most recent World Health Organization (WHO)–endorsed platform from Cepheid (Xpert MTB/RIF) transcends some of those limitations, bringing a sensitive and specific test very close to the POC [24]. It is almost fully automated, with a 2-hour turnaround time, and it provides data on resistance to rifampicin (RIF), the most important tuberculosis drug available. Although not a classic POC diagnostic (eg, lateral flow or dipstick assay), Xpert MTB/RIF meets many of the desired criteria [25]. However, implementation in rural, underserved areas is likely to be hindered by the need for stable electrical supply, temperature control (2°C –28°C kit storage, ≤30°C operating temperature), and off-site maintenance, and it is probably too expensive (currently >17 per test in most developing countries) to be used on the millions of tuberculosis suspects evaluated worldwide. Furthermore, the positive predictive value of a single test indicating RIF resistance may be fairly low [26], and RIF resistance is not a universally reliable proxy marker for multidrug-resistant tuberculosis [27], making it difficult to definitively diagnose multidrug-resistant tuberculosis on the basis of a positive test. ## DEFINING DIAGNOSTIC GOALS AND TESTING ENVIRONMENTS There are a variety of clinical environments in resource-limited settings, ranging from sophisticated urban hospitals to outdoor rural outreach centers. Introduction of novel technologies must be contextualized within health systems where other technologies are already operating. Therefore, selection of sites for the deployment of novel technologies should be carefully considered and should take into account data available from national programs so that the public health impact is maximized. Importantly, existing training and quality assurance programs should be upgraded to include new technologies before deployment takes place. Infrastructure needs and operational requirements are likely to be major obstacles to scaling up POC systems for HIV and tuberculosis. Recent developments in tuberculosis diagnostics and WHO endorsement of the Xpert MTB/RIF system [28], in particular, have highlighted some of the critical elements needed for successful implementation of any POC or near-POC assay in developing country settings. Ideally, POC systems would be usable by general healthcare workers with <1 day of training and require limited preparation, material, and equipment. For a more complete list, the reader is referred to the previously published minimum technical specifications for POC HIV [29] and tuberculosis [30] diagnostics. When deciding where to implement new POC diagnostic equipment, those in charge of a laboratory network must decide into which clinical environments to place equipment and how it would fit into the diagnostic algorithm. Similarly, when designing new diagnostic equipment, product developers must decide the clinical environments for which they are designing the equipment. When data on these clinical environments are not readily available or differ from one place to another, product developers and laboratory network managers must make assumptions (eg, using mathematical models) about the clinical environment’s laboratory infrastructure, human resources, and patient flow [31]. Laboratory infrastructure is a critical consideration for POC equipment design and placement. If a POC device requires stable electricity, then it will not be deployable to a majority of clinical locations in high-burden areas [32]. The frequency and duration of electricity outages and power surges are not typically tracked, making it difficult for laboratory network managers to weigh the investment in an uninterrupted power supply and/or generator to accompany a POC device. In addition to electricity, a POC device may require water or a cold chain, which may similarly exclude most health facilities from the placement decision [33]. As a result of extensive discussions in the public health communities [17, 34], developers have started to consider specific user requirements from resource-limited settings in their product designs. Consequently, new generations of products have been designed with a clear focus on these end-user requirements. This development has been greatly facilitated by technological advances like high-capacity lithium-ion batteries, light emitting diode light sources for imaging, molded optical setups, and rugged fluidics and mechanics. Most importantly, freeze-dried, high-quality reagents, stable without refrigeration, have enabled development of products capable of delivering laboratory-quality test results at the POC. ## QUALITY ASSURANCE FOR POC TESTING Ensuring quality of testing is the cornerstone of assuring the reliability of POC diagnostic testing. The lack in quality assurance has been criticized by clinicians and laboratorians and results in low acceptance of data from these types of devices. The Pima instrument (Alere), the first fully field-deployable CD4 product, has been extensively evaluated and shown to have excellent analytical performance [35, 36]. However, for successful implementation, certain parts of existing systems of quality assurance, such as sample collection, have to be reconsidered. Currently, valid external quality assurance (EQA) standards have been developed and agreed on based on the feasibility on the ground. However EQA practices rely on distributing stabilized samples to each individual test site. New EQA methods may need to be developed for whole blood to verify proper sample collection and processing. Despite significant efforts, no sample used in EQA schemes so far has been proven to be sufficiently stable to fulfill the requirements of CD4 POC testing at the ground level. A distribution network for incorporating proficiency testing must be considered as part of any rollout plan. An additional concern may be the development of quality control procedures for sample collection. One solution to this is the development of internal controls in conjunction with integrated connectivity, thus allowing the transmission of quality data to a central site for independent review. Obstacles include developing the necessary communications infrastructure and establishing standardized connectivity protocols to create a truly decentralized yet fully quality-assured network of test sites in any part of the world. Monitoring quality across a large number of testing sites will require new models of quality assurance that provide meaningful, real-time feedback on test performance. Despite user-friendly design, monitoring of the routine prospective performance of POC and similar systems will be paramount to ensure the integrity and validity of the test result. Turnaround time and performance characteristics associated with a particular POC diagnostic assay, 2 pieces of a much larger programmatic puzzle, must be evaluated. Several additional parameters, including suspect identification, specimen collection and transport, result reporting, result interpretation, treatment initiation, and monitoring, should also be considered. Therefore, performance indicators, potentially in the form of operational research, should be designed to measure the various preanalytic, analytic, and postanalytic phases of diagnostic testing. In doing so, areas for strengthening the laboratory–program interface can be identified, and quality improvement projects can be easily monitored systematically. ## OBSTACLES TO UPTAKE OF POC TECHNOLOGIES A number of elements present barriers to both the uptake and the effective utilization of laboratory testing for service delivery. Here we describe 3 key challenges: legal constraints, clarity of normative and regulatory recommendations, and linking laboratory testing to clinical care. Legal issues governing conducting and reporting laboratory tests may prevent the implementation of POC technologies. In many countries, nonlaboratory personnel cannot perform even the simplest laboratory tests (eg, dipstick urinalysis), which disincentivizes the adoption of POC tests designed for use in the field by clinical staff. Access to appropriate diagnostic assays is essential to provide effective and high-quality clinical care. However, international and national guidelines often lack clarity around how tests should be used, leaving interpretation up to the clinician. This can become an impediment to delivery of services. A good example is the guidance for discretionary use of HIV-1 viral load to confirm treatment failure of patients on ART. If clinicians interpret this to mean that viral load testing is a necessary prerequisite before starting costly second-line drugs, then poor access to viral load may in turn limit access to treatment. Such conditional recommendations without clear interpretation make it difficult for national programs to allocate resources to laboratory testing or for manufacturers to estimate need and therefore determine the types of tests that might best serve the market. All of this is compounded by an inconsistent regulatory environment (see Palamountain et al, this issue). In some settings there are strict national-level regulations and well-described standards that manufacturers and tests must meet. In other settings, national governments might use the reference standards of international regulatory agencies or have no defined policies whatsoever to determine what types of diagnostic assays are acceptable for use. For manufacturers, such inconsistencies make the process of seeking regulatory approval complex and unpredictable, and for national programs, a lack of alignment between normative and regulatory guidance makes it hard to put newer technologies in place. Too often, test results do not translate into care decisions. The recent increase in access to DNA polymerase chain reaction for diagnosis of HIV in infants is a case in point. Unprecedented scale-up of infant diagnosis has been achieved over the past 5 years, and currently an estimated 400 000 infant diagnosis tests are performed annually [37]. However, despite clear guidance from WHO that all HIV-infected infants be treated immediately, program reviews suggest that less than half of all HIV-positive infants are referred to ART treatment centers. Irrespective of the nature of the diagnostic test, whether laboratory based or POC, it is essential that testing and service delivery are linked in the care and treatment program. ## LABORATORY STRENGTHENING TO SUPPORT DECENTRALIZED POC DIAGNOSTICS In recent years, international aid programs to combat HIV, tuberculosis, and malaria have increased investment in developing countries [38]. Conventional centralized testing is typically more efficient, with higher throughput, and is easier to manage and quality assure than highly decentralized testing. However, significant gaps exist because conventional testing often cannot reach patients living in remote areas and the delay required to send specimens to a laboratory results in loss to follow-up [39]. Centralized tests typically require multiple component products; for example, even simple staining tests or CD4 count tests require >10–15 individual products [40]. Insufficient access to health services in resource-limited settings is a major factor impacting expanded treatment. Access to adequate laboratory services is particularly deficient and often nonexistent in rural settings. In addition, poor laboratory results reduce confidence in diagnostic testing by the healthcare providers [41]. The strengthening of laboratory systems in resource-limited settings is therefore becoming an important priority. The implementation of POC testing will place significant demands on the laboratory systems required to ensure high-quality and reliable testing. For a successful laboratory strengthening program, it is important to address several factors [42]. Collaborations are essential, and building alliances to develop local capacity and ownership is imperative for sustainability. With both NGOs and government organizations, the first step is to plan strategically and create a budgeted operational plan. Following this, the program must, through local leadership, use the plan to introduce technology, improve and maintain infrastructure, implement quality programs, and develop human resources. The establishment and implementation of laboratory networks is an important step in understanding the country disease burden and ensuring the quality of results as countries implement EQA programs. In developing economies, maintaining adequate quantities of testing stocks at each site will require enhanced, interactive, and highly responsive supply chains. The supply chain management system of essential health commodities experiences problems leading to stockout and expiry of essential reagents [43], and POC systems are not likely to be the exception. In fact, improved diagnostic capacity is expected to increase the number of individuals on treatment, placing increased pressure on tenuous health systems. To realize the impact of POC devices, it is important to address factors responsible for drug stockouts and expiry, such as inaccurate supply quantification and forecasting, prolonged procedural requirements, and delay in ordering supplies. As testing becomes more dispersed, it will become important to implement systems that check in with test operators on a regular basis and ensure that testing is still being done, that there are adequate supplies of consumables, and that the instruments are in working condition. Point-of-care diagnostic devices (as all laboratory equipment) require regular calibration, maintenance, and repair, regardless of ease of use. Manufacturers of POC systems should develop local capacity for technical assistance where possible because transporting equipment or technical staff is not cost effective in the long run. The use of modular diagnostic equipment with interchangeable internal components makes repair simpler and reduces the number of parts that need to be stocked. Although many diagnostic companies prefer to switch out rather than repair diagnostic machines, many devices are purchased through government programs that do not allow switch-outs. ## POINT-OF-CARE DIAGNOSTICS: HIV OPERATIONAL CHALLENGES Notwithstanding the fact that the majority of HIV rapid tests are easy to perform and appropriate for nonlaboratory settings, implementation and rapid growth of HIV rapid testing has created challenges and raised concerns about the accuracy and reliability of POC diagnostics [44]. These concerns include those associated to both the test (eg, rapid growth of testing not keeping pace with external quality assurance programs) and testers (eg, lack of sufficiently trained personnel to perform accurate POC diagnosis) [45]. Poor-quality test kits and inconsistent test kit lots, which are not evaluated routinely by qualified laboratories, impact accuracy. As an example, WHO, in the last quarter of 2011, discovered unacceptably high rates of invalid results of certain lots of SD Bioline 1/2 3.0 manufactured by Standard Diagnostics from the Republic of Korea [46, 47]. Consequently, on 20 December 2011, the US Agency for International Development removed this test from the list of approved diagnostics, and the US President’s Emergency Plan for AIDS Relief (PEPFAR) advised its country teams to cease procurement and to quarantine unused stocks [48]. As a result of this delisting, countries have been forced to revise their testing algorithms to assure continuity of testing services. In addition to the evaluation of the diagnostic device, test operators should undergo a comprehensive hands-on training that focuses on following the manufacturer’s instructions in test procedures, allowing for accurate interpretation of test results [49]. Despite best efforts to provide extensive training, high attrition rates and personnel turnover can still lead to severe capacity challenges, including inadequate documentation [50]. Systematic approaches to improve test accuracy during the scaling-up of POC HIV rapid testing have included the use of validated kits [48], extensive training with an emphasis on quality, use of standardized logbooks, proficiency testing, new kit lot verification, and postmarket surveillance [51]. ## POC DIAGNOSTICS: TUBERCULOSIS OPERATIONAL CHALLENGES Over the past decade, tuberculosis case detection and treatment rates have improved dramatically, and tuberculosis mortality has declined as a result [52]. Further improvements, however, will likely require the use of better diagnostic tools. There are still significant unmet needs in developing countries; >3.5 million tuberculosis cases were undiagnosed and untreated in 2009 [52]. Although new diagnostic tools developed globally over the past several years may help close this gap, adoption and implementation of these tools remains a substantial challenge in many high-burden countries. The Stop TB Partnership Research Movement, with funding from the Global Fund to Fight AIDS, Tuberculosis and Malaria (GFATM), identified key areas in which knowledge gaps hamper effective implementation of tuberculosis control [53]. Ministries of health, through their national tuberculosis and HIV control programs, must create an environment that can rapidly integrate new diagnostics, develop new policies for diagnostic testing, and scale up the new technology countrywide [52, 54]. Countrywide tuberculosis control is exceedingly complex, requiring coordination of numerous financial, technical, and implementing partners, public and private agencies, powerful professional organizations and advocacy groups, and research/training institutes [54, 55]. All stakeholders have a vested interest in tuberculosis control, and their buy-in is critical to the rapid uptake of new diagnostics, but staffing to coordinate these stakeholders at the national level is often grossly inadequate. Challenges include lack of strong national regulatory frameworks; lack of accountability, transparency, and commitment; weak capacity of program managers to effect change; inadequate infrastructure, equipment, and staff; and severe financial limitations. A potential solution to individual country regulatory agencies is the adoption of regional regulatory bodies, which can assess how effective technologies are and make recommendations for use. The adoption and development of a new policy to introduce a new TB diagnostic in a country is a complex and time-consuming process (see Cobelens et al, this issue). Countries must adapt global recommendations to national needs and resources and conduct analyses to look at the needs and evidence for change, risks, and benefits of new diagnostics and capacity of the current environment to adopt, introduce, and implement the new tuberculosis diagnostic [54]. The successful introduction and implementation of new tuberculosis diagnostics challenges every technical facet of a national tuberculosis control program, from registration of products to guideline development to dissemination of materials and training of staff. Given the association and high burden of HIV/tuberculosis coinfection in many resource-limited settings, implementation of a new tuberculosis diagnostic entails coordination with HIV programs and integration into HIV diagnostic infrastructure. Operationally, program managers must phase out existing tools; ensure availability of new tools in both public and private sectors; adapt procurement, distribution and inventory management plans; and revise monitoring and evaluation systems. ## LOOKING AHEAD: MODELS TO INFORM THE IMPLEMENTATION OF POC DIAGNOSTICS Given that the challenges outlined above are substantial, developers and public health officials must project the cost and public health impact of introducing POC tests before investing in their development. Mathematical models—including dynamic transmission models [56] and economic models [57]—are well suited to this purpose. Existing analyses have evaluated multiple applications of rapid HIV testing [58–60], but unfortunately, extension of these methods to novel POC diagnostics for HIV and tuberculosis has been limited [61, 62]. As such, decisions regarding the development and implementation of such tests are often made without a formal framework for incorporating existing evidence and prevailing knowledge. Modeling the impact and cost effectiveness of POC tests for HIV and tuberculosis is limited not just by the paucity of available evidence but also by methodological challenges [63]. Perhaps greatest among these is the difficulty of estimating the operational and clinical impact of POC tests after implementation in the field. Not only is the accuracy of POC tests often lower in the field than in the laboratory [64, 65], but implementing POC diagnostics can alter healthcare processes and workflows beyond simply increasing sensitivity, specificity, or access to services. For example, integrated HIV testing and POC CD4 testing may raise the initial CD4 count at diagnosis [66], increasing demands on ART programs, and rapid tuberculosis diagnosis with drug susceptibility testing may reduce loss to follow-up and time to diagnosis of drug resistance [24], making certain laboratory infrastructure unnecessary and increasing demand for second-line tuberculosis therapy. Models of POC diagnostics’ cost effectiveness and impact must account for these operational realities. Ultimately, a new generation of translational models must be developed to inform test developers and healthcare decision makers of the likely epidemiologic and economic impacts of POC diagnostics before these tests are fully implemented and to update these estimates as operational data accrue. These models must account for the operational realities of POC testing [67], incorporate a variety of outcomes (including economic, public health, and patient centered), and be made accessible to decision makers across a spectrum of local conditions. Only by creating such models can POC diagnostics for HIV and tuberculosis be developed and implemented in a rational, evidence-based fashion. ## LESSONS LEARNED Because care and treatment programs have shown great success in an emergency phase, there is a renewed impetus for ensuring sustainability and that investments are efficient and have a measurable impact on disease burden. An example of such a smart investment is found in the experience of PEPFAR’s supply chain management system shifting from mainly air freight to predominantly sea freight. Transportation savings are achieved through extended forecasts and synchronized planning, but the magnitude of savings far outweighs the effort expended. For example, shipping commodities equivalent to 2 containers costs approximately20 000 by sea freight (8–10 weeks’ lead time) compared with >$230 000 by air freight (2–3 weeks). As of June 2011, PEPFAR’s intentional shift from air to sea freight has resulted in$52 million in savings through December 2010 (unpublished data). These savings (equal to a year’s supply of antiretrovirals for 575 000 persons) represent funding that can be reallocated to service expansion. Another important and tangible spillover effect is noted in the overall technical capacity building that such advance planning entails.

Important avenues to pursue with regard to diagnostics include:

• Further leveraging of pooled procurement mechanisms, such as PEPFAR’s supply chain management system and GFATM’s voluntary pooled procurement (VPP) program, applying value-for-money criteria to the laboratory sector. The VPP mechanism is a strategic initiative established by GFATM for the recipients of its grants to leverage GFATM’s market influence in order to achieve better prices for quality health products and to enhance the long-term sustainability of the market. It also aims to address some of the procurement problems faced by recipients, such as long processes and stockouts caused by delays in procurement.

• Further development of multipartner collaborations such as the Coordinated Procurement Planning (CPP) initiative. The goal of the CPP initiative is to help countries, donors, and implementing partners improve procurement and supply management planning to support the continuous availability of antiretrovirals and other essential medicines and medical commodities required by HIV/AIDS patients. The CPP initiative originated out of a PEPFAR implementers’ meeting to ensure that commodities are available for uninterrupted ART services; efforts are focused on strengthening country-level coordination and information sharing among funders and technical agencies to avert stockouts and improve procurement effectiveness.

• Evaluating lessons learned through the Foundation for Innovate New Diagnostics experience in deriving best price models with Xpert.

## CONCLUSIONS

Point-of-care testing will reduce many technical and operational challenges associated with laboratory-based testing but will create others as more patients are tested and require treatment and quality assurance becomes more distributed. As new POC devices are deployed, it will be important to find the right balance between POC and conventional testing to achieve the greatest benefit from each type of technology. Striking this balance is not simple and requires consideration of the distribution of test demand across different sites, all-inclusive testing costs (ie, including facility, staff, and overhead costs), human resource and infrastructure requirements, maintenance, quality monitoring, supply chain needs, and potential patient benefits (Table 3). For example, despite the benefits of POC testing, conventional laboratory capacity may be more cost effective or better suited to high volume testing in urban areas with stable infrastructure and transportation systems. Centralized referral networks may be more cost effective than widespread peripheral distribution of POC instruments, especially because peripheral facilities may have low patient volumes. To assess the optimal balance between POC and centralized laboratory testing, market researchers should evaluate the number of affected clinics, existing laboratory infrastructure and human resources, and the volume of patients seen at each level.

Table 3.

Current Implementation Obstacles for Introducing New Diagnostic Tools in Resource-Limited Settings and Possible Solutions

 Issues Challenges Potential Solutions Laboratory strengthening Lack of human resources and infrastructure; rudimentary laboratory network Develop a national laboratory strategic plan that includes funding support from the national budget; utilize telecommunication network to link devices with central laboratory facility Regulatory Lack of international guidelines that translate to local needs; lack of transparency, accountability, commitment and policy at the country level; inconsistent regulatory environment Develop regional bodies to compile available data and guidance; make recommendations based on feasibility studies to meet regional needs; accelerate WHO prequalification of new diagnostics Quality control Lack controls to ensure the assay is functioning properly and guarantee validity of test performance Include internal process controls in the assay; use standardized connectivity methods to transmit data to central laboratory facility Quality assurance Lack controls to ensure the sample is processed correctly; more costly to implement; may require temperature controlled storage of QA reagents; relatively short shelf life Develop new models to incorporate sample processing; use standardized connectivity methods to transmit data to central laboratory facility; centralize procurement Supply chain management Inadequate supply of reagents; need to abrogate stockouts and use of expired reagents Institute programs to accurately forecast supply needs; link with connectivity to monitor use; follow up with warm lines; centralize procurement; develop distribution network; evaluate and release test lots at a central facility Maintenance and repair Lack of qualified human resources to perform routine maintenance and repair of equipment in the field Include modular design for easier repair; develop local capacity; institute switchout programs (if possible) Training and retention Lack of sufficient training for the appropriate persons performing assays; high turnover of staff Develop training requirements that are dictated by where testing will be conducted such as secondary (hospitals), tertiary (clinics), or health outposts; provide performance incentives; secure funding to increase wages
 Issues Challenges Potential Solutions Laboratory strengthening Lack of human resources and infrastructure; rudimentary laboratory network Develop a national laboratory strategic plan that includes funding support from the national budget; utilize telecommunication network to link devices with central laboratory facility Regulatory Lack of international guidelines that translate to local needs; lack of transparency, accountability, commitment and policy at the country level; inconsistent regulatory environment Develop regional bodies to compile available data and guidance; make recommendations based on feasibility studies to meet regional needs; accelerate WHO prequalification of new diagnostics Quality control Lack controls to ensure the assay is functioning properly and guarantee validity of test performance Include internal process controls in the assay; use standardized connectivity methods to transmit data to central laboratory facility Quality assurance Lack controls to ensure the sample is processed correctly; more costly to implement; may require temperature controlled storage of QA reagents; relatively short shelf life Develop new models to incorporate sample processing; use standardized connectivity methods to transmit data to central laboratory facility; centralize procurement Supply chain management Inadequate supply of reagents; need to abrogate stockouts and use of expired reagents Institute programs to accurately forecast supply needs; link with connectivity to monitor use; follow up with warm lines; centralize procurement; develop distribution network; evaluate and release test lots at a central facility Maintenance and repair Lack of qualified human resources to perform routine maintenance and repair of equipment in the field Include modular design for easier repair; develop local capacity; institute switchout programs (if possible) Training and retention Lack of sufficient training for the appropriate persons performing assays; high turnover of staff Develop training requirements that are dictated by where testing will be conducted such as secondary (hospitals), tertiary (clinics), or health outposts; provide performance incentives; secure funding to increase wages

Abbreviations: QA, quality assurance; WHO, World Health Organization.

New testing platforms can also be slow to implement on a national scale, so conventional laboratory platforms are likely to remain the primary diagnostic systems for years after competing POC technology is initially introduced. There are significant barriers to new test adoption, including the large cost of switching from one test platform to another and quality assurance coverage. To be made efficient, decisions about how to deploy POC diagnostics require modeling of their projected impact and cost effectiveness under local conditions. In the end, ongoing dialogue and compromise across a broad partnership of stakeholders is required if we are to fulfill the promise of POC diagnostics for combating the dual epidemics of HIV and tuberculosis.

## Notes

### Acknowledgments.

The authors thank Michael Kimerling of the Bill & Melinda Gates Foundation for reviewing and providing comments for this manuscript.

### Disclaimers.

The opinions expressed herein are those of the authors and should not be construed as representing the official views of the US Department of Health and Human Services, US Department of State, or the US Agency for International Development. The opinions do not necessarily reflect the official policies of, nor does mention of trade names, commercial practices, or organizations imply endorsement by,s the US government.

### Financial support.

This project has been funded in part with federal funds from the Office of the US Global AIDS Coordinator, US Department of State, US Centers for Disease Control and Prevention, and the National Institute of Allergy and Infectious Diseases, National Institutes of Health, Department of Health and Human Services (contract HHSN272200800014C).

### Potential conflicts of interest.

E. E. is employed at Alere Technologies, which developed the CD4 Pima instrument. G. J. Y. is at Becton Dickinson, which has a large portfolio of diagnostic devices. K. M. P. is President of the Northwestern Global Health Foundation to develop medical diagnostics, including a POC infant HIV test, and with Quidel Corporation to develop a molecular platform. All other authors report no potential conflicts of interest.

All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

## References

1.
Peeling
RW
Mabey
D
Point-of-care tests for diagnosing infections in the developing world
Clin Microbiol Infect
,
2010
, vol.
16
(pg.
1062
-
9
)
2.
Crucitti
T
Taylor
D
Beelaert
G
Fransen
K
Van Damme
L
Performance of a rapid and simple HIV testing algorithm in a multicenter phase III microbicide clinical trial
Clin Vaccine Immunol
,
2011
, vol.
18
(pg.
1480
-
5
)
3.
Reynolds
SJ
Muwonga
J
OraQuick ADVANCE Rapid HIV-1/2 antibody test
Expert Rev Mol Diagn
,
2004
, vol.
4
(pg.
587
-
91
)
4.
Centers for Disease Control and Prevention
False-positive oral fluid rapid HIV tests—New York City, 2005–2008
MMWR Morb Mortal Wkly Rep
,
2008
, vol.
57
(pg.
660
-
5
)
5.
Powers
KA
Ghani
AC
Miller
WC
, et al.  .
The role of acute and early HIV infection in the spread of HIV and implications for transmission prevention strategies in Lilongwe, Malawi: a modelling study
Lancet
,
2011
, vol.
378
(pg.
256
-
68
)
6.
Boily
MC
Baggaley
RF
Wang
L
, et al.  .
Heterosexual risk of HIV-1 infection per sexual act: systematic review and meta-analysis of observational studies
Lancet Infect Dis
,
2009
, vol.
9
(pg.
118
-
29
)
7.
Powers
KA
Poole
C
Pettifor
AE
Cohen
MS
Rethinking the heterosexual infectivity of HIV-1: a systematic review and meta-analysis
Lancet Infect Dis
,
2008
, vol.
8
(pg.
553
-
63
)
8.
Gras
L
van Sighem
A
Bezemer
D
Smit
C
Wit
F
de Wolf
F
Lower mortality and earlier start of combination antiretroviral therapy in patients tested repeatedly for HIV than in those with a positive first test
AIDS
,
2011
, vol.
25
(pg.
813
-
8
)
9.
Cohen
MS
Chen
YQ
McCauley
M
, et al.  .
Prevention of HIV-1 infection with early antiretroviral therapy
N Engl J Med
,
2011
, vol.
365
(pg.
493
-
505
)
10.
Hollingsworth
TD
Anderson
RM
Fraser
C
HIV-1 transmission, by stage of infection
J Infect Dis
,
2008
, vol.
198
(pg.
687
-
93
)
11.
Pilcher
CD
Eaton
L
Kalichman
S
Bisol
C
de Souza Rda
S
Approaching “HIV elimination”: interventions for acute HIV infection
Curr HIV/AIDS Rep
,
2006
, vol.
3
(pg.
160
-
8
)
12.
Ciaranello
AL
Park
JE
Ramirez-Avila
L
Freedberg
KA
Walensky
RP
Leroy
V
Early infant HIV-1 diagnosis programs in resource-limited settings: opportunities for improved outcomes and more cost-effective interventions
BMC Med
,
2011
, vol.
9
pg.
59

13.
Creek
TL
Sherman
GG
Nkengasong
J
, et al.  .
Infant human immunodeficiency virus diagnosis in resource-limited settings: issues, technologies, and country experiences
Am J Obstet Gynecol
,
2007
, vol.
197
(pg.
S64
-
71
)
14.
Chohan
BH
Emery
S
Wamalwa
D
, et al.  .
Evaluation of a single round polymerase chain reaction assay using dried blood spots for diagnosis of HIV-1 infection in infants in an African setting
BMC Pediatr
,
2011
, vol.
11
pg.
18

15.
Van Braeckel
E
Koutsoukos
M
Bourguignon
P
Clement
F
McNally
L
Leroux-Roels
G
Vaccine-induced HIV seropositivity: a problem on the rise
J Clin Virol
,
2011
, vol.
50
(pg.
334
-
7
)
16.
Cooper
CJ
Metch
B
Dragavon
J
Coombs
RW
LR
Vaccine-induced HIV seropositivity/reactivity in noninfected HIV vaccine recipients
JAMA
,
2010
, vol.
304
(pg.
275
-
83
)
17.
Schito
ML
D’Souza
MP
Owen
SM
Busch
MP
Challenges for rapid molecular HIV diagnostics
J Infect Dis
,
2010
, vol.
201

Suppl 1
(pg.
S1
-
6
)
18.
Rouet
F
Rouzioux
C
The measurement of HIV-1 viral load in resource-limited settings: how and where?
Clin Lab
,
2007
, vol.
53
(pg.
135
-
48
)
19.
Corbett
EL
Zezai
A
Cheung
YB
, et al.  .
Provider-initiated symptom screening for tuberculosis in Zimbabwe: diagnostic value and the effect of HIV status
Bull World Health Organ
,
2010
, vol.
88
(pg.
13
-
21
)
20.
Getahun
H
Kittikraisak
W
Heilig
CM
, et al.  .
Development of a standardized screening rule for tuberculosis in people living with HIV in resource-constrained settings: individual participant data meta-analysis of observational studies
PLoS Med
,
2011
, vol.
8
pg.
e1000391

21.
Bassett
IV
Wang
B
Chetty
S
, et al.  .
Intensive tuberculosis screening for HIV-infected patients starting antiretroviral therapy in Durban, South Africa
Clin Infect Dis
,
2010
, vol.
51
(pg.
823
-
9
)
22.
von Reyn
CF
Mtei
L
Arbeit
RD
, et al.  .
Prevention of tuberculosis in bacille Calmette-Guerin–primed, HIV-infected adults boosted with an inactivated whole-cell mycobacterial vaccine
AIDS
,
2010
, vol.
24
(pg.
675
-
85
)
23.
Wilson
D
Nachega
J
Morroni
C
Chaisson
R
Maartens
G
Diagnosing smear-negative tuberculosis using case definitions and treatment response in HIV-infected adults
Int J Tuberc Lung Dis
,
2006
, vol.
10
(pg.
31
-
8
)
24.
Boehme
CC
Nicol
MP
Nabeta
P
, et al.  .
Feasibility, diagnostic accuracy, and effectiveness of decentralised use of the Xpert MTB/RIF test for diagnosis of tuberculosis and multidrug resistance: a multicentre implementation study
Lancet
,
2011
, vol.
377
(pg.
1495
-
505
)
25.
Creswell
J
Falzon
D
Getahun
H
, et al.  .
Rapid implementation of the Xpert MTB/RIF diagnostic test: technical and operational “how-to”; practical considerations
,
2011
, vol.
Vol 2011

Geneva, Switzerland
WHO
26.
Lawn
SD
Brooks
SV
Kranzer
K
, et al.  .
Screening for HIV-associated tuberculosis and rifampicin resistance before antiretroviral therapy using the Xpert MTB/RIF assay: a prospective study
PLoS Med
,
2011
, vol.
8
pg.
e1001067

27.
Smith
S
Kurbatova
E
Cavanaugh
J
Cegielski
JP
Global isoniazid resistance patterns in rifampin-resistant and rifampin-susceptible tuberculosis
Int J Tuberc Lung Dis
,
2011
, vol.
16
(pg.
203
-
5
)
28.
World Health Organization
Automated real-time nucleic acid amplification technology for rapid and simultaneous detection of tuberculosis and rifampicin resistance: Xpert MTB/RIF System
,
2011
Geneva, Switzerland
WHO

29.
Usdin
M
Guillerm
M
Calmy
A
Patient needs and point-of-care requirements for HIV load testing in resource-limited settings
J Infect Dis
,
2010
, vol.
201

Suppl 1
(pg.
S73
-
7
)
30.
Batz
HG
Cooke
GS
Reid
SD
Towards lab-free tuberculosis diagnosis
,
2011
Treatment Action Group; Stop TB Partnership, Imperial College London and the MSF Access Campaign, London, UK
31.
Girosi
F
Olmsted
SS
Keeler
E
, et al.  .
Developing and interpreting models to improve diagnostics in developing countries
Nature
,
2006
, vol.
444

Suppl 1
(pg.
3
-
8
)
32.
National Coordinating Agency for Population and Development (NCAPD) Mo Medical Services (MoMS) MoPHaSMK, Kenya National Bureau of Statistics (KNBS)
IMKSP, Assessment Survey 2010. Nairobi KNCAfPa, Development MoMS, Ministry of Public Health and Sanitation, Kenya and National Bureau of Statistics aIM. Kenya Service Provision Assessment Survey 2011
,
2011

33.

Ministry of Health (MOH) [Uganda] and Macro International Inc. 2008. Uganda Service Provision Assessment Survey 2007. Kampala, Uganda: Ministry of Health and Macro International Inc
34.
Thairu
L
Katzenstein
D
Israelski
D
Operational challenges in delivering CD4 diagnostics in sub-Saharan Africa
AIDS Care
,
2011
, vol.
23
(pg.
814
-
21
)
35.
Jani
IV
Sitoe
NE
Chongo
PL
, et al.  .
Accurate CD4 T-cell enumeration and antiretroviral drug toxicity monitoring in primary healthcare clinics using point-of-care testing
AIDS
,
2011
, vol.
25
(pg.
807
-
12
)
36.
Mtapuri-Zinyowera
S
Chideme
M
Mangwanya
D
, et al.  .
Evaluation of the PIMA point-of-care CD4 analyzer in VCT clinics in Zimbabwe
J Acquir Immune Defic Syndr
,
2010
, vol.
55
(pg.
1
-
7
)
37.
Mahy
M
Stover
J
Kiragu
K
, et al.  .
What will it take to achieve virtual elimination of mother-to-child transmission of HIV? An assessment of current progress and future needs
Sex Transm Infect
,
2010
, vol.
86

Suppl 2
(pg.
ii48
-
55
)
38.
Birx
D
de Souza
M
Nkengasong
JN
Laboratory challenges in the scaling up of HIV, TB, and malaria programs: the interaction of health and laboratory systems, clinical research, and service delivery
Am J Clin Pathol
,
2009
, vol.
131
(pg.
849
-
51
)
39.
Jani
IV
Sitoe
NE
Alfai
ER
, et al.  .
Effect of point-of-care CD4 cell count tests on retention of patients and rates of antiretroviral therapy initiation in primary health clinics: an observational cohort study
Lancet
,
2011
, vol.
378
(pg.
1572
-
9
)
40.
Consultation on Technical and Operational Recommendations for Clinical Laboratory Testing Harmonization and Standardization
Helping to expand sustainable quality testing to improve the care and treatment of people infected with and affected by HIV/AIDS, TB and malaria
,
2008
Maputo, Mozambique

41.
Peter
TF
Rotz
PD
Blair
DH
Khine
AA
Freeman
RR
Murtagh
MM
Impact of laboratory accreditation on patient care and the health system
Am J Clin Pathol
,
2010
, vol.
134
(pg.
550
-
5
)
42.
Nkengasong
JN
A shifting paradigm in strengthening laboratory health systems for global health: acting now, acting collectively, but acting differently
Am J Clin Pathol
,
2010
, vol.
134
(pg.
359
-
60
)
43.
Nakyanzi
JK
Kitutu
FE
Oria
H
Kamba
PF
Expiry of medicines in supply outlets in Uganda
Bull World Health Organ
,
2010
, vol.
88
(pg.
154
-
8
)
44.
World Health Organization
Task shifting: global recommendations and guidelines
,
2008
Geneva, Switzerland
WHO

45.
World Health Organization, US Department of Health and Human Services CfDCaP
Guidelines for assuring the accuracy and reliability of HIV rapid testing: applying a quality system approach
,
2005
Geneva, Switzerland
WHO

46.
World Health Organization
Field safety notice no. 1
,
2011
Geneva, Switzerland
WHO

47.
World Health Organization
Update information on the SD Bioline HIV-1/2 v3.0
,
2011
Geneva, Switzerland
WHO

48.
US Agency for International Development
List of approved HIV/AIDS rapid test kits
,
2011

49.
AIDS Institute and Wadsworth Center, New York State Department of Health
Rapid HIV testing workbook and implementation guidelines for limited testing sites in New York State
,
2005
Albany, NY
New York State Department of Health

50.
Yao
K
Wafula
W
Bile
EC
, et al.  .
Ensuring the quality of HIV rapid testing in resource-poor countries using a systematic approach to training
Am J Clin Pathol
,
2010
, vol.
134
(pg.
568
-
72
)
51.
Parekh
BS
Kalou
MB
Alemnji
G
Ou
CY
Gershy-Damet
GM
Nkengasong
JN
Scaling up HIV rapid testing in developing countries: comprehensive approach for implementing quality assurance
Am J Clin Pathol
,
2010
, vol.
134
(pg.
573
-
84
)
52.
World Health Organization
Global tuberculosis control: WHO report 2010
,
2010
Geneva, Switzerland
WHO

53.
World Health Organization
Priorities in operational research to improve tuberculosis care and control
,
2011
Geneva, Switzerland
WHO

54.
World Health Organization
New technologies for TB control: a framework for their adoption, introduction and implementation
,
2007
Geneva, Switzerland
WHO

55.
Parsons
LM
Somoskovi
A
Gutierrez
C
, et al.  .
Laboratory diagnosis of tuberculosis in resource-poor countries: challenges and opportunities
Clin Microbiol Rev
,
2011
, vol.
24
(pg.
314
-
50
)
56.
Vynnycky
E
White
R
An introduction to infectious disease modelling
,
2010
Oxford
Oxford University Press
57.
Drummond
MF
Sculpher
MJ
Torrance
GW
O'Brien
BJ
Stoddart
GL
Methods for the economic evaluation of health care programmes
,
2005
3rd ed
Oxford
Oxford University Press
58.
Granich
RM
Gilks
CF
Dye
C
De Cock
KM
Williams
BG
Universal voluntary HIV testing with immediate antiretroviral therapy as a strategy for elimination of HIV transmission: a mathematical model
Lancet
,
2009
, vol.
373
(pg.
48
-
57
)
59.
Sanders
GD
Bayoumi
AM
Sundaram
V
, et al.  .
Cost-effectiveness of screening for HIV in the era of highly active antiretroviral therapy
N Engl J Med
,
2005
, vol.
352
(pg.
570
-
85
)
60.
Walensky
RP
Wood
R
Fofana
MO
, et al.  .
The clinical impact and cost-effectiveness of routine, voluntary HIV screening in South Africa
J Acquir Immune Defic Syndr
,
2010
, vol.
56
(pg.
26
-
35
)
61.
Dowdy
DW
O’Brien
MA
Bishai
D
Cost-effectiveness of novel diagnostic tools for the diagnosis of tuberculosis
Int J Tuberc Lung Dis
,
2008
, vol.
12
(pg.
1021
-
9
)
62.
Loubiere
S
Moatti
JP
Economic evaluation of point-of-care diagnostic technologies for infectious diseases
Clin Microbiol Infect
,
2010
, vol.
16
(pg.
1070
-
6
)
63.
Dowdy
DW
Cattamanchi
A
Steingart
KR
Pai
M
Is scale-up worth it? Challenges in economic analysis of diagnostic tests for tuberculosis
PLoS Med
,
2011
, vol.
8
pg.
e1001063

64.
Delaney
KP
Branson
BM
Uniyal
A
, et al.  .
Performance of an oral fluid rapid HIV-1/2 test: experience from four CDC studies
AIDS
,
2006
, vol.
20
(pg.
1655
-
60
)
65.
Dinnes
J
Deeks
J
Kunst
H
, et al.  .
A systematic review of rapid diagnostic tests for the detection of tuberculosis infection
Health Technol Assess
,
2007
, vol.
11
(pg.
1
-
196
)
66.
Phanuphak
N
Pattanachaiwit
S
Pankam
T
, et al.  .
Active voluntary counseling and testing with integrated CD4 count service can enhance early HIV testing and early CD4 count measurement: experiences from the Thai Red Cross Anonymous Clinic in Bangkok, Thailand
J Acquir Immune Defic Syndr
,
2010
, vol.
56
(pg.
244
-
52
)
67.
Lin
HH
Langley
I
Mwenda
R
, et al.  .
A modelling framework to support the selection and implementation of new tuberculosis diagnostic tools
Int J Tuberc Lung Dis
,
2011
, vol.
15
(pg.
996
-
1004
)

## Author notes

Presented in part: Workshop on TB and HIV Diagnostics in Adult and Pediatric Populations, Silver Spring, Maryland, 28–30 June 2011.