Integrating HIV, syphilis, malaria and anaemia point-of-care testing (POCT) for antenatal care at dispensaries in western Kenya: discrete-event simulation modelling of operational impact.

BACKGROUND: Despite WHO advocating for an integrated approach to antenatal care (ANC), testing coverage for conditions other than HIV remains low and women are referred to distant laboratories for testing. Using point-of-care tests (POCTs) at peripheral dispensaries could improve access to testing and timely treatment. However, the effect of providing additional services on nurse workload and client wait times are unknown. We use discrete-event simulation (DES) modelling to understand the effect of providing four point-of-care tests for ANC on nurse utilization and wait times for women seeking maternal and child health (MCH) services. METHODS: We collected detailed time-motion data over 20 days from one high volume dispensary in western Kenya during the 8-month implementation period (2014-2015) of the intervention. We constructed a simulation model using empirical arrival distributions, activity durations and client pathways of women seeking MCH services. We removed the intervention from the model to obtain wait times, length-of-stay and nurse utilization rates for the baseline scenario where only HIV testing was offered for ANC. Additionally, we modelled a scenario where nurse consultations were set to have minimum durations for sufficient delivery of all WHO-recommended services. RESULTS: A total of 183 women visited the dispensary for MCH services and 14 of these women received point-of-care testing (POCT). The mean difference in total waiting time was 2 min (95%CI: < 1-4 min, p = 0.026) for MCH women when integrated POCT was given, and 9 min (95%CI: 4-14 min, p < 0.001) when integrated POCT with adequate ANC consult times was given compared to the baseline scenario. Mean length-of-stay increased by 2 min (95%CI: < 1-4 min, p = 0.015) with integrated POCT and by 16 min (95%CI: 10-21 min, p < 0.001) with integrated POCT and adequate consult times compared to the baseline scenario. The two nurses' overall daily utilization in the scenario with sufficient minimum consult durations were 72 and 75%. CONCLUSION: The intervention had a modest overall impact on wait times and length-of-stay for women seeking MCH services while ensuring pregnant women received essential diagnostic testing. Nurse utilization rates fluctuated among days: nurses experienced spikes in workload on some days but were under-utilized on the majority of days. Overall, our model suggests there was sufficient time to deliver all WHO's required ANC activities and offer integrated testing for ANC first and re-visits with the current number of healthcare staff. Further investigations on improving healthcare worker, availability, performance and quality of care are needed. Delivering four point-of-care tests together for ANC at dispensary level would be a low burden strategy to improve ANC.

Outcomes from patients with presumed drug resistant tuberculosis in five reference centers in Brazil.

BACKGROUND: The implementation of rapid drug susceptibility testing (DST) is a current global priority for TB control. However, data are scarce on patient-relevant outcomes for presumptive diagnosis of drug-resistant tuberculosis (pDR-TB) evaluated under field conditions in high burden countries. METHODS: Observational study of pDR-TB patients referred by primary and secondary health units. TB reference centers addressing DR-TB in five cities in Brazil. Patients age 18 years and older were eligible if pDR-TB, culture positive results for Mycobacterium tuberculosis and, if no prior DST results from another laboratory were used by a physician to start anti-TB treatment. The outcome measures were median time from triage to initiating appropriate anti-TB treatment, empirical treatment and, the treatment outcomes. RESULTS: Between February,16th, 2011 and February, 15th, 2012, among 175 pDR TB cases, 110 (63.0%) confirmed TB cases with DST results were enrolled. Among study participants, 72 (65.5%) were male and 62 (56.4%) aged 26 to 45 years. At triage, empirical treatment was given to 106 (96.0%) subjects. Among those, 85 were treated with first line drugs and 21 with second line. Median time for DST results was 69.5 [interquartile - IQR: 35.7-111.0] days and, for initiating appropriate anti-TB treatment, the median time was 1.0 (IQR: 0-41.2) days. Among 95 patients that were followed-up during the first 6 month period, 24 (25.3%; IC: 17.5%-34.9%) changed or initiated the treatment after DST results: 16/29 MDRTB, 5/21 DR-TB and 3/45 DS-TB cases. Comparing the treatment outcome to DS-TB cases, MDRTB had higher proportions changing or initiating treatment after DST results (p = 0.01) and favorable outcomes (p = 0.07). CONCLUSIONS: This study shows a high rate of empirical treatment and long delay for DST results. Strategies to speed up the detection and early treatment of drug resistant TB should be prioritized.

Patient-cost survey for tuberculosis in the context of patient-pathway modelling.

SETTING: Eight tuberculosis treatment sites in Cavite Province, the Philippines, including two sites specialising in management of multidrug-resistant tuberculosis (MDR-TB).OBJECTIVE: To evaluate costs incurred by TB patients and to determine the proportion of households that faced catastrophic costs, then to consider cost survey responses alongside results of detailed patient-pathway modelling.DESIGN: Clustered cross-sectional survey using a field testing version of the WHO TB patient-costing tool and protocol; face-to-face interviews with 194 patients conducted in May-August 2016. Costs included direct-medical, direct non-medical and indirect costs using the human capital approach. Patients were deemed to incur catastrophic expenditure if TB-related costs exceeded 20% of annual household income. Patient pathways were modelled following multiple health staff interviews.RESULTS: Estimated mean cost incurred by patients with drug-susceptible TB was US$321 vs. $2356 for MDR-TB patients. Catastrophic costs were suffered by 28% of drug-susceptible and 80% of MDR-TB patients, with lost income being the largest contributor. Patient-pathway modelling suggested most patients had under-reported health visits.CONCLUSION: Survey results indicate that patient costs are large for all patients in Cavite, particularly for MDR-TB patients. Patient-pathway modelling suggests these costs are an underestimate due to poor recollection of health visits, suggesting that the WHO instrument and protocol could be improved to better capture the diagnostic journey.

Improving rifampicin-resistant tuberculosis diagnosis using Xpert® MTB/RIF: modelling interventions and costs.

SETTING: Cape Town, South Africa. OBJECTIVE: To model the diagnosis of rifampicin-resistant tuberculosis (RR-TB) and laboratory costs of smear/culture and Xpert-based algorithms and the effect of varying adherence and human immunodeficiency virus (HIV) testing in the Xpert-based algorithm. METHODS: We used a validated operational model (100 000 population) and published laboratory cost data. We estimated the number and cost of RR-TB cases identified using the smear/culture- and Xpert-based algorithms. We modelled varying adherence and different levels of known HIV status against the Xpert-based algorithm. RESULTS: The number of RR-TB cases identified increased from 603 with smear/culture to 1178 with the Xpert-based algorithm (100% adherence; 60% knew their HIV status). The overall laboratory cost increased from US$1 073 858 to US$2 430 050 and the cost per RR-TB case identified increased from US$1781 to US$2063 in the respective algorithms. When adherence to the Xpert-based algorithm was increased from 50% to 100% (60% knew their HIV status), the number of RR-TB cases identified increased from 721 to 1178. CONCLUSION: The Xpert-based algorithm is efficient in identifying RR-TB, as the increase in costs is offset by the increase in the number of cases identified. Adherence to the Xpert-based algorithm is important to ensure that all presumptive TB cases receive the benefit of simultaneous TB and RR-TB testing.

High laboratory cost predicted per tuberculosis case diagnosed with increased case finding without a triage strategy.

SETTING: Cape Town, South Africa. OBJECTIVE: To model the effects of increased case finding and triage strategies on laboratory costs per tuberculosis (TB) case diagnosed. METHODS: We used a validated operational model and published laboratory cost data. We modelled the effect of varying the proportion with TB among presumptive cases and Xpert cartridge price reductions on cost per TB case and per additional TB case diagnosed in the Xpert-based vs. smear/culture-based algorithms. RESULTS: In our current scenario (18.3% with TB among presumptive cases), the proportion of cases diagnosed increased by 8.7% (16.7% vs. 15.0%), and the cost per case diagnosed increased by 142% (US$121 vs. US$50). The cost per additional case diagnosed was US$986. This would increase to US$1619 if the proportion with TB among presumptive cases was 10.6%. At 25.9-30.8% of TB prevalence among presumptive cases and a 50% reduction in Xpert cartridge price, the cost per TB case diagnosed would range from US$50 to US$59 (comparable to the US$48.77 found in routine practice with smear/culture). CONCLUSION: The operational model illustrates the effect of increased case finding on laboratory costs per TB case diagnosed. Unless triage strategies are identified, the approach will not be sustainable, even if Xpert cartridge prices are reduced.

Operational modelling: the mechanisms influencing TB diagnostic yield in an Xpert® MTB/RIF-based algorithm.

SETTING: Cape Town, South Africa. OBJECTIVE: To compare the diagnostic yield for smear/culture and Xpert® MTB/RIF algorithms and to investigate the mechanisms influencing tuberculosis (TB) yield. METHOD: We developed and validated an operational model of the TB diagnostic process, first with the smear/culture algorithm and then with the Xpert algorithm. We modelled scenarios by varying TB prevalence, adherence to diagnostic algorithms and human immunodeficiency virus (HIV) status. This enabled direct comparisons of diagnostic yield in the two algorithms to be made. RESULTS: Routine data showed that diagnostic yield had decreased over the period of the Xpert algorithm roll-out compared to the yield when the smear/culture algorithm was in place. However, modelling yield under identical conditions indicated a 13.3% increase in diagnostic yield from the Xpert algorithm compared to smear/culture. The model demonstrated that the extensive use of culture in the smear/culture algorithm and the decline in TB prevalence are the main factors contributing to not finding an increase in diagnostic yield in the routine data. CONCLUSION: We demonstrate the benefits of an operational model to determine the effect of scale-up of a new diagnostic algorithm, and recommend that policy makers use operational modelling to make appropriate decisions before new diagnostic algorithms are scaled up.

Impact and cost-effectiveness of current and future tuberculosis diagnostics: the contribution of modelling.

The landscape of diagnostic testing for tuberculosis (TB) is changing rapidly, and stakeholders need urgent guidance on how to develop, deploy and optimize TB diagnostics in a way that maximizes impact and makes best use of available resources. When decisions must be made with only incomplete or preliminary data available, modelling is a useful tool for providing such guidance. Following a meeting of modelers and other key stakeholders organized by the TB Modelling and Analysis Consortium, we propose a conceptual framework for positioning models of TB diagnostics. We use that framework to describe modelling priorities in four key areas: Xpert(®) MTB/RIF scale-up, target product profiles for novel assays, drug susceptibility testing to support new drug regimens, and the improvement of future TB diagnostic models. If we are to maximize the impact and cost-effectiveness of TB diagnostics, these modelling priorities should figure prominently as targets for future research.

Outcomes from patients with presumed drug resistant tuberculosis in five reference centers in Brazil

Background: The implementation of rapid drug susceptibility testing (DST) is a current global priority for TBcontrol. However, data are scarce on patient-relevant outcomes for presumptive diagnosis of drug-resistanttuberculosis (pDR-TB) evaluated under field conditions in high burden countries.Methods: Observational study of pDR-TB patients referred by primary and secondary health units. TB referencecenters addressing DR-TB in five cities in Brazil. Patients age 18 years and older were eligible if pDR-TB, culturepositive results for Mycobacterium tuberculosis and, if no prior DST results from another laboratory were used by aphysician to start anti-TB treatment. The outcome measures were median time from triage to initiating appropriateanti-TB treatment, empirical treatment and, the treatment outcomes.Results: Between February,16th, 2011 and February, 15th, 2012, among 175 pDR TB cases, 110 (63.0%) confirmed TBcases with DST results were enrolled. Among study participants, 72 (65.5%) were male and 62 (56.4%) aged 26 to45 years. At triage, empirical treatment was given to 106 (96.0%) subjects. Among those, 85 were treated with firstline drugs and 21 with second line. Median time for DST results was 69.5 [interquartile - IQR: 35.7–111.0] days and,for initiating appropriate anti-TB treatment, the median time was 1.0 (IQR: 0–41.2) days. Among 95 patients thatwere followed-up during the first 6 month period, 24 (25.3%; IC: 17.5%–34.9%) changed or initiated the treatmentafter DST results: 16/29 MDRTB, 5/21 DR-TB and 3/45 DS-TB cases. Comparing the treatment outcome to DS-TBcases, MDRTB had higher proportions changing or initiating treatment after DST results (p = 0.01) and favorableoutcomes (p = 0.07).Conclusions: This study shows a high rate of empirical treatment and long delay for DST results. Strategies tospeed up the detection and early treatment of drug resistant TB should be prioritized.

A modelling framework to support the selection and implementation of new tuberculosis diagnostic tools.

Efforts to stimulate technological innovation in the diagnosis of tuberculosis (TB) have resulted in the recent introduction of several novel diagnostic tools. As these products come to market, policy makers must make difficult decisions about which of the available tools to implement. This choice should depend not only on the test characteristics (e.g., sensitivity and specificity) of the tools, but also on how they will be used within the existing health care infrastructure. Accordingly, policy makers choosing between diagnostic strategies must decide: 1) What is the best combination of tools to select? 2)Who should be tested with the new tools? and 3)Will these tools complement or replace existing diagnostics? The best choice of diagnostic strategy will likely vary between settings with different epidemiology (e.g., levels of TB incidence, human immunodeficiency virus co-infection and drug-resistant TB) and structural and resource constraints (e.g., existing diagnostic pathways, human resources and laboratory capacity). We propose a joint modelling framework that includes a tuberculosis (TB) transmission component (a dynamic epidemiological model) and a health system component (an operational systems model) to support diagnostic strategy decisions. This modelling approach captures the complex feedback loops in this system: new diagnostic strategies alter the demands on and performance of health systems that impact TB transmission dynamics which, in turn, result in further changes to demands on the health system. We demonstrate the use of a simplified model to support the rational choice of a diagnostic strategy based on health systems requirements, patient outcomes and population-level TB impact.

Making innovations accessible to the poor through implementation research.

Within countries, poorer populations have greater health needs and less access to good medical care than better-off populations. This is particularly true for tuberculosis (TB), the archetypal disease of poverty. Innovations also tend to become available to better-off populations well before they become available to those who need them the most. In a new era of innovations for TB diagnosis and treatment, it is increasingly important not only to be sure that these innovations can work in terms of accuracy and efficacy, but also that they will work, especially for the poor. We argue that after an innovation or a group of innovations has been endorsed, based on demonstrated accuracy and/or efficacy, introduction into routine practice should proceed through implementation by research. Cluster-randomised pragmatic trials are suited to this approach, and permit the prospective collection of evidence needed for full impact assessment according to a previously published framework. The novel approach of linking transmission modelling with operational modelling provides a methodology for expanding and enhancing the range of evidence, and can be used alongside evidence from pragmatic implementation trials. This evidence from routine practice should then be used to ensure that innovations in TB control are used for positive action for all, and particularly the poor.