Diagnostic Prediction Model for Tuberculous Meningitis: An Individual Participant Data Meta-Analysis.

No accurate and rapid diagnostic test exists for tuberculous meningitis (TBM), leading to delayed diagnosis. We leveraged data from multiple studies to improve the predictive performance of diagnostic models across different populations, settings, and subgroups to develop a new predictive tool for TBM diagnosis. We conducted a systematic review to analyze eligible datasets with individual-level participant data (IPD). We imputed missing data and explored three approaches: stepwise logistic regression, classification and regression tree (CART), and random forest regression. We evaluated performance using calibration plots and C-statistics via internal-external cross-validation. We included 3,761 individual participants from 14 studies and nine countries. A total of 1,240 (33%) participants had "definite" (30%) or "probable" (3%) TBM by case definition. Important predictive variables included cerebrospinal fluid (CSF) glucose, blood glucose, CSF white cell count, CSF differential, cryptococcal antigen, HIV status, and fever presence. Internal validation showed that performance varied considerably between IPD datasets with C-statistic values between 0.60 and 0.89. In external validation, CART performed the worst (C = 0.82), and logistic regression and random forest had the same accuracy (C = 0.91). We developed a mobile app for TBM clinical prediction that accounted for heterogeneity and improved diagnostic performance (https://tbmcalc.github.io/tbmcalc). Further external validation is needed.

Pneumocystis jirovecii pneumonia in people living with HIV: a review.

Pneumocystis jirovecii is a ubiquitous opportunistic fungus that can cause life-threatening pneumonia. People with HIV (PWH) who have low CD4 counts are one of the populations at the greatest risk of Pneumocystis jirovecii pneumonia (PCP). While guidelines have approached the diagnosis, prophylaxis, and management of PCP, the numerous studies of PCP in PWH are dominated by the 1980s and 1990s. As such, most studies have included younger male populations, despite PCP affecting both sexes and a broad age range. Many studies have been small and observational in nature, with an overall lack of randomized controlled trials. In many jurisdictions, and especially in low- and middle-income countries, the diagnosis can be challenging due to lack of access to advanced and/or invasive diagnostics. Worldwide, most patients will be treated with 21 days of high-dose trimethoprim sulfamethoxazole, although both the dose and the duration are primarily based on historical practice. Whether treatment with a lower dose is as effective and less toxic is gaining interest based on observational studies. Similarly, a 21-day tapering regimen of prednisone is used for patients with more severe disease, yet other doses, other steroids, or shorter durations of treatment with corticosteroids have not been evaluated. Now with the widespread availability of antiretroviral therapy, improved and less invasive PCP diagnostic techniques, and interest in novel treatment strategies, this review consolidates the scientific body of literature on the diagnosis and management of PCP in PWH, as well as identifies areas in need of more study and thoughtfully designed clinical trials.

Guidelines for Diagnosis and Management of Infective Endocarditis in Adults: A WikiGuidelines Group Consensus Statement.

Importance: Practice guidelines often provide recommendations in which the strength of the recommendation is dissociated from the quality of the evidence. Objective: To create a clinical guideline for the diagnosis and management of adult bacterial infective endocarditis (IE) that addresses the gap between the evidence and recommendation strength. Evidence Review: This consensus statement and systematic review applied an approach previously established by the WikiGuidelines Group to construct collaborative clinical guidelines. In April 2022 a call to new and existing members was released electronically (social media and email) for the next WikiGuidelines topic, and subsequently, topics and questions related to the diagnosis and management of adult bacterial IE were crowdsourced and prioritized by vote. For each topic, PubMed literature searches were conducted including all years and languages. Evidence was reported according to the WikiGuidelines charter: clear recommendations were established only when reproducible, prospective, controlled studies provided hypothesis-confirming evidence. In the absence of such data, clinical reviews were crafted discussing the risks and benefits of different approaches. Findings: A total of 51 members from 10 countries reviewed 587 articles and submitted information relevant to 4 sections: establishing the diagnosis of IE (9 questions); multidisciplinary IE teams (1 question); prophylaxis (2 questions); and treatment (5 questions). Of 17 unique questions, a clear recommendation could only be provided for 1 question: 3 randomized clinical trials have established that oral transitional therapy is at least as effective as intravenous (IV)-only therapy for the treatment of IE. Clinical reviews were generated for the remaining questions. Conclusions and Relevance: In this consensus statement that applied the WikiGuideline method for clinical guideline development, oral transitional therapy was at least as effective as IV-only therapy for the treatment of IE. Several randomized clinical trials are underway to inform other areas of practice, and further research is needed.

Clinical utility of WHO-recommended screening tools and development and validation of novel clinical prediction models for pulmonary tuberculosis screening among outpatients living with HIV: an individual participant data meta-analysis.

BACKGROUND: The World Health Organization (WHO) recommends that outpatient people living with HIV (PLHIV) undergo tuberculosis screening with the WHO four-symptom screen (W4SS) or C-reactive protein (CRP) (5 mg·L-1 cut-off) followed by confirmatory testing if screen positive. We conducted an individual participant data meta-analysis to determine the performance of WHO-recommended screening tools and two newly developed clinical prediction models (CPMs). METHODS: Following a systematic review, we identified studies that recruited adult outpatient PLHIV irrespective of tuberculosis signs and symptoms or with a positive W4SS, evaluated CRP and collected sputum for culture. We used logistic regression to develop an extended CPM (which included CRP and other predictors) and a CRP-only CPM. We used internal-external cross-validation to evaluate performance. RESULTS: We pooled data from eight cohorts (n=4315 participants). The extended CPM had excellent discrimination (C-statistic 0.81); the CRP-only CPM had similar discrimination. The C-statistics for WHO-recommended tools were lower. Both CPMs had equivalent or higher net benefit compared with the WHO-recommended tools. Compared with both CPMs, CRP (5 mg·L-1 cut-off) had equivalent net benefit across a clinically useful range of threshold probabilities, while the W4SS had a lower net benefit. The W4SS would capture 91% of tuberculosis cases and require confirmatory testing for 78% of participants. CRP (5 mg·L-1 cut-off), the extended CPM (4.2% threshold) and the CRP-only CPM (3.6% threshold) would capture similar percentages of cases but reduce confirmatory tests required by 24, 27 and 36%, respectively. CONCLUSIONS: CRP sets the standard for tuberculosis screening among outpatient PLHIV. The choice between using CRP at 5 mg·L-1 cut-off or in a CPM depends on available resources.

Diagnostic options for pulmonary fungal diseases in Africa.

Background: Fungal lung diseases are global in distribution and require specific tests for diagnosis. We report a survey of diagnostic service provision in Africa. Methods: A written questionnaire was followed by a video conference call with each respondent(s) and external validation. To disseminate the questionnaire, a snowball sample was used. Results: Data were successfully collected from 50 of 51 African countries with populations >1 million. The questionnaire was completed by respondents affiliated with 72 health facilities. Of these 72 respondents, 33 (45.8%) reported data for the whole country while others reported data for a specific region/province within their country. In the public sector, chest X-ray and computed tomography are performed often in 49 countries (98%) and occasionally in 37 countries (74%), and less often in the private sector. Bronchoscopy and spirometry were done often in 28 countries (56%) and occasionally in 18 countries (36%) in the tertiary health facilities of public sector. The most conducted laboratory diagnostic assay was fungal culture (often or occasionally) in 29 countries (58%). In collaboration with the Africa Centre for Disease Control and Prevention, regional webinars and individual country profiles provided further data validation. Conclusion: This survey has found a huge disparity of diagnostic test capability across the African continent. Some good examples of good diagnostic provision and very high-quality care were seen, but this was unusual. The unavailability of essential testing such as spirometry was noted, which has a high impact in the diagnosis of lung diseases. It is important for countries to implement tests based on the World Health Organization Essential Diagnostics List.

The therapeutic threshold in clinical decision-making for TB.

Because TB control is still hampered by the limitations of diagnostic tools, diagnostic uncertainty is common. The decision to offer treatment is based on clinical decision-making. The therapeutic threshold, test threshold and test-treatment threshold can guide in making these decisions. This review summarizes the literature on methods to estimate the therapeutic threshold that have been applied for TB. Only five studies estimated the threshold for the diagnosis of TB. The therapeutic threshold can be estimated by prescriptive methods, based on calculations, and by descriptive methods, deriving the threshold from observing clinical practice. Test and test-treatment thresholds can be calculated using the therapeutic threshold and the characteristics of an available diagnostic test. Estimates of the therapeutic threshold for pulmonary TB from intuitive descriptive approaches (20%-50%) are higher than theoretical prescriptive calculations (2%-3%). In conclusion, estimates of the therapeutic threshold for pulmonary TB depend on the method used. Other methods exist within the field of decision-making that have yet to be implemented or adapted as tools to estimate the TB therapeutic threshold. Because clinical decision-making is a core element of TB management, it is necessary to find a new, clinician-friendly way to unbiasedly estimate context-specific, agreed upon therapeutic thresholds.

Adherence to a care bundle for Staphylococcus aureus bacteraemia: A retrospective cohort study.

Background: Staphylococcus aureus bacteraemia is associated with high hospital mortality. Improvements in outcome have been described with standardised bundles of care. Objectives: To study the adherence of a standardised bundle of care (BOC) recommendations using a consultation pro forma, for all patients admitted with S. aureus bacteraemia to Groote Schuur Hospital over a year. The study further aimed to describe the 90-day mortality in these patients and to assess for an association between adherence to the bundle of care and outcome. Method: A retrospective audit of all unsolicited infectious disease consultations for patients with S. aureus bacteraemia admitted to Groote Schuur Hospital during 2018. Adherence to recommendations of a standard bundle of care was audited. Results: A total of 86 patients were included in the study: 61 (71%) with hospital-associated infection and 25 (29%) with community-associated infection. Over 80% of adherence to treatment recommendations was achieved regarding antibiotic (including vancomycin) usage, source control and use of echocardiography as required. In-hospital mortality was 16%, while the overall 90-day mortality was 18%, with only age as an independent predictor of mortality. No association between adherence to the bundle of care and outcome was found. Conclusion: Adherence to a simple, structured bundle of care was good when using standardised pro forma as communication tools for advice and a structured antibiotic chart for vancomycin administration. Although adherence was not associated with outcome, the overall mortality for S. aureus bacteraemia was improving in the institution under study. Contribution: Our findings support feasibility and ongoing use of bundles of care for S. aureus bacteraemia in similar settings.

Outcomes of a model integrating tuberculosis testing into COVID-19 services in South Africa.

The coronavirus disease 2019 (COVID-19) pandemic led to a reordering of healthcare priorities. Health resources were turned to the screening and diagnosis of COVID-19, leading to a reduction in tuberculosis (TB) testing and treatment initiation. An innovative model that integrated TB and COVID-19 services was adopted at primary care facilities in Johannesburg Health District, Gauteng. This short report illustrates results from this model's implementation in two facilities. Patients were screened for COVID-19 at a single point of entry and separated according to screening result. Self-reported human immunodeficiency virus (HIV) status, symptom, and symptom duration were then used to determine TB risk amongst those screening positive for COVID-19. Data from clinical records were extracted. Approximately 9% of patients with a positive symptom screen (n = 76) were sent for a TB test and 84% were sent for a COVID-19 test. Amongst those sent for a TB test, 8% (n = 6) had TB detected, and amongst those sent for a COVID-19 test, 18% (n = 128) were positive. Amongst those with COVID-19-related symptoms, 15% (n = 130) presented with a cough or fever and were known HIV positive and 121 (93%) of these were sent for a COVID-19 test and 31 (24%) were sent for a TB test. Given the HIV prevalence and symptoms in our study, our results show lower-than-expected TB tests conducted.Contribution: Our study documents the outcomes of an innovative way to combine operational workflows for TB and COVID-19. This provides a starting point for countries seeking to integrate TB and COVID-19 screening and testing.

Consensus study on the health system and patient-related barriers for lung cancer management in South Africa.

BACKGROUND: Lung cancer is the highest incident cancer globally and is associated with significant morbidity and mortality particularly if identified at a late stage. Poor patient outcomes in low- and middle-income countries (LMIC's) might reflect contextual patient and health system constraints at multiple levels, that act as barriers to prevention, disease recognition, diagnosis, and treatment. Lung cancer screening, even for high-risk patients, is not available in the public health sector in South Africa (SA), where the current HIV and tuberculosis (TB) epidemics often take precedence. Yet, there has been no formal assessment of the individual and health-system related barriers that may delay patients with lung cancer from seeking and accessing help within the public health care system and receiving the appropriate and effective diagnosis and treatment. This study aimed to derive consensus from health-system stakeholders in the urban Gauteng Province of SA on the most important challenges faced by the health services and patients in achieving optimum lung cancer management and to identify potential solutions. METHODS: The study was undertaken among 27 participant stakeholders representing clinical managers, clinicians, opinion leaders from the public health sector and non-governmental organisation (NGO) representatives. The study compromised two components: consensus and engagement. For the consensus component, the Delphi Technique was employed with open-ended questions and item ranking from five rounds of consensus-seeking, to achieve collective agreement on the most important challenges faced by patients and the health services in achieving optimal lung cancer management. For the engagement component, the Nominal Group Technique was used to articulate ideas and reach an agreement on the group's recommendations for solution strategies and approaches. RESULTS: Public health sector stakeholders suggested that a lack of knowledge and awareness of lung cancer, and the apparent stigma associated with the disease and its risk factors, as well as symptoms and signs, are critical to treatment delay. Furthermore, delays in up-referral of patients with suspected lung cancer from district health care level were attributed to inadequate knowledge arising from a lack of in-service training of nurses and doctors regarding oncologic symptoms, risk factors, need for further investigation, interpretation of x-rays and available treatments. At a tertiary level, participants suggested that insufficient availability of specialised diagnostic resources (imaging, cytological and pathological services including biomolecular assessment of lung cancer), theatres, cardiothoracic surgeons, and appropriate therapeutic modalities (chemotherapeutic agents and radiation oncology) are the main barriers to the provision of optimal care. It was suggested that a primary prevention programme initiated by the government that involves private-public partnerships may improve lung cancer management nationally. CONCLUSIONS: Considerable barriers to the early identification and treatment of lung cancer exist. Finding solutions to overcome both individual and health-system level obstacles to lung cancer screening and management are vital to facilitate early identification and treatment, and to improve survival. Furthermore, research on inexpensive biomarkers for asymptomatic disease detection, the introduction of diagnostic imaging tools that utilise artificial intelligence to compensate for inadequate human resources and improving clinical integration across all levels of the healthcare system are essential.

Diagnostic Accuracy of the INSHI Consensus Case Definition for the Diagnosis of Paradoxical Tuberculosis-IRIS.

BACKGROUND: The diagnosis of paradoxical tuberculosis-associated immune reconstitution inflammatory syndrome (TB-IRIS) relies on characteristic clinical features synthesized as the International Network for the Study of HIV-associated IRIS (INSHI) case definition. There is no confirmatory laboratory test. SETTING: Site B HIV-TB clinic in Khayelitsha, Cape Town, South Africa. METHODS: Using data of participants with HIV-associated tuberculosis starting antiretroviral treatment from a prospective trial evaluating prednisone for TB-IRIS prevention, we applied latent class analysis to model a gold standard for TB-IRIS. The model-predicted probability of TB-IRIS for each participant was used to assess the performance of the INSHI case definition and compare its diagnostic accuracy with several adapted case definitions. RESULTS: Data for this analysis were complete for 217 participants; 41% developed TB-IRIS. Our latent class model included the following parameters: respiratory symptoms; night sweats; INSHI major criteria 1, 2, and 4; maximum C-reactive protein >90 mg/L; maximum heart rate >120/min; maximum temperature >37.7°C; and preantiretroviral therapy CD4 count <50 cells/µL. The model estimated a TB-IRIS incidence of 43% and had optimal goodness of fit (χ2 = 337, P = 1.0). The INSHI case definition displayed a sensitivity of 0.77 and a specificity of 0.86. Replacing all the minor INSHI criteria with objectives measures (C-reactive protein elevation, fever, and/or tachycardia) resulted in a definition with better diagnostic accuracy, with a sensitivity of 0.89 and a specificity of 0.88. CONCLUSION: The INSHI case definition identifies TB-IRIS with reasonable accuracy. Amending the case definition by replacing INSHI minor criteria with objective variables improved sensitivity without loss of specificity.

Decision-making in the diagnosis of tuberculous meningitis.

Tuberculous meningitis (TBM) is the most devastating form of tuberculosis (TB) but diagnosis is difficult and delays in initiating therapy increase mortality. All currently available tests are imperfect; culture of Mycobacterium tuberculosis from the cerebrospinal fluid (CSF) is considered the most accurate test but is often negative, even when disease is present, and takes too long to be useful for immediate decision making. Rapid tests that are frequently used are conventional Ziehl-Neelsen staining and nucleic acid amplification tests such as Xpert MTB/RIF and Xpert MTB/RIF Ultra. While positive results will often confirm the diagnosis, negative tests frequently provide insufficient evidence to withhold therapy. The conventional diagnostic approach is to determine the probability of TBM using experience and intuition, based on prevalence of TB, history, examination, analysis of basic blood and CSF parameters, imaging, and rapid test results. Treatment decisions may therefore be both variable and inaccurate, depend on the experience of the clinician, and requests for tests may be inappropriate. In this article we discuss the use of Bayes' theorem and the threshold model of decision making as ways to improve testing and treatment decisions in TBM. Bayes' theorem describes the process of converting the pre-test probability of disease to the post-test probability based on test results and the threshold model guides clinicians to make rational test and treatment decisions. We discuss the advantages and limitations of using these methods and suggest that new diagnostic strategies should ultimately be tested in randomised trials.

CheXaid: deep learning assistance for physician diagnosis of tuberculosis using chest x-rays in patients with HIV.

Tuberculosis (TB) is the leading cause of preventable death in HIV-positive patients, and yet often remains undiagnosed and untreated. Chest x-ray is often used to assist in diagnosis, yet this presents additional challenges due to atypical radiographic presentation and radiologist shortages in regions where co-infection is most common. We developed a deep learning algorithm to diagnose TB using clinical information and chest x-ray images from 677 HIV-positive patients with suspected TB from two hospitals in South Africa. We then sought to determine whether the algorithm could assist clinicians in the diagnosis of TB in HIV-positive patients as a web-based diagnostic assistant. Use of the algorithm resulted in a modest but statistically significant improvement in clinician accuracy (p = 0.002), increasing the mean clinician accuracy from 0.60 (95% CI 0.57, 0.63) without assistance to 0.65 (95% CI 0.60, 0.70) with assistance. However, the accuracy of assisted clinicians was significantly lower (p < 0.001) than that of the stand-alone algorithm, which had an accuracy of 0.79 (95% CI 0.77, 0.82) on the same unseen test cases. These results suggest that deep learning assistance may improve clinician accuracy in TB diagnosis using chest x-rays, which would be valuable in settings with a high burden of HIV/TB co-infection. Moreover, the high accuracy of the stand-alone algorithm suggests a potential value particularly in settings with a scarcity of radiological expertise.

A Clinical Prediction Score Including Trial of Antibiotics and C-Reactive Protein to Improve the Diagnosis of Tuberculosis in Ambulatory People With HIV.

BACKGROUND: The use of a "trial of antibiotics" as empiric therapy for bacterial pneumonia as a diagnostic tool for tuberculosis in people with HIV (PWH) was removed from World Health Organization (WHO) recommendations in 2007, based on expert opinion. Current guidelines recommend antibiotics only after 2 Xpert MTB/RIF tests (if available), chest x-ray, and clinical assessment have suggested that tuberculosis is unlikely. Despite this, a "trial of antibiotics" remains common in algorithms in low-resource settings, but its value is uncertain. C-reactive protein (CRP), which has been proposed as a "rule-out" test for tuberculosis, may be an objective marker of response to antibiotics. METHODS: We performed a passive case-finding cohort study of adult PWH with a positive WHO symptom screen. All participants received antibiotics at first visit according to the local protocol and were reviewed to ascertain clinical response. Point-of-care CRP was measured at both visits. All patients had sputum tested with Xpert MTB/RIF Ultra (Ultra), and the reference standard was based on 2 sputum mycobacterial cultures. We explored multivariable prediction models (MPM) for tuberculosis based on 1 or 2 visits. RESULTS: Seventy-five of 207 patients (36%) had confirmed tuberculosis. Clinical response to antibiotics after 2 days was a good predictor of disease. An MPM based on 2 visits, without CRP, had acceptable discrimination (c-statistic, 0.75) and calibration (goodness-of-fit P = .07). Addition of CRP after antibiotics improved the model moderately (c-statistic, 0.78). CRP at first visit was not an independent predictor of tuberculosis. CONCLUSIONS: In adult PWH seeking care for symptoms suggestive of tuberculosis, lack of response to antibiotics is a strong predictor of disease and is likely to be useful, particularly when access to Ultra is limited. CRP adds value when measured after antibiotics but is of limited value at first visit.

South African Society of Clinical Microbiology Clostridioides difficile infection diagnosis, management and infection prevention and control guideline.

Clostridioides difficile infection (CDI) is a problem in both developed and developing countries and is a common hospital-acquired infection. This guideline provides evidence-based practical recommendations for South Africa and other developing countries. The scope of the guideline includes CDI diagnostic approaches; adult, paediatric and special populations treatment options; and surveillance and infection prevention and control recommendations.

Low-dose ritonavir-boosted darunavir once daily versus ritonavir-boosted lopinavir for participants with less than 50 HIV RNA copies per mL (WRHI 052): a randomised, open-label, phase 3, non-inferiority trial.

BACKGROUND: Pilot studies suggest that ritonavir-boosted darunavir could show high efficacy at doses below those currently approved. We investigated whether switch to 400 mg of darunavir boosted with 100 mg ritonavir once daily could show equivalent efficacy to continuation of ritonavir-boosted lopinavir (a protease inhibitor commonly used in low-income and middle-income countries) for individuals with HIV RNA suppression. METHODS: In the WRHI 052 study, a randomised, parallel-group, open-label, non-inferiority phase 3 trial, adults who were HIV-1 positive were enrolled in Charlotte Maxeke Johannesburg Academic Hospital, Johannesburg, South Africa. Eligible participants were 18 years or older, who tolerated ritonavir-boosted lopinavir in combination with two nucleoside analogues (standard of care) for 6 months or more, and had plasma HIV-1 RNA of less than 50 copies per mL within 60 days of enrolment. We randomly assigned participants (1:1), using a computer-generated randomisation plan, to switch to darunavir (400 mg) boosted with ritonavir (100 mg) once daily or remain on ritonavir-boosted lopinavir (800 mg [plus 200 mg ritonavir]), with nucleoside analogues left unchanged. The primary endpoint was the proportion of patients with less than 50 HIV-1 RNA copies per mL at week 48 (US Food and Drug Administration snapshot algorithm; non-inferiority margin -4%). Primary and safety analyses included participants receiving at least one dose of darunavir boosted with ritonavir. This trial is registered with ClinicalTrials.gov, number NCT02671383. FINDINGS: Between June 30, 2016, and June 15, 2017, 148 participants were assigned to ritonavir-boosted darunavir 400 mg and 152 continued on their lopinavir-containing regimen. Four (3%) patients in the darunavir group and three (2%) in the lopinavir group discontinued before week 48. At week 48, darunavir was non-inferior to lopinavir for the primary outcome (142 [96%] of 148 participants on darunavir had <50 HIV-1 RNA copies per mL vs 143 [94%] of 152 participants on lopinavir; difference 1·9% [95% CI -3·4 to 7·3]), with a predefined margin of -4%. More participants taking darunavir (30 [20%] participants) had drug-related adverse events than those on lopinavir (eight [5%]), but the adverse events were generally asymptomatic and resolved when switching back to lopinavir. Elevated liver transaminase in three (1%; one symptomatic) darunavir participants led to study withdrawal; all transaminase elevations resolved on restarting lopinavir. INTERPRETATION: Low-dose ritonavir-boosted darunavir might be a safe and efficacious switch option to maintain HIV suppression for patients on lopinavir. However, an adequately powered and designed study in viraemic participants is needed. FUNDING: South African Medical Research Council, United States Agency for International Development, and US National Institute of Allergy and Infectious Diseases.

Knowledge, attitudes and perceptions of antibiotic use and resistance among patients in South Africa: A cross-sectional study.

BACKGROUND: Antibiotic resistance (ABR) is a global health crisis. We conducted a cross-sectional survey to describe South African patients' (n = 782) ABR knowledge, attitudes and perceptions (KAP), differences in KAP between public (n = 379, 48%) and private (n = 403, 52%) practice respondents and associations between attitudes, perceptions and knowledge scores. METHODS: Knowledge scores (15 questions) were placed into low (0% - 53%) and high (> 54%) categories (below and above overall mean). Comparisons were conducted using chi-squared and t-tests. RESULTS: Of all respondents, 72% believed it was the human body that becomes resistant to antibiotics, 66% stated that antibiotics are good for treating viruses and 25% of patients believed that people should be given antibiotics on demand. Mean knowledge scores were lower in public sector respondents (public 45%, s.d. 15%; private 60%, s.d. 30%; p ≤ 0.001). Public practice patients with high knowledge scores were more likely to report both negative KAP (antibiotic prescriptions justify doctors' visits, scientists will discover new antibiotics) and protective KAP (finishing a course of antibiotics, antibiotics do not treat all illnesses). Private practice patients with high knowledge scores were marginally less likely to report negative KAP (wanting antibiotics after long illnesses or when very sick) and more likely to report protective KAP (antibiotics have side effects and are a strong treatment). CONCLUSION: Our study shows differences in KAP by practice type and that greater knowledge increases the likelihood of protective attitudes, perceptions and behaviours.

The diagnosis of tuberculous meningitis in adults and adolescents: protocol for a systematic review and individual patient data meta-analysis to inform a multivariable prediction model.

Background: Tuberculous meningitis (TBM) is the most lethal and disabling form of tuberculosis. Delayed diagnosis and treatment, which is a risk factor for poor outcome, is caused in part by lack of availability of diagnostic tests that are both rapid and accurate. Several attempts have been made to develop clinical scoring systems to fill this gap, but none have performed sufficiently well to be broadly implemented. We aim to identify and validate a set of clinical predictors that accurately classify TBM using individual patient data (IPD) from published studies. Methods: We will perform a systematic review and obtain IPD from studies published from the year 1990 which undertook diagnostic testing for TBM in adolescents or adults using at least one of, microscopy for acid-fast bacilli, commercial nucleic acid amplification test for Mycobacterium tuberculosis or mycobacterial culture of cerebrospinal fluid.  Clinical data that have previously been shown to be associated with TBM, and can inform the final diagnosis, will be requested. The data-set will be divided into training and test/validation data-sets for model building. A predictive logistic model will be built using a training set with patients with definite TBM and no TBM. Should it be warranted, factor analysis may be employed, depending on evidence for multicollinearity or the case for including latent variables in the model. Discussion: We will systematically identify and extract key clinical parameters associated with TBM from published studies and use a 'big data' approach to develop and validate a clinical prediction model with enhanced generalisability. The final model will be made available through a smartphone application. Further work will be external validation of the model and test of efficacy in a randomised controlled trial.