On the Origin of Sensitivity and Specificity.

Although it is commonly said that the notions of sensitivity and specificity were first defined by Jacob Yerushalmy in 1947, the sensitivity and specificity of diagnostic tests have been assessed as far back as the early 1900s. These notions share a common origin with the development of serology. They were originally immunologic concepts, closely associated with the development of complement fixation reactions for syphilis. Here, the authors trace how immunologic sensitivity and specificity were transformed into diagnostic sensitivity and specificity. By relocating the origins of these concepts to the early 20th century, they highlight how these origins were bound to then-commonplace assumptions about specific infectious disease entities.

QUADAS-C: A Tool for Assessing Risk of Bias in Comparative Diagnostic Accuracy Studies.

Comparative diagnostic test accuracy studies assess and compare the accuracy of 2 or more tests in the same study. Although these studies have the potential to yield reliable evidence regarding comparative accuracy, shortcomings in the design, conduct, and analysis may bias their results. The currently recommended quality assessment tool for diagnostic test accuracy studies, QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies-2), is not designed for the assessment of test comparisons. The QUADAS-C (Quality Assessment of Diagnostic Accuracy Studies-Comparative) tool was developed as an extension of QUADAS-2 to assess the risk of bias in comparative diagnostic test accuracy studies. Through a 4-round Delphi study involving 24 international experts in test evaluation and a face-to-face consensus meeting, an initial version of the tool was developed that was revised and finalized following a pilot study among potential users. The QUADAS-C tool retains the same 4-domain structure of QUADAS-2 (Patient Selection, Index Test, Reference Standard, and Flow and Timing) and comprises additional questions to each QUADAS-2 domain. A risk-of-bias judgment for comparative accuracy requires a risk-of-bias judgment for the accuracy of each test (resulting from QUADAS-2) and additional criteria specific to test comparisons. Examples of such additional criteria include whether participants either received all index tests or were randomly assigned to index tests, and whether index tests were interpreted with blinding to the results of other index tests. The QUADAS-C tool will be useful for systematic reviews of diagnostic test accuracy addressing comparative questions. Furthermore, researchers may use this tool to identify and avoid risk of bias when designing a comparative diagnostic test accuracy study.

Symptom scores in the diagnosis of pediatric cow's milk protein allergy: A systematic review.

BACKGROUND: Cow's milk protein allergy (CMPA) is an immune-mediated allergic response to proteins in milk that is common in infants. Broad CMPA symptoms make diagnosis a challenge, particularly in primary care. Symptom scores may improve a clinician's awareness of symptoms, indicating a need for further testing. This systematic review examined the development and evaluation of such symptom scores for use in infants. METHODS: CENTRAL, MEDLINE, EMBASE and CINAHL databases were searched from inception to 3 December 2019 (Updated 14 November 2020) for diagnostic accuracy studies, randomised controlled trials, observational studies, economic evaluations, qualitative studies and studies reporting development of the tools. Data were not suitable for meta-analysis due to clinical and methodological heterogeneity, so were narratively synthesised. RESULTS: We found two symptom scores evaluated in one and fourteen studies, respectively. Estimated sensitivity and specificity ranged from 37% to 98% and 38% to 93%. The evaluations of each tool were at high risk of bias or failed to address issues such as clinical and cost-effectiveness. CONCLUSIONS: Estimates of accuracy of symptom scores for CMPA offered so far should be interpreted cautiously. Rigorous, conflict-free research based on well-defined roles for the tools is urgently required.

Development, validation and effectiveness of diagnostic prediction tools for colorectal cancer in primary care: a systematic review.

BACKGROUND: Tools based on diagnostic prediction models are available to help general practitioners (GP) diagnose colorectal cancer. It is unclear how well they perform and whether they lead to increased or quicker diagnoses and ultimately impact on patient quality of life and/or survival. The aim of this systematic review is to evaluate the development, validation, effectiveness, and cost-effectiveness, of cancer diagnostic tools for colorectal cancer in primary care. METHODS: Electronic databases including Medline and Web of Science were searched in May 2017 (updated October 2019). Two reviewers independently screened titles, abstracts and full-texts. Studies were included if they reported the development, validation or accuracy of a prediction model, or assessed the effectiveness or cost-effectiveness of diagnostic tools based on prediction models to aid GP decision-making for symptomatic patients presenting with features potentially indicative of colorectal cancer. Data extraction and risk of bias were completed by one reviewer and checked by a second. A narrative synthesis was conducted. RESULTS: Eleven thousand one hundred thirteen records were screened and 23 studies met the inclusion criteria. Twenty-studies reported on the development, validation and/or accuracy of 13 prediction models: eight for colorectal cancer, five for cancer areas/types that include colorectal cancer. The Qcancer models were generally the best performing. Three impact studies met the inclusion criteria. Two (an RCT and a pre-post study) assessed tools based on the RAT prediction model. The third study looked at the impact of GP practices having access to RAT or Qcancer. Although the pre-post study reported a positive impact of the tools on outcomes, the results of the RCT and cross-sectional survey found no evidence that use of, or access to, the tools was associated with better outcomes. No study evaluated cost effectiveness. CONCLUSIONS: Many prediction models have been developed but none have been fully validated. Evidence demonstrating improved patient outcome of introducing the tools is the main deficiency and is essential given the imperfect classification achieved by all tools. This need is emphasised by the equivocal results of the small number of impact studies done so far.

Accuracy and precision analysis for a biophotonic assay of C-reactive protein.

A multiplexed biophotonic assay platform has been developed using the localised particle plasmon in gold nanoparticles assembled in an array and functionalised for two assays: total IgG and C-reactive protein (CRP). A protein A/G (PAG) assay, calibrated with a NIST reference material, shows a maximum surface coverage of θmax = 7.13 ± 0.19 mRIU, equivalent to 1.5 ng mm-2 of F(ab)-presenting antibody. The CRP capture antibody has an equivalent surface binding density of θmax = 2.95 ± 0.41 mRIU indicating a 41% capture antibody availability. Free PAG binding to the functionalised anti-CRP surface shows that only 47 ± 3% of CRP capture antibodies are correctly presenting Fab regions for antigen capture. The accuracy and precision of the CRP sensor assay was assessed with 54 blood samples containing spiked CRP in the range 2-160 mg L-1. The mean accuracy was 0.42 mg L-1 with Confidence Interval (CI) at 95% from -14.7 to 13.8 mg L-1 and the precision had a Coefficient of Variation (CV) of 10.6% with 95% CI 0.9%-20.2%. These biophotonic platform performance metrics indicate a CRP assay with 2-160 mg L-1 dynamic range, performed in 8 minutes from 5 µL of whole blood without sample preparation.

Real-world evaluation of a novel technology for quantitative simultaneous antibody detection against multiple SARS-CoV-2 antigens in a cohort of patients presenting with COVID-19 syndrome.

An evaluation of a rapid portable gold-nanotechnology measuring SARS-CoV-2 IgM, IgA and IgG antibody concentrations against spike 1 (S1), spike 2 (S) and nucleocapsid (N) was conducted using serum samples from 74 patients tested for SARS-CoV-2 RNA on admission to hospital, and 47 historical control patients from March 2019. 59 patients were RNA(+) and 15 were RNA(-). A serum (±) classification was derived for all three antigens and a quantitative serological profile was obtained. Serum(+) was identified in 30% (95% CI 11-48) of initially RNA(-) patients, in 36% (95% CI 17-54) of RNA(+) patients before 10 days, 77% (95% CI 67-87) between 10 and 20 days and 95% (95% CI 86-100) after 21 days. The patient-level diagnostic accuracy relative to RNA(±) after 10 days displayed 88% sensitivity (95% CI 75-95) and 75% specificity (95% CI 22-99), although specificity compared with historical controls was 100% (95%CI 91-100). This study provides robust support for further evaluation and validation of this novel technology in a clinical setting and highlights challenges inherent in assessment of serological tests for an emerging disease such as COVID-19.

A rapid and quantitative technique for assessing IgG monomeric purity, calibrated with the NISTmAb reference material.

The fraction of intact monomer in a sample (moles/moles), the monomeric purity, is measured as a quality control in therapeutic monoclonal antibodies but is often unknown in research samples and remains a major source of variation in quantitative antibody-based techniques such as immunoassay development. Here, we describe a novel multiplex technique for estimating the monomeric purity and antigen affinity of research grade antibody samples. Light scattering was used to simultaneously observe the mass of antibody binding to biosensor surfaces functionalised with antigen (revealing Fab binding kinetics) or protein A/G (PAG). Initial estimates of monomeric purity in 7 antibody samples including a therapeutic infliximab biosimilar were estimated by observing a mass deficit on the PAG surface compared to the NISTmAb standard of high monomeric purity. Monomeric purity estimates were improved in a second step by observing the mass of antigen binding to the mass of antibody on the PAG surface. The NISTmAb and infliximab biosimilar displayed tightly controlled stoichiometries for antigen binding of 1.31 ± 0.57 and 1.71 ± 0.16 (95% confidence interval)-within the theoretical limit of 1-2 antigens per antibody depending on avidity. The other antibodies in the panel displayed antigen binding stoichiometries in the range 0.06-1.15, attributed to lower monomeric purity. The monomeric purity estimates were verified by electrospray ionization mass spectrometry (ESI), the gold standard technique for structural characterization of antibodies. ESI data indicated that the NISTmAb and infliximab biosimilar samples had monomeric purity values of 93.5% and 94.7%, respectively, whilst the research grade samples were significantly lower (54-89%). Our results demonstrate rapid quality control testing for monomeric purity of antibody samples (< 15 min) which could improve the reproducibility of antibody-based experiments.

A UK nationwide prospective study of treatment change in MODY: genetic subtype and clinical characteristics predict optimal glycaemic control after discontinuing insulin and metformin.

AIMS/HYPOTHESIS: Treatment change following a genetic diagnosis of MODY is frequently indicated, but little is known about the factors predicting future treatment success. We therefore conducted the first prospective study to determine the impact of a genetic diagnosis on individuals with GCK-, HNF1A- or HNF4A-MODY in the UK, and to identify clinical characteristics predicting treatment success (i.e. HbA1c ≤58 mmol/mol [≤7.5%]) with the recommended treatment at 2 years. METHODS: This was an observational, prospective, non-selective study of individuals referred to the Exeter Molecular Genetic Laboratory for genetic testing from December 2010 to December 2012. Individuals from the UK with GCK- or HNF1A/HNF4A-MODY who were not on recommended treatment at the time of genetic diagnosis, and who were diagnosed below the age of 30 years and were currently aged less than 50 years, were eligible to participate. RESULTS: A total of 44 of 58 individuals (75.9%) changed treatment following their genetic diagnosis. Eight individuals diagnosed with GCK-MODY stopped all diabetes medication without experiencing any change in HbA1c (49.5 mmol/mol [6.6%] both before the genetic diagnosis and at a median of 1.25 years' follow-up without treatment, p = 0.88). A total of 36 of 49 individuals (73.5%) diagnosed with HNF1A/HNF4A-MODY changed treatment; however, of the 21 of these individuals who were being managed with diet or sulfonylurea alone at 2 years, only 13 (36.1% of the population that changed treatment) had an HbA1c ≤58 mmol/mol (≤7.5%). These individuals had a shorter diabetes duration (median 4.6 vs 18.1 years), lower HbA1c (58 vs 73 mmol/mol [7.5% vs 8.8%]) and lower BMI (median 24.2 vs 26.0 kg/m2) at the time of genetic diagnosis, compared with individuals (n = 23/36) with an HbA1c >58 mmol/mol (>7.5%) (or <58 mmol/mol [<7.5%] on additional treatment) at the 2 year follow-up. Overall, 64% (7/11) individuals with a diabetes duration of ≤11 years and an HbA1c of ≤69 mmol/mol (≤8.5%) at time of the genetic test achieved good glycaemic control (HbA1c ≤58 mmol/mol [≤7.5%]) with diet or sulfonylurea alone at 2 years, compared with no participants with a diabetes duration of >11 years and an HbA1c of >69 mmol/mol (>8.5%) at the time of genetic diagnosis. CONCLUSIONS/INTERPRETATION: In participants with GCK-MODY, treatment cessation was universally successful, with no change in HbA1c at follow-up. In those with HNF1A/HNF4A-MODY, a shorter diabetes duration, lower HbA1c and lower BMI at genetic diagnosis predicted successful treatment with sulfonylurea/diet alone, supporting the need for early genetic diagnosis and treatment change. Our study suggests that, in individuals with HNF1A/HNF4A-MODY with a longer duration of diabetes (>11 years) at time of genetic test, rather than ceasing current treatment, a sulfonylurea should be added to existing therapy, particularly in those who are overweight or obese and have a high HbA1c.

The simultaneous assessment of and relations between children's sympathetic and parasympathetic psychophysiology and their reactive and proactive aggression.

The goal of the current study was to examine the link between children's psychophysiology and aggression when both constructs were assessed simultaneously in scenarios designed to provide the opportunity to aggress for either a reactive reason or a proactive reason. Both sympathetic nervous system (SNS) activity (skin conductance) and parasympathetic nervous system (PNS) activity (respiratory sinus arrhythmia or RSA), as well as their interaction, were included as physiological measures. Participants were 35 5th-grade children who were placed in two virtual-peer scenarios; one scenario provided the opportunity to aggress in response to peer provocation (i.e., reactive aggression) and the other scenario provided the opportunity to aggress for instrumental gain (i.e., proactive aggression). Both skin conductance and RSA were assessed at the time that children were given the opportunity to aggress; this simultaneous assessment of psychophysiology and aggression allowed for an examination of in-the-moment relations between the two constructs. For the reactive scenario, RSA moderated the in-the-moment relation between skin conductance and aggression such that the association was positive at low RSA but negative at high RSA. For the proactive scenario, skin conductance negatively predicted aggression in-the-moment, and RSA positively predicted aggression in-the-moment, but their interaction was not a significant predictor of aggression. Theoretical implications for reactive and proactive aggression and underlying physiological processes are discussed.

Preferred Reporting Items for a Systematic Review and Meta-analysis of Diagnostic Test Accuracy Studies: The PRISMA-DTA Statement.

Importance: Systematic reviews of diagnostic test accuracy synthesize data from primary diagnostic studies that have evaluated the accuracy of 1 or more index tests against a reference standard, provide estimates of test performance, allow comparisons of the accuracy of different tests, and facilitate the identification of sources of variability in test accuracy. Objective: To develop the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) diagnostic test accuracy guideline as a stand-alone extension of the PRISMA statement. Modifications to the PRISMA statement reflect the specific requirements for reporting of systematic reviews and meta-analyses of diagnostic test accuracy studies and the abstracts for these reviews. Design: Established standards from the Enhancing the Quality and Transparency of Health Research (EQUATOR) Network were followed for the development of the guideline. The original PRISMA statement was used as a framework on which to modify and add items. A group of 24 multidisciplinary experts used a systematic review of articles on existing reporting guidelines and methods, a 3-round Delphi process, a consensus meeting, pilot testing, and iterative refinement to develop the PRISMA diagnostic test accuracy guideline. The final version of the PRISMA diagnostic test accuracy guideline checklist was approved by the group. Findings: The systematic review (produced 64 items) and the Delphi process (provided feedback on 7 proposed items; 1 item was later split into 2 items) identified 71 potentially relevant items for consideration. The Delphi process reduced these to 60 items that were discussed at the consensus meeting. Following the meeting, pilot testing and iterative feedback were used to generate the 27-item PRISMA diagnostic test accuracy checklist. To reflect specific or optimal contemporary systematic review methods for diagnostic test accuracy, 8 of the 27 original PRISMA items were left unchanged, 17 were modified, 2 were added, and 2 were omitted. Conclusions and Relevance: The 27-item PRISMA diagnostic test accuracy checklist provides specific guidance for reporting of systematic reviews. The PRISMA diagnostic test accuracy guideline can facilitate the transparent reporting of reviews, and may assist in the evaluation of validity and applicability, enhance replicability of reviews, and make the results from systematic reviews of diagnostic test accuracy studies more useful.

Intra-patient comparison of reduced-dose model-based iterative reconstruction with standard-dose adaptive statistical iterative reconstruction in the CT diagnosis and follow-up of urolithiasis.

OBJECTIVES: To evaluate the accuracy of reduced-dose CT scans reconstructed using a new generation of model-based iterative reconstruction (MBIR) in the imaging of urinary tract stone disease, compared with a standard-dose CT using 30% adaptive statistical iterative reconstruction. METHODS: This single-institution prospective study recruited 125 patients presenting either with acute renal colic or for follow-up of known urinary tract stones. They underwent two immediately consecutive scans, one at standard dose settings and one at the lowest dose (highest noise index) the scanner would allow. The reduced-dose scans were reconstructed using both ASIR 30% and MBIR algorithms and reviewed independently by two radiologists. Objective and subjective image quality measures as well as diagnostic data were obtained. RESULTS: The reduced-dose MBIR scan was 100% concordant with the reference standard for the assessment of ureteric stones. It was extremely accurate at identifying calculi of 3 mm and above. The algorithm allowed a dose reduction of 58% without any loss of scan quality. CONCLUSIONS: A reduced-dose CT scan using MBIR is accurate in acute imaging for renal colic symptoms and for urolithiasis follow-up and allows a significant reduction in dose. KEY POINTS: • MBIR allows reduced CT dose with similar diagnostic accuracy • MBIR outperforms ASIR when used for the reconstruction of reduced-dose scans • MBIR can be used to accurately assess stones 3 mm and above.

EXPLICIT: a feasibility study of remote expert elicitation in health technology assessment.

BACKGROUND: Expert opinion is often sought to complement available information needed to inform model-based economic evaluations in health technology assessments. In this context, we define expert elicitation as the process of encoding expert opinion on a quantity of interest, together with associated uncertainty, as a probability distribution. When availability for face-to-face expert elicitation with a facilitator is limited, elicitation can be conducted remotely, overcoming challenges of finding an appropriate time to meet the expert and allowing access to experts situated too far away for practical face-to-face sessions. However, distance elicitation is associated with reduced response rates and limited assistance for the expert during the elicitation session. The aim of this study was to inform the development of a remote elicitation tool by exploring the influence of mode of elicitation on elicited beliefs. METHODS: An Excel-based tool (EXPLICIT) was developed to assist the elicitation session, including the preparation of the expert and recording of their responses. General practitioners (GPs) were invited to provide expert opinion about population alcohol consumption behaviours. They were randomised to complete the elicitation by either a face-to-face meeting or email. EXPLICIT was used in the elicitation sessions for both arms. RESULTS: Fifteen GPs completed the elicitation session. Those conducted by email were longer than the face-to-face sessions (13 min 30 s vs 10 min 26 s, p = 0.1) and the email-elicited estimates contained less uncertainty. However, the resulting aggregated distributions were comparable. CONCLUSIONS: EXPLICIT was useful in both facilitating the elicitation task and in obtaining expert opinion from experts via email. The findings support the opinion that remote, self-administered elicitation is a viable approach within the constraints of HTA to inform policy making, although poor response rates may be observed and additional time for individual sessions may be required.

A comparison of two methods for expert elicitation in health technology assessments.

BACKGROUND: When data needed to inform parameters in decision models are lacking, formal elicitation of expert judgement can be used to characterise parameter uncertainty. Although numerous methods for eliciting expert opinion as probability distributions exist, there is little research to suggest whether one method is more useful than any other method. This study had three objectives: (i) to obtain subjective probability distributions characterising parameter uncertainty in the context of a health technology assessment; (ii) to compare two elicitation methods by eliciting the same parameters in different ways; (iii) to collect subjective preferences of the experts for the different elicitation methods used. METHODS: Twenty-seven clinical experts were invited to participate in an elicitation exercise to inform a published model-based cost-effectiveness analysis of alternative treatments for prostate cancer. Participants were individually asked to express their judgements as probability distributions using two different methods - the histogram and hybrid elicitation methods - presented in a random order. Individual distributions were mathematically aggregated across experts with and without weighting. The resulting combined distributions were used in the probabilistic analysis of the decision model and mean incremental cost-effectiveness ratios and the expected values of perfect information (EVPI) were calculated for each method, and compared with the original cost-effectiveness analysis. Scores on the ease of use of the two methods and the extent to which the probability distributions obtained from each method accurately reflected the expert's opinion were also recorded. RESULTS: Six experts completed the task. Mean ICERs from the probabilistic analysis ranged between £162,600-£175,500 per quality-adjusted life year (QALY) depending on the elicitation and weighting methods used. Compared to having no information, use of expert opinion decreased decision uncertainty: the EVPI value at the £30,000 per QALY threshold decreased by 74-86 % from the original cost-effectiveness analysis. Experts indicated that the histogram method was easier to use, but attributed a perception of more accuracy to the hybrid method. CONCLUSIONS: Inclusion of expert elicitation can decrease decision uncertainty. Here, choice of method did not affect the overall cost-effectiveness conclusions, but researchers intending to use expert elicitation need to be aware of the impact different methods could have.

First do no harm: pain relief for the peripheral venous cannulation of adults, a systematic review and network meta-analysis.

BACKGROUND: Peripheral venous cannulation is an everyday practice in hospitals, which many adults find painful. However, anaesthesia for cannulation is usually only offered to children. Inadequate pain relief is not only unpleasant for patients but may cause anxiety about further treatment and deter patients from seeking medical care in the future. The aim of this study is to discover the most effective local anaesthetic for adult peripheral venous cannulation and to find out how the pain of local anaesthetic application compares with that of unattenuated cannulation. METHODS: These aims are addressed through a systematic review, network meta-analysis and random-effects meta-analysis. Searching covered 12 databases including MEDLINE and EMBASE from 1990 to August 2015. The main included study design was RCTs. The primary outcome measure is self-reported pain, measured on a 100 mm visual analogue scale. RESULTS: The systematic review found 37 includable studies, 27 of which were suitable for network meta-analysis and two for random-effects meta-analysis. The results of the network meta-analysis indicate that none of the 17 anaesthetic considered had a very high probability of being the most effective when compared to each other; 2 % lidocaine had the highest probability (44 %). When the anaesthetics were compared to no treatment, the network meta-analysis showed that again 2 % lidocaine was estimated to be the most effective (mean difference -25.42 (95 % CI -32.25, -18.57). Other members of the 'caine' family were also estimated to be more effective than no treatment as were Ametop®, EMLA® and Rapydan® patch. The meta-analysis compared the pain of anaesthetic application with the unattenuated pain of cannulation. This found that all applications of local anaesthetic were less painful than cannulation without local anaesthetic. In particular a 1 % lidocaine injection was estimated to be -12.97 (95 % CI -15.71, -10.24) points (100 mm VAS) less painful than unattenuated cannulation. CONCLUSIONS: The pain of peripheral venous cannulation in adults can be successfully treated. The pain of application of any local anaesthetic is less than that of unattenuated cannulation. Local anaesthetic prior to cannulation should become normal practice and a marker of high quality care. PROTOCOL REGISTRATION: The protocol for the larger study was registered with PROSPERO no. CRD42012002093 .

Accuracy of an artificial intelligence as a medical device as part of a UK-based skin cancer teledermatology service.

Introduction: An artificial intelligence as a medical device (AIaMD), built on convolutional neural networks, has demonstrated high sensitivity for melanoma. To be of clinical value, it needs to safely reduce referral rates. The primary objective of this study was to demonstrate that the AIaMD had a higher rate of correctly classifying lesions that did not need to be referred for biopsy or urgent face-to-face dermatologist review, compared to teledermatology standard of care (SoC), while achieving the same sensitivity to detect malignancy. Secondary endpoints included the sensitivity, specificity, positive and negative predictive values, and number needed to biopsy to identify one case of melanoma or squamous cell carcinoma (SCC) by both the AIaMD and SoC. Methods: This prospective, single-centre, single-arm, masked, non-inferiority, adaptive, group sequential design trial recruited patients referred to a teledermatology cancer pathway (clinicaltrials.gov NCT04123678). Additional dermoscopic images of each suspicious lesion were taken using a smartphone with a dermoscopic lens attachment. The images were assessed independently by a consultant dermatologist and the AIaMD. The outputs were compared with the final histological or clinical diagnosis. Results: A total of 700 patients with 867 lesions were recruited, of which 622 participants with 789 lesions were included in the per-protocol (PP) population. In total, 63.3% of PP participants were female; 89.0% identified as white, and the median age was 51 (range 18-95); and all Fitzpatrick skin types were represented including 25/622 (4.0%) type IV-VI skin. A total of 67 malignant lesions were identified, including 8 diagnosed as melanoma. The AIaMD sensitivity was set at 91 and 92.5%, to match the literature-defined clinician sensitivity (91.46%) as closely as possible. In both settings, the AIaMD identified had a significantly higher rate of identifying lesions that did not need a biopsy or urgent referral compared to SoC (p-value = 0.001) with comparable sensitivity for skin cancer. Discussion: The AIaMD identified significantly more lesions that did not need to be referred for biopsy or urgent face-to-face dermatologist review, compared to teledermatologists. This has the potential to reduce the burden of unnecessary referrals when used as part of a teledermatology service.

Evolution of the evidence on the effectiveness and cost-effectiveness of acetylcholinesterase inhibitors and memantine for Alzheimer's disease: systematic review and economic model.

INTRODUCTION: in 2007 the National Institute of Health and Clinical Excellence (NICE) restricted the use of acetylcholinesterase inhibitors and memantine. METHODS: we conducted a health technology assessment (HTA) of the effectiveness and cost-effectiveness of donepezil, galantamine, rivastigmine and memantine for the treatment of AD to re-consider and up-date the evidence base used to inform the 2007 NICE decision. The systematic review of effectiveness targeted randomised controlled trials. A comprehensive search, including MEDLINE, Embase and the Cochrane Library, was conducted from January 2004 to March 2010. All key review steps were done by two reviewers. Random effects meta-analysis was conducted. The cost-effectiveness was assessed using a cohort-based model with three health states: pre-institutionalised, institutionalised and dead. The perspective was NHS and Personal Social Services and the cost year 2009. RESULTS: confidence about the size and statistical significance of the estimates of effect of galantamine, rivastigmine and memantine improved on function and global impact in particular. Cost-effectiveness also changed. For donepezil, galantamine and rivastigmine, the incremental cost per quality-adjusted life year (QALY) in 2004 was above £50,000; in 2010 the same drugs 'dominated' best supportive care (improved clinical outcome at reduced cost). This was primarily because of changes in the modelled costs of introducing the drugs. For memantine, the cost-effectiveness also improved from a range of £37-53,000 per QALY gained to a base-case of £32,000. CONCLUSION: there has been a change in the evidence base between 2004 and 2010 consistent with the change in NICE guidance. Further evolution in cost-effectiveness estimates is possible particularly if there are changes in drug prices.

Children's physiological and emotional reactions to witnessing bullying predict bystander intervention.

Study goals were to explore whether children clustered into groups based on reactions to witnessing bullying and to examine whether these reactions predicted bullying intervention. Seventy-nine children (M = 10.80 years) watched bullying videos in the laboratory while their heart rate (HR) was measured, and they self-reported on negative emotion after each video. Bullying intervention was assessed by school peers. Two groups emerged based on reactions to the bullying videos: The Emotional group (43% of children) displayed HR acceleration and reported high negative emotion, whereas the Unemotional group (57% of children) showed HR deceleration and reported low negative emotion. Group membership predicted bullying intervention, with peers reporting that Emotional children were more likely to stop a bully than Unemotional children.

Psychological consequences of false-positive screening mammograms in the UK.

OBJECTIVES: To identify the psychological effects of false-positive screening mammograms in the UK. METHODS: Systematic review of all controlled studies and qualitative studies of women with a false-positive screening mammogram. The control group participants had normal mammograms. All psychological outcomes including returning for routine screening were permitted. All studies had a narrative synthesis. RESULTS: The searches returned seven includable studies (7/4423). Heterogeneity was such that meta-analysis was not possible. Studies using disease-specific measures found that, compared to normal results, there could be enduring psychological distress that lasted up to 3 years; the level of distress was related to the degree of invasiveness of the assessment. At 3 years the relative risks were, further mammography, 1.28 (95% CI 0.82 to 2.00), fine needle aspiration 1.80 (95% CI 1.17 to 2.77), biopsy 2.07 (95% CI 1.22 to 3.52) and early recall 1.82 (95% CI 1.22 to 2.72). Studies that used generic measures of anxiety and depression found no such impact up to 3 months after screening. Evidence suggests that women with false-positive mammograms have an increased likelihood of failing to reattend for routine screening, relative risk 0.97 (95% CI 0.96 to 0.98) compared with women with normal mammograms. CONCLUSIONS: Having a false-positive screening mammogram can cause breast cancer-specific distress for up to 3 years. The degree of distress is related to the invasiveness of the assessment. Women with false-positive mammograms are less likely to return for routine assessment than those with normal ones.

Test accuracy of artificial intelligence-based grading of fundus images in diabetic retinopathy screening: A systematic review.

OBJECTIVES: To systematically review the accuracy of artificial intelligence (AI)-based systems for grading of fundus images in diabetic retinopathy (DR) screening. METHODS: We searched MEDLINE, EMBASE, the Cochrane Library and the ClinicalTrials.gov from 1st January 2000 to 27th August 2021. Accuracy studies published in English were included if they met the pre-specified inclusion criteria. Selection of studies for inclusion, data extraction and quality assessment were conducted by one author with a second reviewer independently screening and checking 20% of titles. Results were analysed narratively. RESULTS: Forty-three studies evaluating 15 deep learning (DL) and 4 machine learning (ML) systems were included. Nine systems were evaluated in a single study each. Most studies were judged to be at high or unclear risk of bias in at least one QUADAS-2 domain. Sensitivity for referable DR and higher grades was ≥85% while specificity varied and was <80% for all ML systems and in 6/31 studies evaluating DL systems. Studies reported high accuracy for detection of ungradable images, but the latter were analysed and reported inconsistently. Seven studies reported that AI was more sensitive but less specific than human graders. CONCLUSIONS: AI-based systems are more sensitive than human graders and could be safe to use in clinical practice but have variable specificity. However, for many systems evidence is limited, at high risk of bias and may not generalise across settings. Therefore, pre-implementation assessment in the target clinical pathway is essential to obtain reliable and applicable accuracy estimates.