Search

Background Machine learning (ML) tools exist that can reduce or replace human activities in repetitive or complex tasks. Yet, ML is underutilized within evidence synthesis, despite the steadily growing rate of primary study publication and the need to periodically update reviews to reflect new evidence. Underutilization may be partially explained by a paucity of evidence on how ML tools can reduce resource use and time-to-completion of reviews. Methods This protocol describes how we will answer two research questions using a retrospective study design: Is there a difference in resources used to produce reviews using recommended ML versus not using ML, and is there a difference in time-to-completion? We will also compare recommended ML use to non-recommended ML use that merely adds ML use to existing procedures. We will retrospectively include all reviews conducted at our institute from 1 August 2020, corresponding to the commission of the first review in our institute that used ML. Conclusion The results of this study will allow us to quantitatively estimate the effect of ML adoption on resource use and time-to-completion, providing our organization and others with better information to make high-level organizational decisions about ML.

Book review - No Abstract available.

BackgroundOpioids are strong pain medications that can be essential for acute pain. However, opioids are also commonly used for chronic conditions and illicitly where there are well recognised concerns about the balance of their benefits and harms. Technologies using artificial intelligence (AI) are being developed to examine and optimise the use of opioids. Yet, this research has not been synthesised to determine the types of AI models being developed and the application of these models.MethodsWe aimed to synthesise studies exploring the use of AI in people taking opioids. We searched three databases: the Cochrane Database of Systematic Reviews, EMBASE, and Medline on 4 January 2021. Studies were included if they were published after 2010, conducted in a real-life community setting involving humans, and used AI to understand opioid use. Data on the types and applications of AI models were extracted and descriptively analysed.ResultsEighty-one articles were included in our review, representing over 5.3 million participants and 14.6 million social media posts. Most (93%) studies were conducted in the USA. The types of AI technologies included natural language processing (46%) and a range of machine learning algorithms, the most common being random forest algorithms (36%). AI was predominately applied for the surveillance and monitoring of opioids (46%), followed by risk prediction (42%), pain management (10%), and patient support (2%). Few of the AI models were ready for adoption, with most (62%) being in preliminary stages.ConclusionsMany AI models are being developed and applied to understand opioid use. However, there is a need for these AI technologies to be externally validated and robustly evaluated to determine whether they can improve the use and safety of opioids.SUMMARY BOXKey PointsAcross the landscape of opioid research, natural language processing models (46%) and ensemble algorithms, particularly random forest algorithms (36%), were the most common types of AI technologies studied.There were four domains to which AI was applied to assess the use of opioids, including surveillance and monitoring (46%), risk prediction (42%), pain management (10%), and patient support (2%).The AI technologies were at various stages of development, validation, and deployment, with most (62%) models in preliminary stages, 11% required external validation, and few models were openly available to access (6%).

Introduction: Electrolyte derangements, acidosis, and volume overload remain life-threatening emergencies in people with acute kidney injury in austere environments. A single-lumen alternating micro-batch (SLAMB) dialysis technique was designed to perform renal replacement therapy using a single-lumen access, low-cost disposable bags and tubing, widely available premade fluids, and a dialysis filter. A manual variation (mSLAMB) works without electricity, battery, or a pump. We modeled mSLAMB dialysis and predicted it could achieve adequate small solute clearance, blood flow rates, and ultrafiltration accuracy. Methods: A 25- to 30-kg pediatric patient’s blood volume was simulated by a 2-L bag of expired blood and spiked with 5 g of urea initially, then with 1–2 g between experiments. Experiments had 8 cycles totaling prescription volumes of 800–2,400 mL and were conducted with different ratios of hemofiltration fluid to blood volume. Concentrations of urea and potassium, final effluent volumes, and cycle duration were measured at the end of each cycle to determine clearance, ultrafiltration accuracy, and blood flow rates. Results: Each cycle lasted 70–145 s. Experiments achieved a mean urea reduction ratio of 27.4 ± 7.1% and a mean potassium reduction of 23.4 ± 9.3%. The largest urea and potassium reduction percentage occurred with the first cycle. Increased hemofiltration fluid to blood volume ratio did not increase clearance. Mean (+/− standard deviation) blood flow ranged from 79.7 +/− 4.4 mL/min to 90.8 +/− 6.5 mL/min and increased with larger batch volume and height difference between reservoirs. Ultrafiltration accuracy ranged from 0 to 2.4% per cycle. Discussion: mSLAMB dialysis is a simple, manual, cost-effective mode of dialysis capable of providing clearance and accurate ultrafiltration. With further refinement of technique, we believe this can be a potentially lifesaving treatment in austere conditions and low-resource settings.

Significant controversy has arisen over the role of the renin-angiotensin-aldosterone system (RAAS) in COVID-19 pathophysiology. In this prospective, observational study, we evaluated plasma angiot...

AimsNetwork meta-analysis (NMA) has been used in several systematic reviews on relapsing refractory multiple myeloma (RRMM). NMAs have been questioned on the basis that effect modification may invalidate the underpinning assumptions. We aimed to systematically review and meta-analyze the evidence for effect modification of hazard ratios (HRs) for overall survival (OS) and progression-free survival (PFS) with respect to refractory status and number of treatment lines.MethodsWe extracted stratified HR estimates from 42 phase 2 and 3 randomized controlled trials (RCTs). We tested for within-study effect modification and used meta-analyses to estimate ratios of hazard ratios (RHRs) across trial under assumptions that strongly favor the modification hypothesis. RHR estimates were used in simulations to estimate how many NMA results would be expected to differ in the presence versus absence of effect modification.ResultsMost (95%) publications could have reported stratified estimates but only 14% (OS) and 43% (PFS) did. Within-study evidence for effect modification is very weak (p > 0.05 for 47 of 49 sets of stratified estimates). The largest RHR estimated was 1.31 (95% CI 1.16–1.47), for the modifying effect of refractory status on HR for PFS. Simulations suggest that, in the worst case, effect modification would result in 4.48% (95% CI 4.42%–4.53%) of NMA estimates differing statistically significantly in the presence versus absence of effect modification.ConclusionsEffect modification is essentially undetectable in phase 2 and 3 trials. In the worst case, it is unlikely to affect more than about 5% of random-effects NMA estimates.

Purpose This paper aims to propose a multi-dimensional hierarchical K-means clustering algorithm for the purpose of intrusion detection. Initially, the clustering set of rules is proposed to shape some of clusters in the network and then the most beneficial clusters are decided on by the use of Cuckoo search optimization set of rules. Finally, an Artificial Bee Colony primarily based selection tree (ABC-DT) classifier is rented to classify the regular and unusual instances present in the network with the aid of the extracted features. Design/methodology/approach Intrusion detection system (IDS) is crucial for the network system; the intruder can take sensitive details about the network. IDS are said to be more effective when it has both high intrusion detection rate and low false alarm rate. Numerous strategies including gadget mastering, records mining and statistical techniques were tested for IDS mission. Recent study reveals that combining multiple classifiers, i.e. classifiers ensemble, can also own better performance than unmarried classifier. In this paper, a comparative study is conducted of the overall performance of four classifiers, i.e. hybrid ABC-DT particle swarm optimization-based K-means clustering (PSO-KM), help vector device (SVM) and K-Nearest neighbour (KNN). All the four classifiers are tested with exceptional packet sizes 1470, 1024, 512 and 256. The experiment is carried out for the speed ranging from turned into done for the velocity ranging from 250Mbps, 500Mbps, 750Mbps, 1.0Gpbs, 1.5Gbps, and 2.0Gbps in terms of accuracy, detection charge, specificity, false alarm charge and computational time. The experimental results reveals that the hybridization of classifiers performs better than the base classifiers in all scenarios. Findings This study analyses the performance of hybrid ABC-DT classifier and compares the performance against three well-known classifiers such as PSO-KM, SVM and K-NN. The performances of all the four classifiers are tested with Discovery in Data Mining (KDD) CUP 99 dataset with different packet sizes 1470, 1024, 512 and 256. The results show the classifier performance variations with different speed ranges. From the experimental results and analysis, the hybridization of classifiers such as ABC-DT outperforms the base classifiers in all scenarios. Originality/value The novel approach in this paper is used to study the hybrid ABC-DT classifier and compare the performance against three well-known classifiers such as PSO-KM, SVM and K-NN. The discussed concept is used within the network to monitor the traffic to and from all the devices connected in that network.

ObjectivesThe renal activity index for lupus (RAIL) measures lupus nephritis (LN) activity considering urine levels of 6 biomarkers (neutrophil gelatinase-associated lipocalin, monocyte chemoattractant protein-1, kidney injury molecule-1, adiponectin, haemopexin, ceruloplasmin). We aimed to compare the accuracy of the RAIL and the renal domain-score of the SLE disease activity index (rSLEDAI) in detecting LN activity.MethodsRandom urine samples of patients with childhood-onset SLE with and without LN were assayed and scores of the RAIL, and RAIL standardised for urine creatinine (RAIL-Cr) were calculated. Clinical LN activity was measured by the rSLEDAI, and histological activity of LN was categorised as inactive/low-moderate/high for National Institute of Health-activity index scores of 10, respectively.Results115 patients were included in the analysis (47 patients without and 68 with LN). RAIL, RAIL-Cr and rSLEDAI scores at the time (±3 months) of kidney biopsy were available for 32 patients. Median rSLEDAI, RAIL and RAIL-Cr values were 4, –0.04, 0.02 for inactive LN, 12, 0.7 and 0.9 for low-moderate LN activity and 12, 2 and 1.8 for high LN activity, respectively. The area under the receiver operating characteristic curve (AUC) to capture high LN activity was the lowest for the rSLEDAI (AUC=0.62), followed by the RAIL-Cr (AUC=0.73) and RAIL (AUC=0.79). Notably, when testing urine samples collected during routine clinic visits remote (>3 months) from a kidney biopsy, 50% patients with rSLEDAI scores of 0 had RAIL scores reflecting low-moderate LN activity.ConclusionMonitoring of renal inflammation in children and adolescents with SLE can be improved by the measurement of urine biomarkers. The RAIL may constitute important auxiliary tool for the surveillance of LN in a clinical setting and assist with the decision to obtain a kidney biopsy.

BACKGROUND Digital health interventions have been shown to improve data quality and health services in low- and middle-income countries (LMICs). Nonetheless, in LMICs, systematic assessments of time saved with the use of digital tools are rare. We ran a set of cluster-randomized controlled trials as part of the implementation of a digital maternal and child health registry (eRegistry) in the West Bank, Palestine. OBJECTIVE In the eRegTime study, we compared time spent on health information management in clinics that use the eRegistry versus the existing paper-based documentation system. METHODS Intervention (eRegistry) and control (paper documentation) arms were defined by a stratified random subsample of primary health care clinics from the concurrent eRegQual trial. We used time-motion methodology to collect data on antenatal care service provision. Four observers used handheld tablets to record time-use data during one working day per clinic. We estimated relative time spent on health information management for booking and follow-up visits and on client care using mixed-effects linear regression. RESULTS In total, 22 of the 24 included clinics (12 intervention, 10 control) contributed data; no antenatal care visits occurred in the other two clinics during the study period. A total of 123 and 118 consultations of new pregnancy registrations and follow-up antenatal care visits were observed in the intervention and control groups, respectively. Average time spent on health information management for follow-up antenatal care visits in eRegistry clinics was 5.72 minutes versus 8.10 minutes in control clinics (adjusted relative time 0.69, 95% CI 0.60-0.79; PP=.85) for booking visits. The average time spent on documentation, a subcategory of health information management, was 5.50 minutes in eRegistry clinics versus 8.48 minutes in control clinics (adjusted relative time 0.68, 95% CI 0.56-0.83; PP=.27). CONCLUSIONS The eRegistry captures digital data at point of care during client consultations and generates automated routine reports based on the clinical data entered. Markedly less time (plausibly a saving of at least 18%) was spent on health information management in eRegistry clinics compared to those that use paper-based documentation. This is likely explained by the fact that the eRegistry requires lesser repetitive documentation work than paper-based systems. Adoption of eRegistry-like systems in comparable settings may save valuable and scarce health care resources. CLINICALTRIAL ISRCTN registry ISRCTN18008445; https://doi.org/10.1186/ISRCTN18008445 INTERNATIONAL REGISTERED REPORT RR2-10.2196/13653

Series arc fault is widely concerned in DC distribution system, as they could introduce transient electromagnetic interference to nearby equipment and cause serious system failure to electrical system and signalling system. Extracting the signatures of series arc fault has an important role in arc-fault detection algorithms and protection techniques for sensitive equipment. Current studies show that characteristic analysis methods based on frequency domain are deficient in identifying arc fault from other transient emission [1] - [3] , therefore the time-frequency representation is considered as an efficient tool in extracting the arc fault signatures.

Background and Purpose— Many patients with acute ischemic stroke are not eligible for thrombolysis or mechanical reperfusion therapies due to contraindications, inaccessible vascular occlusions, late presentation, or large infarct core. Sphenopalatine ganglion (SPG) stimulation to enhance collateral flow and stabilize the blood-brain barrier offers an alternative, potentially more widely deliverable, therapy. Methods— In a randomized, sham-controlled, double-masked trial at 41 centers in 7 countries, patients with anterior circulation ischemic stroke not treated with reperfusion therapies within 24 hours of onset were randomly allocated to active SPG stimulation or sham control. The primary efficacy outcome was improvement beyond expectations on the modified Rankin Scale of global disability at 90 days (sliding dichotomy), assessed in the modified intention-to-treat population. The initial planned sample size was 660 patients, but the trial was stopped early when technical improvements in device placement occurred, so that analysis of accumulated experience could be conducted to inform a successor trial. Results— Among 303 enrolled patients, 253 received at least one active SPG or sham stimulation, constituting the modified intention-to-treat population (153 SPG stimulation and 100 sham control). Age was median 73 years (interquartile range, 64–79), 52.6% were female, deficit severity on the National Institutes of Health Stroke Scale was median 11 (interquartile range, 9–15), and time from last known well median 18.6 hours (interquartile range, 14.5–22.5). For the primary outcome, improved 3-month disability beyond expectations, rates in the SPG versus sham treatment groups were 49.7% versus 40.0%; odds ratio, 1.48 (95% CI, 0.89–2.47); P =0.13. A significant treatment interaction with stroke location (cortical versus noncortical) was noted, P =0.04. In the 87 patients with confirmed cortical involvement, rates of improvement beyond expectations were 50.0% versus 27.0%; odds ratio, 2.70 (95% CI, 1.08–6.73); P =0.03. Similar response patterns were observed for all prespecified secondary efficacy outcomes. No differences in mortality or serious adverse event safety end points were observed. Conclusions— SPG stimulation within 24 hours of onset is safe in acute ischemic stroke. SPG stimulation was not shown to statistically significantly improve 3-month disability above expectations, though favorable outcomes were nominally higher with SPG stimulation. Beneficial effects may distinctively be conferred in patients with confirmed cortical involvement. The results of this study need to be confirmed in a larger pivotal study. Clinical Trial Registration— URL: https://www.clinicaltrials.gov . Unique identifier: NCT03767192.

Decades of reports and policy have drawn attention to the significant social and occupational impairment of many young Aboriginal men in Central Australia. However, the role of mental ill-health as a contributing factor to this impairment, and culturally appropriate intervention targets have received insufficient attention in the psychiatry literature. Despite having the worst health outcomes of any population in Australia, Aboriginal men chronically underuse primary health care services. It's proposed that interventions ensuring cultural continuity through Identity-strengthening with a particular focus on positive Aboriginal masculinities will address a critical mental health gap for young men. In Central Australian and broader Indigenous populations, tangible and measurable kinship, language, religious and economic (KLRE) activities are catalytic vehicles for restoring traditional knowledge that suffer ongoing pressures as a result of colonization and assimilationist Government policy. By transforming KLRE knowledge content from ethnographic archives, these culturally rich repositories may be utilized to create education and engagement materials that will support young Aboriginal men's efforts to obtain and maintain positive mental health. This proposal focuses on building resilience through the acquisition of KLRE knowledge which young Aboriginal men can utilize as resources for enhancing positive identity and mental health outcomes.

This paper provides a record and analysis of a site in east-midland England, at which organic and soil material are found between two Middle Pleistocene tills. This is the first discovery of its kind in the area, and demonstrates unequivocally that the region was glaciated on two separate occasions, something that has long been inferred and articulated, but not actually demonstrated. The landforms, sediments and soils are studied with respect to their geomorphological, lithological, pedological, palaeobotanical and structural properties. The organic and soil material along with soil structures indicate, sequentially, a periglacial climate, a long period of warm temperate weathering and a cool temperate climate. Evaluation of this evidence in terms of existing published work identifies a number of problems with existing models and suggests that the most likely model for the glacial history of this part of midland England is an early Middle Pleistocene glaciation which is represented only by trace erratics, a Marine Isotope Stage (MIS) 12 age glaciation which moved across the area from the NW and deposited a chalk-free till, and an MIS 8 age glaciation that transported and deposited an upper chalky till from the NE.

BackgroundSleepio is an automated digital programme that delivers cognitive behavioural therapy for insomnia. Sleepio has been proven effective in improving sleep difficulties. However, evidence for the possible impact of Sleepio use on health care costs in the United Kingdom has not previously been developed. In this study, we assessed the effect of a population-wide rollout of Sleepio in terms of primary care costs in the National Health Service (NHS) in England.MethodsThe study was conducted in the Thames Valley region of England, where access to Sleepio was made freely available to all residents between October 2018 and January 2020. We use primary care data for people with relevant characteristics from nine general practices in Buckinghamshire. The study relies on a quasi-experimental design, using an interrupted time series to compare the trend in primary care costs before and after the rollout of Sleepio. Primary care costs include general practice contacts and prescriptions. Segmented regression analysis was used to estimate primary and secondary outcomes.ResultsFor the 10,704 patients included in our sample, the total saving over the 65-week follow-up period was £71,027. This corresponds to £6.64 per person in our sample or around £70.44 per Sleepio user. Secondary analyses suggest that savings may be driven primarily by reductions in prescriptions.ConclusionSleepio rollout reduced primary care costs. National adoption of Sleepio may reduce primary care costs by £20 million in the first year. The expected impact on primary care costs in any particular setting will depend on the uptake of Sleepio.

Objectives Severe coronavirus disease 2019 (COVID-19) is associated with a dysregulated immune state. While research has focused on the hyperinflammation, little research has been performed on the compensatory anti-inflammatory response. The aim of this study was to evaluate the anti-inflammatory cytokine response to COVID-19, by assessing interleukin-10 (IL-10) and IL-10/lymphocyte count ratio and their association with outcomes. Methods Adult patients presenting to the emergency department (ED) with laboratory-confirmed COVID-19 were recruited. The primary endpoint was maximum COVID-19 severity within 30 days of index ED visit. Results A total of 52 COVID-19 patients were enrolled. IL-10 and IL-10/lymphocyte count were significantly higher in patients with severe disease (p Conclusions The hyperinflammatory response to COVID-19 is accompanied by a simultaneous anti-inflammatory response, which is associated with poor outcomes and may increase the risk of new positive bacterial cultures. IL-10 and IL-10/lymphocyte count at ED presentation were independent predictors of COVID-19 severity. Moreover, elevated IL-10 was more strongly associated with outcomes than pro-inflammatory IL-6 or IL-8. The anti-inflammatory response in COVID-19 requires further investigation to enable more precise immunomodulatory therapy against SARS-CoV-2.

Background Digital health interventions have been shown to improve data quality and health services in low- and middle-income countries (LMICs). Nonetheless, in LMICs, systematic assessments of time saved with the use of digital tools are rare. We ran a set of cluster-randomized controlled trials as part of the implementation of a digital maternal and child health registry (eRegistry) in the West Bank, Palestine. Objective In the eRegTime study, we compared time spent on health information management in clinics that use the eRegistry versus the existing paper-based documentation system. Methods Intervention (eRegistry) and control (paper documentation) arms were defined by a stratified random subsample of primary health care clinics from the concurrent eRegQual trial. We used time-motion methodology to collect data on antenatal care service provision. Four observers used handheld tablets to record time-use data during one working day per clinic. We estimated relative time spent on health information management for booking and follow-up visits and on client care using mixed-effects linear regression. Results In total, 22 of the 24 included clinics (12 intervention, 10 control) contributed data; no antenatal care visits occurred in the other two clinics during the study period. A total of 123 and 118 consultations of new pregnancy registrations and follow-up antenatal care visits were observed in the intervention and control groups, respectively. Average time spent on health information management for follow-up antenatal care visits in eRegistry clinics was 5.72 minutes versus 8.10 minutes in control clinics (adjusted relative time 0.69, 95% CI 0.60-0.79; P Conclusions The eRegistry captures digital data at point of care during client consultations and generates automated routine reports based on the clinical data entered. Markedly less time (plausibly a saving of at least 18%) was spent on health information management in eRegistry clinics compared to those that use paper-based documentation. This is likely explained by the fact that the eRegistry requires lesser repetitive documentation work than paper-based systems. Adoption of eRegistry-like systems in comparable settings may save valuable and scarce health care resources. Trial Registration ISRCTN registry ISRCTN18008445; https://doi.org/10.1186/ISRCTN18008445 International Registered Report Identifier (IRRID) RR2-10.2196/13653

ObjectivesSevere coronavirus disease 2019 (COVID-19) is associated with a dysregulated immune state, called cytokine storm. While research has focused on the hyperinflammation, little research has been performed on the compensatory anti-inflammatory response which if severe may lead to a state of functional immunoparalysis. The aim of this study was to evaluate the anti-inflammatory response to COVID-19, by assessing interleukin-10 (IL-10) and IL-10/lymphocyte count ratio and their association with patient outcomes.MethodsAdult patients presenting to the emergency department (ED) with laboratory-confirmed COVID-19 were recruited. The primary endpoint was peak COVID-19 severity within 30 days of index ED visit. Additional endpoints included COVID-19 severity at ED disposition, development of severe acute kidney injury (AKI) or secondary bacterial infections.ResultsA total of 52 COVID-19 patients were enrolled. IL-10 and IL-10/lymphocyte count were significantly higher in patients with severe disease at both time points (all pConclusionsThe hyperinflammatory response to COVID-19 is accompanied by a simultaneous anti-inflammatory response, which is associated with poor outcomes and may increase the risk of secondary bacterial infections. IL-10 and IL-10/lymphocyte ratio at ED presentation were independent predictors of COVID-19 severity. Functional immunoparalysis in COVID-19 requires further investigation to enable more precise immunomodulatory therapy against SARS-CoV-2.

The aim of this paper is to understand the effects of low frequency (0-2kHz) conducted emissions on power quality from a microgrid based DC/DC converter under steady state and transient loading conditions. The experimental data obtained from the lab set-up is analysed using Fast Fourier Transform and low frequency sinusoidal disturbance indices for input DC voltage and current are calculated. The results show that power quality of a DC/DC converter with varying loads does get affected by the conducted emissions, but the indices obtained cannot be quantified due to lack of standards for DC power quality. The indices decrease with increasing load and have a non-zero value at no-load condition. Further, the calculated results from amplitude probability distribution are in agreement with the indices as well.

Former glaciolacustrine systems are an important archive of palaeoglaciological, palaeoenvironmental and palaeoclimatic change. The annually laminated (varved) sediments that, under certain conditions, accumulate in former glacial lakes, offer a rare opportunity to reconstruct such changes (e.g. glacier advance and retreat cycles, glacier ablation trends, permafrost melt, nival events) at annual or even sub-annual temporal resolution. Data of this kind are desirable for their ability to guide and test numerical model simulations of glacier dynamics and palaeoclimatic change that occur over rapid time intervals, with implications for predicting future glacier response to climatic change, or the effects of weather and climate events on lake sedimentation. The most valuable records preserved in glaciolacustrine systems are continuous varved sequences formed in the distal parts of glacial lakes, where microscale lamination structures can accumulate relatively undisturbed. Technological advances, in the last few decades, have enabled improved characterisation of glaciolacustrine varve microfacies and the precise measurement of varve thickness at the micrometre scale. However, unlike in cognate fields (e.g. soil science), protocols for the robust and consistent description and interpretation of glaciolacustrine varve sediments are lacking. To fill this gap, and to provide a resource for future studies of glaciolacustrine varved sediments, this paper reviews the processes of sedimentation in glacial lake basins, and presents the defining microfacies characteristics of glacial varves using a descriptive protocol that uses consistent examination of grain size, sorting, structure, nature of contacts, development of plasmic fabrics and features such as dropgrains and intraclasts within individual laminations. These lamination types are then combined into lamination sets, whose structures can be interpreted as glaciolacustrine varves. Within this framework, we define five principal assemblages of glaciolacustrine varve microfacies which, if clearly identified in palaeoglaciolacustrine settings, enable more detailed palaeoenvironmental interpretations to be made. Finally, we discuss the utility and complexities of reconstructing the evolution of former glacial lake systems using varve microfacies and thickness datasets.

Although substantial work has been done on the pre-glacial terraces of East Anglia, very little systematic work has been done to understand the origin of river terraces in East Anglia that have formed since ice last covered the region. This paper records the results of studies of exposures and borehole records in ‘classical’ Quaternary terrace landforms that are considered to have formed since the Anglian (MIS 12) Glaciation, in the middle Waveney Valley. These features have been examined in terms of their morphological and sedimentological properties, in order to provide a detailed record of their form and composition, understand their processes of formation, and identify their stratigraphical status. The results show that the main body of the highest terrace (Homersfield Terrace, Terrace 3) is not composed of river sediments, but of shallow marine sediments, and is a remnant of early Middle Pleistocene Wroxham Crag. River sediments, in the form of Anglian age (MIS 12) glaciofluvial Aldeby Sands and Gravels also exist in the area as a channel fill, cut through the Wroxham Crag, and reflect outwash erosion and sedimentation from a relatively proximal ice margin to the west. The results mean that the interpretations previously presented for the terrace landforms of the middle Waveney valley are not applicable. The issue of why the terrace stratigraphy, hitherto identified in East Anglia cannot be related to that for the River Thames to the south and the rivers of Midland England to the west, still requires further research.