Current recommendations/practices for anonymising data from clinical trials in order to make it available for sharing: A scoping review.

BACKGROUND/AIMS: There are increasing pressures for anonymised datasets from clinical trials to be shared across the scientific community, and differing recommendations exist on how to perform anonymisation prior to sharing. We aimed to systematically identify, describe and synthesise existing recommendations for anonymising clinical trial datasets to prepare for data sharing. METHODS: We systematically searched MEDLINE®, EMBASE and Web of Science from inception to 8 February 2021. We also searched other resources to ensure the comprehensiveness of our search. Any publication reporting recommendations on anonymisation to enable data sharing from clinical trials was included. Two reviewers independently screened titles, abstracts and full text for eligibility. One reviewer extracted data from included papers using thematic synthesis, which then was sense-checked by a second reviewer. Results were summarised by narrative analysis. RESULTS: Fifty-nine articles (from 43 studies) were eligible for inclusion. Three distinct themes are emerging: anonymisation, de-identification and pseudonymisation. The most commonly used anonymisation techniques are: removal of direct patient identifiers; and careful evaluation and modification of indirect identifiers to minimise the risk of identification. Anonymised datasets joined with controlled access was the preferred method for data sharing. CONCLUSIONS: There is no single standardised set of recommendations on how to anonymise clinical trial datasets for sharing. However, this systematic review shows a developing consensus on techniques used to achieve anonymisation. Researchers in clinical trials still consider that anonymisation techniques by themselves are insufficient to protect patient privacy, and they need to be paired with controlled access.

Dealing with missing standard deviation and mean values in meta-analysis of continuous outcomes: a systematic review.

BACKGROUND: Rigorous, informative meta-analyses rely on availability of appropriate summary statistics or individual participant data. For continuous outcomes, especially those with naturally skewed distributions, summary information on the mean or variability often goes unreported. While full reporting of original trial data is the ideal, we sought to identify methods for handling unreported mean or variability summary statistics in meta-analysis. METHODS: We undertook two systematic literature reviews to identify methodological approaches used to deal with missing mean or variability summary statistics. Five electronic databases were searched, in addition to the Cochrane Colloquium abstract books and the Cochrane Statistics Methods Group mailing list archive. We also conducted cited reference searching and emailed topic experts to identify recent methodological developments. Details recorded included the description of the method, the information required to implement the method, any underlying assumptions and whether the method could be readily applied in standard statistical software. We provided a summary description of the methods identified, illustrating selected methods in example meta-analysis scenarios. RESULTS: For missing standard deviations (SDs), following screening of 503 articles, fifteen methods were identified in addition to those reported in a previous review. These included Bayesian hierarchical modelling at the meta-analysis level; summary statistic level imputation based on observed SD values from other trials in the meta-analysis; a practical approximation based on the range; and algebraic estimation of the SD based on other summary statistics. Following screening of 1124 articles for methods estimating the mean, one approximate Bayesian computation approach and three papers based on alternative summary statistics were identified. Illustrative meta-analyses showed that when replacing a missing SD the approximation using the range minimised loss of precision and generally performed better than omitting trials. When estimating missing means, a formula using the median, lower quartile and upper quartile performed best in preserving the precision of the meta-analysis findings, although in some scenarios, omitting trials gave superior results. CONCLUSIONS: Methods based on summary statistics (minimum, maximum, lower quartile, upper quartile, median) reported in the literature facilitate more comprehensive inclusion of randomised controlled trials with missing mean or variability summary statistics within meta-analyses.

Correction: Supported Telemonitoring and Glycemic Control in People with Type 2 Diabetes: The Telescot Diabetes Pragmatic Multicenter Randomized Controlled Trial.

[This corrects the article DOI: 10.1371/journal.pmed.1002098.].

Supported Telemonitoring and Glycemic Control in People with Type 2 Diabetes: The Telescot Diabetes Pragmatic Multicenter Randomized Controlled Trial.

BACKGROUND: Self-monitoring of blood glucose among people with type 2 diabetes not treated with insulin does not appear to be effective in improving glycemic control. We investigated whether health professional review of telemetrically transmitted self-monitored glucose results in improved glycemic control in people with poorly controlled type 2 diabetes. METHODS AND FINDINGS: We performed a randomized, parallel, investigator-blind controlled trial with centralized randomization in family practices in four regions of the United Kingdom among 321 people with type 2 diabetes and glycated hemoglobin (HbA1c) >58 mmol/mol. The supported telemonitoring intervention involved self-measurement and transmission to a secure website of twice-weekly morning and evening glucose for review by family practice clinicians who were not blinded to allocation group. The control group received usual care, with at least annual review and more frequent reviews for people with poor glycemic or blood pressure control. HbA1c assessed at 9 mo was the primary outcome. Intention-to-treat analyses were performed. 160 people were randomized to the intervention group and 161 to the usual care group between June 6, 2011, and July 19, 2013. HbA1c data at follow-up were available for 146 people in the intervention group and 139 people in the control group. The mean (SD) HbA1c at follow-up was 63.0 (15.5) mmol/mol in the intervention group and 67.8 (14.7) mmol/mol in the usual care group. For primary analysis, adjusted mean HbA1c was 5.60 mmol/mol / 0.51% lower (95% CI 2.38 to 8.81 mmol/mol/ 95% CI 0.22% to 0.81%, p = 0·0007). For secondary analyses, adjusted mean ambulatory systolic blood pressure was 3.06 mmHg lower (95% CI 0.56-5.56 mmHg, p = 0.017) and mean ambulatory diastolic blood pressure was 2.17 mmHg lower (95% CI 0.62-3.72, p = 0.006) among people in the intervention group when compared with usual care after adjustment for baseline differences and minimization strata. No significant differences were identified between groups in weight, treatment pattern, adherence to medication, or quality of life in secondary analyses. There were few adverse events and these were equally distributed between the intervention and control groups. In secondary analysis, there was a greater number of telephone calls between practice nurses and patients in the intervention compared with control group (rate ratio 7.50 (95% CI 4.45-12.65, p < 0.0001) but no other significant differences between groups in use of health services were identified between groups. Key limitations include potential lack of representativeness of trial participants, inability to blind participants and health professionals, and uncertainty about the mechanism, the duration of the effect, and the optimal length of the intervention. CONCLUSIONS: Supported telemonitoring resulted in clinically important improvements in control of glycaemia in patients with type 2 diabetes in family practice. Current Controlled Trials, registration number ISRCTN71674628. TRIAL REGISTRATION: Current Controlled Trials ISRCTN 71674628.

Increased Hospital-Based Physical Rehabilitation and Information Provision After Intensive Care Unit Discharge: The RECOVER Randomized Clinical Trial.

IMPORTANCE: Critical illness results in disability and reduced health-related quality of life (HRQOL), but the optimum timing and components of rehabilitation are uncertain. OBJECTIVE: To evaluate the effect of increasing physical and nutritional rehabilitation plus information delivered during the post-intensive care unit (ICU) acute hospital stay by dedicated rehabilitation assistants on subsequent mobility, HRQOL, and prevalent disabilities. DESIGN, SETTING, AND PARTICIPANTS: A parallel group, randomized clinical trial with blinded outcome assessment at 2 hospitals in Edinburgh, Scotland, of 240 patients discharged from the ICU between December 1, 2010, and January 31, 2013, who required at least 48 hours of mechanical ventilation. Analysis for the primary outcome and other 3-month outcomes was performed between June and August 2013; for the 6- and 12-month outcomes and the health economic evaluation, between March and April 2014. INTERVENTIONS: During the post-ICU hospital stay, both groups received physiotherapy and dietetic, occupational, and speech/language therapy, but patients in the intervention group received rehabilitation that typically increased the frequency of mobility and exercise therapies 2- to 3-fold, increased dietetic assessment and treatment, used individualized goal setting, and provided greater illness-specific information. Intervention group therapy was coordinated and delivered by a dedicated rehabilitation practitioner. MAIN OUTCOMES AND MEASURES: The Rivermead Mobility Index (RMI) (range 0-15) at 3 months; higher scores indicate greater mobility. Secondary outcomes included HRQOL, psychological outcomes, self-reported symptoms, patient experience, and cost-effectiveness during a 12-month follow-up (completed in February 2014). RESULTS: Median RMI at randomization was 3 (interquartile range [IQR], 1-6) and at 3 months was 13 (IQR, 10-14) for the intervention and usual care groups (mean difference, -0.2 [95% CI, -1.3 to 0.9; P = .71]). The HRQOL scores were unchanged by the intervention (mean difference in the Physical Component Summary score, -0.1 [95% CI, -3.3 to 3.1; P = .96]; and in the Mental Component Summary score, 0.2 [95% CI, -3.4 to 3.8; P = .91]). No differences were found for self-reported symptoms of fatigue, pain, appetite, joint stiffness, or breathlessness. Levels of anxiety, depression, and posttraumatic stress were similar, as were hand grip strength and the timed Up & Go test. No differences were found at the 6- or 12-month follow-up for any outcome measures. However, patients in the intervention group reported greater satisfaction with physiotherapy, nutritional support, coordination of care, and information provision. CONCLUSIONS AND RELEVANCE: Post-ICU hospital-based rehabilitation, including increased physical and nutritional therapy plus information provision, did not improve physical recovery or HRQOL, but improved patient satisfaction with many aspects of recovery. TRIAL REGISTRATION: isrctn.com Identifier: ISRCTN09412438.

Hydrogel-coated coils versus bare platinum coils for the endovascular treatment of intracranial aneurysms (HELPS): a randomised controlled trial.

BACKGROUND: Coated coils for endovascular treatment of cerebral aneurysm were developed to reduce recurrence and retreatment rates, and have been in clinical use for 8-9 years without robust evidence to determine their efficacy. We assessed the efficacy and safety of hydrogel-coated coils. METHODS: This randomised trial was undertaken in 24 centres in seven countries. Patients aged 18-75 years with a previously untreated ruptured or unruptured cerebral aneurysm of 2-25 mm in maximum diameter were randomly allocated (1:1) to aneurysm coiling with either hydrogel-coated coils or standard bare platinum coils (control). Randomisation was done with a computer-generated sequence, stratified by aneurysm size, shape, and dome-to-neck ratio; intention to use assist device; and by region. Participants and those assessing outcomes were masked to allocation. Analysis was by modified intention to treat (excluding missing data). Primary outcome was a composite of angiographic and clinical outcomes at 18-month follow-up. We also did prespecified subgroup analyses of characteristics likely to be relevant to angiographic outcome. This study is registered as an International Standard Randomised Controlled Trial, number ISRCTN30531382. FINDINGS: 249 patients were allocated to the hydrogel coil group and 250 to the control group. In 44 of 467 patients for whom an 18-month composite primary outcome was unavailable, 6-month angiographic results were used. 70 (28%) patients in the hydrogel group and 90 (36%) control patients had an adverse composite primary outcome, giving an absolute reduction in the proportion of adverse composite primary outcomes with hydrogel of 7·0% (95% CI -1·6 to 15·5), odds ratio (OR) 0·73 (0·49-1·1, p=0·13). In a prespecified subgroup analysis in recently ruptured aneurysms, there were more adverse composite primary outcomes in the control group than in the hydrogel group-OR 2·08 (1·24-3·46, p=0·014). There were 8·6% fewer major angiographic recurrences in patients allocated to hydrogel coils-OR 0·7 (0·4-1·0, p=0·049). There were five cases of unexplained hydrocephalus in not-recently-ruptured aneurysms in the hydrogel coil group and one case in the control group. INTERPRETATION: Whether use of hydrogel coils reduces late aneurysm rupture or improves long-term clinical outcome is not clear, but our results indicate that their use lowers major recurrence. FUNDING: MicroVention Inc.

Assessment of physiological parameters within glioblastomas in awake patients: a prospective clinical study.

OBJECT: Multiparametric brain monitoring probes now make it possible to measure cerebral physiology. This prospective clinical study was designed to evaluate the pathophysiological environment of tumoural and peritumoural tissue O(2), CO(2), pH, HCO(3)- and temperature of awake patients with glioblastoma. METHODS: A Neurotrend multiparametric sensor was placed using intraoperative image guidance into glioblastoma after biopsy under general anesthetic. Postoperative monitoring was then performed in awake patients. RESULTS: Twelve patients were recruited and monitoring was performed, and well tolerated in 9 for up to 22 hrs. Mean glioblastoma tumour values were: tissue oxygen pressure (PtiO(2)) 21.0 mmHg, standard deviation +/- 7.9; PtiCO(2) 60.2 +/- 17.2 mmHg; temperature 36.9 +/- 0.4 degrees C, pH 7.08 + 0.2; and HCO(3) 17.1 +/- 3.7. Mean peritumoural brain values in 5 patients were PtiO(2) 29.1 +/- 27.6 mmHg; PtiCO(2) 48.6 +/- 7.0 mmg; temperature 36.4 +/- 0.6 degrees C; pH 7.20 +/- 0.09 and HCO(3) 19.1 +/- 3.5. There were trends for the PtiO(2) to decrease with increasing brain depth. As glioblastoma PtiCO(2) levels decreased, pH increased. There were no relationships between either tumoural PtiO(2) and pH, or PtiO(2) and PtiCO(2), however there were large intra- and inter-tumoural variation in monitoring values. There were technical problems in some patients with the Neurotrend sensor that limited its application, and that compromised aspects of data collection and interpretation, particularly of PtiO(2). CONCLUSION: This study has shown that this novel approach to monitoring glioma pathophysiology is feasible and well tolerated by patients. The data, much of which is novel, contributes to the knowledge of glioblastoma pathophysiology. However, further study and clinical exploitation awaits the development of a more reliable multiparametric sensor.

No evidence of bias in the process of publication of diagnostic accuracy studies in stroke submitted as abstracts.

OBJECTIVE: There is little empirical evidence on publication bias in diagnostic test accuracy studies. We evaluated the proportion of abstracts presented at international stroke meetings, which were later published in full, and investigated which study features characterized publication. METHODS: We reviewed all diagnostic abstracts presented at two international stroke conferences between 1995 and 2004. We assessed the characteristics and findings of the identified abstracts. We identified full publications through electronic databases and by contacting the authors. Determinants of publication were assessed by Cox regression. RESULTS: Seventy-six percent (121 out of 160) of identified abstracts were subsequently published in full. Sixty-two percent were published within 24 months of presentation. The median time to publication was 16 months. Assessment of interobserver agreement between test readers was a significant predictor of full publication (P=0.02). No other study characteristic (including clinical utility of results, multicenter status, or Youden's index) was predictive. CONCLUSIONS: We found no clear evidence of bias in the publication process that occurs after abstract acceptance. We were unable to assess bias in abstract submission or acceptance. "Interobserver agreement" was the only characteristic statistically associated with publication. Clinical utility of results and other study characteristics did not predict publication. Diagnostic abstracts often did not report many relevant methodological aspects.

Cerebral microbleeds are associated with lacunar stroke defined clinically and radiologically, independently of white matter lesions.

BACKGROUND AND PURPOSE: Associations among microbleeds, white matter lesions (WMLs), and small deep infarcts on imaging have been reported. Because many of these imaging infarcts were asymptomatic, the relationship of microbleeds to clinical lacunar stroke is unclear. An association between microbleeds and clinically defined lacunar stroke might suggest a common causal microangiopathy. METHODS: Patients with lacunar, partial anterior circulation or posterior circulation stroke syndromes and older healthy subjects underwent MRI. Microhemorrhages, infarcts, hemorrhages, and WMLs were coded blind to clinical details. A final clinicoradiologic stroke subtype diagnosis was assigned. RESULTS: Among 308 subjects (67 older healthy and 241 with stroke), 54 patients had microbleeds (17%). Microbleeds were twice as frequent in lacunar than cortical strokes (26% versus 13%, P=0.03) or healthy older subjects (9%) and associated with increasing WML scores (P<0.0001). Lacunar and cortical stroke subtypes and healthy older subjects had similar WML scores. CONCLUSIONS: Microbleeds are associated with lacunar stroke defined clinicoradiologically more than other stroke subtypes but not simply by association with WMLs. This suggests that microbleeds and lacunar stroke have a similar microvascular abnormality.

Frequency and risk factors for spontaneous hemorrhagic transformation of cerebral infarction.

OBJECTIVE: Hemorrhagic transformation of cerebral infarction (HTI) occurs spontaneously but its frequency and risk factors are uncertain with mixed results in previous studies. We aimed to determine the overall frequency of and risk factors for HTI. METHODS: We performed a systematic review according to Cochrane Collaboration methods of published reports of HTI with reliable, systematic follow-up with computed tomography or magnetic resonance imaging. RESULTS: In all, 28 observational studies and 19 randomized controlled trials in stroke were identified that included follow-up imaging data. Problems with inconsistent definitions or small and biased patient populations limited detailed interpretation. The overall frequency of any HTI in untreated patients was 8.5% (95% confidence interval 7%-10%). Severe HTI (i.e., HTI accompanied by neurologic deterioration or parenchymal hematoma formation) occurred in 1.5% (95% confidence interval 0.8%-2.2%). The frequency of HTI increased markedly with increasing use of antithrombotic or thrombolytic drugs. Magnetic resonance imaging detected more HTI than did computed tomography. The published data were generally inadequate to undertake more detailed analysis of risk factors. However, in the 8 studies that did provide the information, HTI was associated with large infarcts, mass effect, hypodensity observed early after the stroke, and age older than 70 years, but not hypertension or cardioembolic stroke. CONCLUSIONS: Severe HTI is uncommon in patients not receiving antithrombotic or thrombolytic agents. The methods used to assess the frequency of and risk factors for HTI, particularly a standard of definitions in future prospective studies, could be improved.

Extensive haemorrhagic transformation of infarct: might it be an important cause of primary intracerebral haemorrhage?

Data from prior studies using serial imaging and post mortem data support the possibility that at least some cases of apparent primary intracerebral haemorrhage are due to early haemorrhagic transformation of infarct. If some primary intracerabral haemorrhage is actually early haemorrhagic transformation of infarct, then secondary stroke prevention for ischaemic stroke might be appropriate and so future studies should obtain data to determine the frequency of early major haemorrhagic transformation of infarct.

Comparison of risk factors in patients with transient and prolonged eye and brain ischemic syndromes.

BACKGROUND AND PURPOSE: Patients with ischemic stroke, cerebral transient ischemic attacks (TIAs), retinal artery occlusion (RAO), and amaurosis fugax are thought to have similar risk factors and underlying vascular disease. However, if risk factors are different in patients with eye compared with brain symptoms and in those whose symptoms last <24 hours (transient) compared with those lasting >24 hours (prolonged), more focused prevention strategies are possible in the future. METHODS: All patients with ischemic stroke, cerebral TIA, RAO, and amaurosis fugax presenting to our hospital from 1994 to 1999 were examined by a stroke physician. Risk factors were documented, and patients underwent carotid Doppler ultrasound. RESULTS: We registered 1491 patients with ischemic stroke, 580 with cerebral TIA, 79 with RAO, and 138 with amaurosis fugax. Atrial fibrillation was more common in brain than eye events, whether prolonged [ischemic stroke versus RAO: odds ratio (OR), 3.6; 95% confidence interval (CI), 1.1 to 12] or transient (cerebral TIA versus amaurosis fugax: OR, 2.9; 95% CI, 0.7 to 13), and more common in prolonged than transient events, whether brain (stroke versus cerebral TIA: OR, 3.3; 95% CI, 2.1 to 5.1) or eye (RAO versus amaurosis fugax: OR, 2.7; 95% CI, 0.4 to 16). Severe ipsilateral carotid disease was less common in brain than eye events, whether prolonged (ischemic stroke versus RAO: OR, 0.6; 95% CI, 0.3 to 1.0) or transient (cerebral TIA versus amaurosis fugax: OR, 0.4; 95% CI, 0.2 to 0.6). CONCLUSIONS: These data suggest that there are pathogenetic differences between transient and permanent eye and brain ischemic syndromes. Improved understanding of these mechanisms could lead to more effective stroke prevention.

Which Doppler velocity is best for assessing suitability for carotid endarterectomy?

OBJECTIVE: To evaluate which velocity, or combination of velocities, from carotid Doppler ultrasonography (DU), achieved the closest agreement with an assessment of suitability for carotid endarterectomy from intra-arterial angiograms (IAA). METHODS: We prospectively collected data from 148 consecutive patients (288 carotids), who had DU and IAA (blinded assessment) before possible carotid endarterectomy. We halved our data by randomly selecting the left or right carotid artery for each patient. We used one half to calibrate our DU results to IAA (to decide which velocity corresponded with what degree of angiographic stenosis). Using this analysis, each artery in the other half of the data was defined as suitable (80-99% stenosed) or unsuitable for carotid endarterectomy. We evaluated every individual, and combination of, velocities (strategies) to see which gave the closest agreement with IAA. RESULTS: Of all 80 strategies, six resulted in better agreement than others of the same or lower complexity. Five of these strategies gave better agreement than the internal carotid artery peak systolic velocity (ICA PSV) (kappa 0.78), but the improvement was small. CONCLUSION: Using the ICA PSV alone is adequate for assessing carotid stenosis before endarterectomy using DU, as long as the machine is calibrated to IAA. However, the addition of the ratio of the ICA PSV to the common carotid artery PSV adds only one further measurement, slightly increases the agreement with IAA, and would be reasonable to use on a daily basis.