Search

Differential privacy (DP) is a mathematical privacy notion increasinglydeployed across government and industry. With DP, privacy protections areprobabilistic: they are bounded by the privacy budget parameter, $\epsilon$.Prior work in health and computational science finds that people struggle toreason about probabilistic risks. Yet, communicating the implications of$\epsilon$ to people contributing their data is vital to avoiding privacytheater -- presenting meaningless privacy protection as meaningful -- andempowering more informed data-sharing decisions. Drawing on best practices inrisk communication and usability, we develop three methods to conveyprobabilistic DP guarantees to end users: two that communicate odds and oneoffering concrete examples of DP outputs. We quantitatively evaluate these explanation methods in a vignette surveystudy ($n=963$) via three metrics: objective risk comprehension, subjectiveprivacy understanding of DP guarantees, and self-efficacy. We find thatodds-based explanation methods are more effective than (1) output-based methodsand (2) state-of-the-art approaches that gloss over information about$\epsilon$. Further, when offered information about $\epsilon$, respondents aremore willing to share their data than when presented with a state-of-the-art DPexplanation; this willingness to share is sensitive to $\epsilon$ values: asprivacy protections weaken, respondents are less likely to share data.

Background: It has been suggested that compounds present in silicone breast implants (eg, silicone particles or heavy metals such as platinum) migrate into the body and can cause systemic symptoms in susceptible women, causing what is known as breast implant illness. This pilot study investigates possible associations between hair platinum levels in patients with breast implants and breast implant illness, and evaluates its possible use for diagnostic purposes. Methods: Patients were included from the silicone outpatient clinic at Amsterdam University Medical Centre. Platinum concentration in hair samples of 10 women with breast implants and systemic symptoms (group A) was compared with that in 10 women with breast implants but no symptoms (group B), and a control group of 10 women without implants or symptoms (group C), using laser ablation inductively coupled plasma mass spectrometry. Radiological imaging was used to assess implant ruptures or silicone leakage. Results: A median platinum concentration of 0.09 μg per kg [IQR 0.04-0.15] was found in group A, 0.08 μg per kg [IQR 0.04-0.12] in group B, and 0.04 μg per kg [IQR 0.02-0.13] in group C, with no statistical significant difference between the groups (Kruskal-Wallis test, P = 0.43). No correlation between radiologically proven implant leakage and platinum level was found. Conclusions: There was no statistically significant difference in hair platinum levels in women with or without silicone breast implants or breast implant illness. Therefore, based on this pilot study, we do not recommend this test for clinical use. Given the small sample size, more research is required to fully assess its possible use for diagnostic purposes.

Background: Sepsis in patients with cancer is increasingly common and associated with high mortality. To date, no studies have examined the effectiveness of prehospital antibiotics in septic patients with cancer. This study aimed to compare survival of septic patients with cancer to those without and to evaluate the effect of prehospital antibiotics in septic patients with cancer. Methods: We conducted a post-hoc sub-analysis of the PHANTASi (PreHospital ANTibioitcs Against Sepsis) trial database: a randomised controlled trial which enrolled patients with suspected sepsis who were transported to the emergency department by ambulance. Patients in the intervention group were administered prehospital intravenous antibiotics while those in the control group received usual care. We compared patients who had cancer to those who did not. Primary outcome was 28-day mortality; among the secondary outcomes, we included in-hospital mortality and 90-day mortality. Results: 357(13.4%) of the 2658 included patients had cancer in the past five years, of which, 209 (58.5%) were included in the intervention and 148 (41.5%) usual care groups; 28-day mortality was significantly higher in patients who were diagnosed with cancer in the past five years than those without cancer in the past five years: 15.2% vs. 7.1%, respectively (p < 0.001). Prehospital antibiotics in the group of patients with cancer in the last five years yielded no significant effect on survival. There were however, significantly fewer 30-day readmissions (p = 0.031) in the intervention group of cancer patients (12.2% vs 5.7%). Conclusion: Prehospital antibiotics did not improve overall survival. However, there was a significant reduction in 30-day readmissions. Published version

Background: This study investigates (1) whether the hospital standardised mortality ratio (HSMR) model underestimates or overestimates disease severity and (2) the completeness of the data collected by administrators to calculate HSMR in a cohort of deceased patients with the diagnosis of pneumonia. Methods: In this retrospective cohort study Pneumonia Severity Index (PSI) and Abbreviated Mortality in Emergency Department Sepsis (abbMEDS) scores and associated mortality probabilities were obtained from 32 deceased pneumonia patients over the year 2014 in the VU University Medical Centre. These were compared with mortality probabilities of the Central Bureau for Statistics (CBS) calculated for every patient using the HSMR model. Clinical charts were examined to extract relevant comorbidities to determine the reliability of data sent to the national registration of hospital care. Results: Risk categories determined by using the PSI and the abbMEDS were significantly higher compared with the risk categories according to HSMR (p = 0.001, respectively p = 0.000). The mean difference between the number of comorbidities in our registration and the coders’ registration was 1.97 (p = 0.00). The mean difference was 0.97 (p = 0.000) for the number of comorbidities of influence on the Charlson Comorbidity Index (CCI) and 1.25 (p = 0.001) for the calculated CCI. Conclusion: The results of this study suggest that the mortality probabilities as calculated by the CBS are an underestimation of the risk of dying for each patient. Our study also showed that the registration of data sent to the CBS underestimated the actual comorbidities of the patient, and could possibly influence the HSMR.

Background: This study investigates (1) whether the hospital standardised mortality ratio (HSMR) model underestimates or overestimates disease severity and (2) the completeness of the data collected by administrators to calculate HSMR in a cohort of deceased patients with the diagnosis of pneumonia. Methods: In this retrospective cohort study Pneumonia Severity Index (PSI) and Abbreviated Mortality in Emergency Department Sepsis (abbMEDS) scores and associated mortality probabilities were obtained from 32 deceased pneumonia patients over the year 2014 in the VU University Medical Centre. These were compared with mortality probabilities of the Central Bureau for Statistics (CBS) calculated for every patient using the HSMR model. Clinical charts were examined to extract relevant comorbidities to determine the reliability of data sent to the national registration of hospital care. Results: Risk categories determined by using the PSI and the abbMEDS were significantly higher compared with the risk categories according to HSMR (p = 0.001, respectively p = 0.000). The mean difference between the number of comorbidities in our registration and the coders’ registration was 1.97 (p = 0.00). The mean difference was 0.97 (p = 0.000) for the number of comorbidities of influence on the Charlson Comorbidity Index (CCI) and 1.25 (p = 0.001) for the calculated CCI. Conclusion: The results of this study suggest that the mortality probabilities as calculated by the CBS are an underestimation of the risk of dying for each patient. Our study also showed that the registration of data sent to the CBS underestimated the actual comorbidities of the patient, and could possibly influence the HSMR.

Of the warning scores in use for recognition of high-risk patients at the Emergency Department (ED), few incorporate laboratory results. Although hematological characteristics have shown prognostic value in small studies, large studies in elderly ED populations are lacking. We studied the association between blood cell and platelet counts and characteristics as well as C-reactive protein (CRP) at ED presentation with mortality in non-multitrauma patients ≥ 65 years. Comparison between survivors and non-survivors showed small, significant differences with AUROCs ranging between 56.6% and 65.2% for 30-day mortality. Combining parameters yielded an evident improvement (AUROC of 70.4%). Efforts should be pursued to study the added value of hematological parameters on top of clinical data when assessing patient risk.

Objective: To compare the effects of central versus peripheral drug administration on the rate of return of organised electrical activity and/or spontaneous circulation during CPR. Study design: Randomized clinical trial. Study population: Hospitalized patients and patients presenting at the emergency department, older than 18 years, requiring CPR. Intervention: Central venous access Main study parameters/endpoints: Combined primary endpoint: rate of appearance of organised electrical activity or return of spontaneous circulation. Nature and extent of the burden and risks associated with participation, benefit and group relatedness: All patients are treated according to the guidelines of the European Resuscitation Council, which are endorsed by the local VUMC CPR-committee. Central access will be obtained by cannulation of the external or internal jugular vein. To avoid interference with initial management, central venous access will be obtained after initiation of chest compressions, first attempt at defibrillation (if applicable), securing the airway and obtaining a peripheral access. All resuscitated patients require vascular access and almost all successfully resuscitated patients require central venous access. Obtaining central access during CPR may be associated with a slightly higher complication rate, such as arterial puncture and pneumothorax. Possible benefits for study subjects are a higher success rate of CPR.

International audience; Background: A long completion time in the Emergency Department (ED) is associated with higher morbidity and in-hospital mortality. A completion time of more than four hours is a frequently used cutoff point. Mostly, older and sicker patients exceed a completion time of four hours on the ED. The primary aim was to examine which factors currently contribute to overcrowding and a time to completion of more than four hours on the EDs of two different hospitals, namely: the VU Medical Center (VUmc), an academic level 1 trauma centre and the St. Antonius Hospital, a large community hospital in Nieuwegein. In addition, we compared the differences between these hospitals. Methods: In this observational study, the time steps in the process of diagnosing and treatment of all patients visiting the EDs of the two hospitals were measured for four weeks. Patients triaged as Emergency Severity Index (ESI) category 2/3 or Manchester Triage System (MTS) orange/yellow were followed more closely and prospectively by researchers for detailed information in the same period from 12.00-23.00 hrs. Results: In the VUmc, 89% of the patients had a completion time of less than four hours. The average completion time (n = 2262) was 2:10 hours, (median 1:51 hours, range: 0:05-12:08). In the St. Antonius Hospital, 77% of patients had a completion time shorter than four hours (n = 1656). The average completion time in hours was 2:49 (n = 1655, median 2:34, range: 0:08-11:04). In the VUmc, a larger percentage of ESI 1, 2 and 3 patients did not achieve the 4-hour target (14%, 20% and 19%) compared with ESI 4 and 5 patients (2.7% and 0%), p < 0.001. At the St. Antonius Hospital, a greater percentage of orange and yellow categorised patients exceeded four hours on the ED (32% and 28%) compared with red (8%) and green/blue (13%), p < 0.001. For both hospitals there was a significant dependency between exceeding four hours on the ED and the following: whether a consultation was performed (p < 0.001), the number of radiology tests performed (p < 0.001), and an age above 65 years. Conclusion: Factors leading to ED stagnation were similar in both hospitals, namely old age, treatment by more than one speciality and undergoing radiological tests. Uniform remedial measures should be taken on a nationwide level to deal with these factors to reduce stagnation in the EDs. Long completion time in the emergency department (ED) can lead to overcrowding and is associated with negative outcomes, such as increased risk of hospital admission and in-hospital mortality. 1 Therefore, optimising ED patient flow is an important and frequently discussed topic. Because the frequency and type of presentations

Background: Cancer patients receiving chemotherapy in an ambulatory setting are at risk of venous thromboembolism (VTE). Randomized controlled trials have shown that thromboprophylaxis with lowmolecular- weight-heparin reduces the occurrence of VTE in cancer patients but the overall incidence of 4% is generally considered too low to justify thromboprophylaxis in all patients. A logical next step is to focus on tools for selecting cancer patients at the highest risk of VTE, in whom thromboprophylaxis is expected to have a more favorable risk-benefit ratio. Aim: We aimed to determine the performance of tissue factor (TF)- dependent microparticle coagulant activity in predicting VTE, and to compare it to that of the Khorana score, and previously evaluated biomarkers. Methods: In six hospitals blood was obtained from consenting ambulatory patients receiving chemotherapy for stage III or IV cancer. Patients were prospectively followed for 6 months for development of VTE. At inclusion, TF-dependent microparticle coagulant activity was measured in a fibrin generation test (FGT) on fresh plasma in the local laboratories. The outcome measure of this plasma recalcification test was the prolongation of the clotting time in the presence of antibody to factor VII/TF, expressed as percentage of the clotting time in the absence of antibody. The result was considered positive above 13%. Remaining plasma was frozen and stored for later central measurement of D-dimer, P-selectin, pro-thrombin fragment 1 + 2 (F1 + 2) and factor VIII; for these tests positivity cut-offs from the original publications were used. Clinical data were collected for calculation of the Khorana score, which assigns patients to three risk categories. Estimates of sensitivity and positive predictive value (PPV) were calculated, taking into account death as a competing risk, with 95% confidence intervals (CI) based on bootstrap re-sampling. Results: The prospective cohort has recruited 443 patients, with a mean age of 61 years, of which 49% women. In total, 23 patients developed VTE, after a mean time of 2.1 months (5.2%); 77 patients died, after a mean follow-up of 3.3 months (18%). Sensitivity was estimated at 61% (95% CI: 31-84%) for FGT, 84% (61-97) for D-dimer, 85% (56-96) for P-selectin, 44% (61-97) for factor VIII, and 70% (36-89) for F1 + 2. The positive predictive value was 4.7% (1.9-9.3%) for FGT, 6.9% (3.5-12) for D-dimer, 14% (4.8-27) for P-selectin, 11% (4.5-21) for factor VIII and 5.9% (2.4-12) for F1 + 2. A high Khorana score had a sensitivity of 63% (28-83) and PPV of 5.1% (1.6-12). None of the differences were statistically significant. Summary/Conclusions: There are no substantial differences in predictive performance between the studied biomarkers. None of the individual positive predictive values exceeded 15%. In this cohort of patients with advanced cancer, 18% died, which makes adjustment for death as competing risk of utmost importance. In our opinion single predictors will probably not solve the problem and combinations are likely to be helpful.

Background: Because plasma of cancer patients presenting with venous thrombosis contains high numbers of tissue factor (TF)-exposing microparticles (TF-MP1), TF-MP have been causally linked to the occurrence of venous thrombosis in cancer patients. The relationship between numbers of TF-exposing MP and the TF-MP dependent coagulant activity, however, is unclear. In addition, to which extent TF-MP originate from cancer cells is also unknown. Aim: We investigated the relationship between TF-MP numbers and coagulant activity, and the cellular origin of circulating TF-MP in cancer patients. Methods: Citrate-anticoagulated blood was collected via a single blood withdrawal from cancer patients undergoing chemotherapy. The number of TF-MP was measured by flow cytometry and TF-MP coagulant activity was measured in a fibrin generation test (FGT) in the presence or absence of an inhibitory antibody to human factor VII. Patients were categorised as having a low or high number of TF-MP (≤ or > 95th percentile), and as having low or high TF-MP coagulant activity (≤ or > 13% TF-dependent prolongation of clotting time as determined by FGT). The cellular origin was determined by flow cytometry in those patients with number of TF-MP above the 95th percentile of the total cohort. Results: Of the total cohort of 209 cancer patients, 98 had metastasis (47%). Overall, no correlation was present between the numbers of TF-MP and the coagulant activity (r = 0.029, P = 0.69). When comparing patients with a high and low number of TF-MP, respectively three of the 12 (25%) and 60 of the 192 (31%) patients had high TFMP coagulant activity. Conversely, when comparing patients with high and low TF-MP coagulant activity, respectively only three of the 63 (4.8%) and nine of the 141 (6.4%) patients had a high number of TF-exposing MP. There was a marked variation between patients with regard to the cellular origin of the TF-MP. Of the 13 patients with high TF-MP numbers, the cellular origin of TF in five patients was variable and added up to above 100%; including 67% and 44% staining for two independent tumor markers, 66% for a platelet marker, and 59% for a monocyte marker (all medians). In the other eight patients, the origin of > 25% of TF-MP could be established, and most of these TF-MP originated from platelets and no tumor-derived vesicles were detectable. Summary/Conclusions: Because increased numbers of TF-MP and coagulant activity are almost mutually exclusive, we postulate that two forms of TF associated with MP circulate in cancer patients, a coagulant form which is present in minute quantities and below the detection limit of flow cytometry, and a non-coagulant form which is detectable by flow cytometry. Taken together, although at least part of the TFMP can derive from tumor cells, the cellular origin of circulating and coagulant TF-exposing MP in cancer patients remains to be elucidated.

Erythermalgia is a rare clinical syndrome characterised by intermittent, usually symmetrical burning pain, warmth and dermal erythema of the extremities with an amelioration of discomfort by cooling of the extremity. In this report, we describe a patient with erythermalgia caused by long-term verapamil use. After discontinuing the verapamil, the symptoms improved dramatically within two weeks.

A 39-year-old woman presented with a ruptured aneurysm of the splenic artery. The postoperative course was complicated by poor wound healing. This, in combination with a history of easy bruising and joint hypermobility, made us consider a connective tissue disease as underlying cause. The vascular type of Ehlers-Danlos syndrome was diagnosed by identifying collagen III deficiency and the corresponding gene mutation in cultured fibroblasts from a skin biopsy.

Three male patients aged 29, 30 and 34 years and a 36-year-old female are reported with nasal septum perforation and a history of cocaine abuse. Two of the patients also had a perforation of the hard palate. In all four, antineutrophil cytoplasmic antibodies (ANCA) were found. One had a cytoplasmic immunofluorescence-staining pattern (c-ANCA), the other three showed a perinuclear staining pattern (p-ANCA). Furthermore, all patients were found to be nasal carriers of S. aureus. We hypothesise that tissue damage to the nasal and palatal area in patients using cocaine may partly be mediated by the presence of ANCA antibodies. Furthermore, we speculate that S. aureus facilitates the development of these ANCA antibodies.

A 51-year-old woman presented at the internal medical outpatients' clinic complaining of fatigue. She had been having frightening suicidal and homicidal thoughts for about 9 months prior to coming to the clinic. Physical examination and extensive laboratory examination revealed no abnormalities. She had been taking vitamin C and a herbal extract of St. John's wort. She was asked to stop taking this herbal extract and her suicidal and homicidal thoughts disappeared within about 3 weeks. She then recalled having used the preparation for some time 3 years earlier, which had given her similar psychiatric symptoms. At that time these symptoms had also disappeared after discontinuation of the preparation. This case illustrates that it is very important to ask about the use of non-regular medication during history taking.

Hereditary haemochromatosis is an autosomal recessive disorder, leading to progressive iron overload, which is very common among the Caucasian population. In the vast majority of the cases, the hereditary iron overload is caused by mutations in the HFE gene. Most prominently this is the homozygous Cys282Tyr mutation. We report two Dutch families in which both propositi were found to be heterozygous for CYS282Tyr in the work-up of hyperferritinaemia. Frequent phlebotomies had no effect on the ferritin level, but led to microcytic anaemia. Finally, the family history with bilateral cataracts was the clue for the correct diagnosis. Hereditary hyperferritinaemia-cataract syndrome (HHCS) is an autosomal dominant disease characterised by elevated serum ferritin levels and bilateral cataracts in the absence of iron overload. Several point mutations and deletions within the iron-responsive element (IRE) in the 5′ noncoding region of the L-ferritin gene have been found in HHCS families. In the first Dutch family a G to C transition at position 32 was found and a G to A mutation at the same location was found in the second Dutch family. In individuals with an isolated hyperferritinaemia (normal transferrin saturation), the presence of early onset (familial) cataract should raise the possibility of HHCS, even when CYs282Tyr heterozygosity is found.

WHAT IS KNOWN AND OBJECTIVE: Drugs with anticholinergic properties increase the risk of falls, delirium, chronic cognitive impairment, and mortality and counteract procholinergic medications used in the treatment of dementia. Medication review and optimisation to reduce anticholinergic burden in patients at risk is recommended by specialist bodies. Little is known how effective this review is in patients who present acutely and how often drugs with anticholinergic properties are used temporarily during an admission. The aim of the study was to describe the changes in the anticholinergic cognitive burden (ACB) in patients admitted to hospital with a diagnosis of delirium, chronic cognitive impairment or falls and to look at the temporary use of anticholinergic medications during hospital stay. METHODS: This is a multi-centre observational study that was conducted in seven different hospitals in the UK, Finland, The Netherlands and Italy. RESULTS AND DISCUSSION: 21.1% of patients had their ACB score reduced by a mean of 1.7%, 19.7% had their ACB increased by a mean of 1.6%, 22.8% of DAP naïve patients were discharged on anticholinergic medications. There was no change in the ACB scores in 59.2% of patients. 54.1% of patients on procholinergics were taking anticholinergics. Out of the 98 medications on the ACB scale, only 56 were seen. Medications with a low individual burden were accounting for 64.9% of the total burden. Anticholinergic drugs were used temporarily during the admission in 21.9% of all patients. A higher number of DAPs used temporarily during admission was associated with a higher risk of ACB score increase on discharge (OR = 1.82, 95% CI for OR: 1.36-2.45, P < .001). WHAT IS NEW AND CONCLUSION: There was no reduction in anticholinergic cognitive burden during the acute admissions. This was the same for all diagnostic subgroups. The anticholinergic load was predominantly caused by medications with a low individual burden. More than 1 in 5 patients not taking anticholinergics on admission were discharged on them and similar numbers saw temporary use of these medications during their admission. More than half of patients on cholinesterase-inhibitors were taking anticholinergics at the same time on admission, potentially directly counteracting their effects.

Readmissions are treated as adverse events in many healthcare systems. Causes can be physiological deterioration or breakdown of social support systems. We investigated data from a European multi-centre study of readmissions for changes in vital signs between index admission and readmission. Data sets were graded according to the National Early Warning Score (NEWS). Of 487 patients in whom NEWS could be calculated on discharge and again on re-admission, 39.6% had worse vital signs with a NEWS score difference ≥ 2 points while only 7.6% had improved by ≤ 2 points. Changes in individual vital signs of 20% or more were most common in respiratory rate and heart rate. Monitoring of respiratory rate and pulse rate post-discharge might predict some deteriorations.

Background: Policy makers struggle with unplanned readmissions as a quality indicator since integrating preventability in such indicators is difficult. Most studies on the preventability of readmissions questioned physicians whether they consider a given readmission to be preventable, from which conclusions on factors predicting preventable readmissions were derived. There is no literature on the interobserver agreement of physician judgement. Aim: To assess the degree of agreement among physicians regarding predictability and preventability of medical readmissions. Design: An online survey based on eight real-life case scenarios was distributed to European physicians. Methods: Physicians were requested to rate from the first four (index admission) scenarios whether they expected these patients to be readmitted within 30 days (the predictability). The remaining four cases, describing a readmission, were used to assess the preventability. The main outcome was the degree of agreement among physicians determined using the intra class correlation coefficient (ICC). Results: 526 European medical physicians completed the survey. Most physicians had internal medicine as primary specialism. The median years of clinical experience was 11. ICC for predictability of readmission was 0.67 (moderate to good) and ICC for preventability of readmission was 0.13 (poor). Conclusion: There was moderate to good agreement among physicians on the predictability of readmissions while agreement on preventability was poor. This study indicates that assessing preventability of readmissions based solely on the judgement of physicians is far from perfect. Current literature on the preventability of readmissions and conclusions derived on the basis of physician opinion should be interpreted with caution.