Major cardiovascular events and subsequent risk of kidney failure with replacement therapy: a CKD Prognosis Consortium study.

AIMS: Chronic kidney disease (CKD) increases risk of cardiovascular disease (CVD). Less is known about how CVD associates with future risk of kidney failure with replacement therapy (KFRT). METHODS AND RESULTS: The study included 25 903 761 individuals from the CKD Prognosis Consortium with known baseline estimated glomerular filtration rate (eGFR) and evaluated the impact of prevalent and incident coronary heart disease (CHD), stroke, heart failure (HF), and atrial fibrillation (AF) events as time-varying exposures on KFRT outcomes. Mean age was 53 (standard deviation 17) years and mean eGFR was 89 mL/min/1.73 m2, 15% had diabetes and 8.4% had urinary albumin-to-creatinine ratio (ACR) available (median 13 mg/g); 9.5% had prevalent CHD, 3.2% prior stroke, 3.3% HF, and 4.4% prior AF. During follow-up, there were 269 142 CHD, 311 021 stroke, 712 556 HF, and 605 596 AF incident events and 101 044 (0.4%) patients experienced KFRT. Both prevalent and incident CVD were associated with subsequent KFRT with adjusted hazard ratios (HRs) of 3.1 [95% confidence interval (CI): 2.9-3.3], 2.0 (1.9-2.1), 4.5 (4.2-4.9), 2.8 (2.7-3.1) after incident CHD, stroke, HF and AF, respectively. HRs were highest in first 3 months post-CVD incidence declining to baseline after 3 years. Incident HF hospitalizations showed the strongest association with KFRT [HR 46 (95% CI: 43-50) within 3 months] after adjustment for other CVD subtype incidence. CONCLUSION: Incident CVD events strongly and independently associate with future KFRT risk, most notably after HF, then CHD, stroke, and AF. Optimal strategies for addressing the dramatic risk of KFRT following CVD events are needed.

Identifying Profiles of Patients With Uncontrolled Type 2 Diabetes Who Would Benefit From Referral to an Endocrinologist.

OBJECTIVE: To compare the outcomes of glycemic uncontrolled diabetes mellitus type 2 patients receiving treatment from endocrinologists with those treated by primary care physicians. Additionally, this research aims to identify patient profiles that would benefit from personalized referral-a novel medical approach that aims to match the most suitable specialist for effectively managing patient while considering the patient's profile. METHODS: This retrospective cohort study uses the Maccabi Healthcare Services diabetes registry to match 508 pairs of glycemic uncontrolled diabetes mellitus type 2 patients treated by endocrinologists (EndoG) and primary care physicians (PcPG). Using a generalized additive model, we analyzed the hemoglobin A1c (HbA1c) trend over 1 year for each group. We employed the odds ratio (OR) from conditional logistic regression to determine the likelihood of favorable outcomes in the EndoG compared to the PcPG, using the entire cohort and subcohort profiles. RESULTS: The generalized additive model comparison indicated an improvement in HbA1c levels in both groups, with the EndoG outperforming the PcPG. Furthermore, the EndoG group had an OR = 2.27 (95% confidence interval, 1.6 to 3.2) for reducing HbA1c by at least 1% within a year and an OR = 1.68 (95% confidence interval, 1.02 to 2.76) for achieving low-density lipoprotein levels< 100 mg/dl. We identified 96 profiles with positive outcomes, all favoring treatment by endocrinologists. CONCLUSIONS: EndoG demonstrated superior HbA1c control over time and achieved better outcomes compared to PcPG. The identification of 96 profiles benefiting from endocrinologist referral emphasizes the potential of personalized referral.

The Association Between Depression and Invasive and In-situ Cervical Tumors: A Large Population Based Cohort Study.

BACKGROUND: Depression has been shown to be associated with cervical tumors (CTs), an association mostly demonstrated in studies in which temporality could not have been ascertained. OBJECTIVES: To study the association between depression and CTs and the influence of co-morbidities of this association in a large cohort study. METHODS: A retrospective computer-based cohort study was conducted. The cohort included 357,450 female members of Maccabi Healthcare Services. The cohort was classified as depressed or non-depressed using the International Classification of Diseases 9/10 codes. For each subgroup, demographic characteristics, behavioral characteristics, co-morbidities, and CTs diagnosis were obtained. The burden of co-morbidities was defined as the sum of major co-morbidities. We used zero-inflated negative binomial regression analysis due to over-dispersion to estimate the relative risk (RR) for CTs with 95% confidence interval (95%CI). RESULTS: Depression was diagnosed in 15,789 women. Among this group, CTs were diagnosed in 1585 (10.0%). Among the 341,661 non-depressed, CTs were diagnosed in 4185 (1.2%). After adjustment to age and socioeconomic status, the association between depression and CTs was RR=9.2 (95%CI 8.7-9.9, P-value < 0.0001). The association between depression and CTs increased as the burden of clinical conditions increased (P-value < 0.0001). CONCLUSIONS: Women with depression are at a higher risk for CTs, especially among those who have several co-morbidities. Tighter gynecology surveillance is crucial among these women.

Epidemiology of treatment resistant depression among major depressive disorder patients in Israel.

INTRODUCTION: Major depressive disorder (MDD) is one of the most common mental disorders worldwide, estimated to affect 10-15% of the population per year. Treatment resistant depression (TRD) is estimated to affect a third of these patients who show difficulties in social and occupational function, decline of physical health, suicidal thoughts and increased health care utilization. We describe the prevalence of MDD, TRD and associated healthcare resource utilization in Maccabi Healthcare Services (MHS), a 2.5 million-member state-mandated health service in Israel. METHODS: All MHS members with an MDD diagnosis were identified within the years 2017-2018 and prevalence assessed by age, sex and TRD. To assess the incidence of MDD, members aged 18-65 years at the start of any MDD episode were identified between 1st January 2016 and 31st May 2018 with at least one systemic first-line antidepressant treatment within three months before or after the initial episode. Treatment patterns, time on first-line treatment, and healthcare resource utilization were compared by TRD. RESULTS: A total of 4960 eligible MDD patients were identified (median age = 51 years, 65% female), representing a period prevalence of 0.218%, and of those, a high proportion of patients received drug treatment (92%). Among incident MDD cases (n = 2553), 24.4% had TRD. Factors associated with TRD included increasing age and personality disorder. Median time on treatment was 3.7 months (longer for those without TRD than those with) and 81.9% of patients purchased more than one month's supply of therapy. In the year after index, patients with TRD had a significant increased number of visits to primary care physicians, psychiatrists, emergency room visits, general hospitalizations, and psychiatric hospitalizations. CONCLUSION: Our study shows that prevalence of MDD in Israel is low compared to other countries, however once diagnosed, patients' are likely to receive drug treatment. Among patients diagnosed with MDD, the proportion of TRD is similar to other countries, increases with age and is associated with increased healthcare utilization, therefore should be a focus of continued research for finding effective long term treatment options.

Associations of Maternal Androgen-Related Conditions With Risk of Autism Spectrum Disorder in Progeny and Mediation by Cardiovascular, Metabolic, and Fertility Factors.

Fetal exposure to elevated androgens is thought to contribute to autism spectrum disorder (ASD) risk. However, data rely heavily on in utero androgens measurements, which also reflect fetal secretions. Thus, in utero hyperandrogenemia might indicate adverse autism-related neurogenesis that has already occurred affecting fetal androgen homeostasis, rather than being a cause of the disorder. Associations between maternal androgen-related conditions and ASD could more directly implicate androgens' etiological role. We examined the association between maternal hyperandrogenemia-related conditions, focusing primarily on polycystic ovarian syndrome (PCOS), and progeny ASD, in an Israeli cohort of 437,222 children born in 1999-2013. Odds ratios and 95% confidence intervals were estimated using generalized estimating equations. Multiple mediation analyses using natural effect models were conducted to evaluate combined mediation of the PCOS effect by androgen-related cardiovascular, metabolic, and fertility factors. Results indicated that children of mothers with PCOS had higher ASD odds compared with children of mothers without PCOS (odds ratio = 1.42, 95% confidence interval: 1.24,1.64), and this effect was only partly mediated by the factors considered. Elevated odds were also observed for other hyperandrogenemia-related conditions. Findings provide support for direct involvement of maternal hyperandrogenemia in ASD etiology. Alternatively, findings might reflect shared genetic and/or environmental factors independently affecting maternal androgen homeostasis and fetal neurodevelopment.

The impact of a Facebook campaign among mothers on HPV vaccine uptake among their daughters: A randomized field study.

OBJECTIVE: The popularity of social networks provide an incredible opportunity to enhance the impact of preventive medicine programs. We aimed to assess whether a targeted Facebook campaign among mothers may increase the uptake of human Papilloma virus (HPV) immunization among their 8th-grade daughters. METHODS: This field study was conducted among the members of a state-mandated health organization in Israel. Included were all 21,592 members who were mothers to 14 year-old daughters in the 2018-19 school-year. A total of 17,271 (80%) were randomly allocated to the campaign arm and the rest (n=4,321) were selected as a reference group. The Facebook ads addressed issues and concerns regarding HPV-related diseases and HPV vaccine. Main outcome measures were Facebook metrics on exposure to campaign and HPV immunization among eighth grade daughters of the study participants. RESULTS: Between 8/2018-10/2018, Facebook ads were shown 1.8-million times (a reach of 88%). The uptake of HPV vaccine among daughters of women allocated to the campaign arm (55.3%) was similar (p = 0.749) to 55.0% in the control group. The only significant differences between study groups were observed when stratifying by SES level. In the lowest SES quartile, Facebook campaign significantly (p = .02) reduced vaccine uptake (35% vs. 39.0%), with a relative risk of 0.90 (95%CI: 0.82-0.98), while in the second SES quartile, Facebook campaign increased vaccine uptake from 52.6% to 55.8%, with a RR of 1.06 (95%CI,1.00-1.12). Among mothers in higher SES levels, daughters of exposed and unexposed mothers had similar immunization rates. CONCLUSIONS: Facebook campaign may increase the uptake of HPV vaccine among daughters to mothers of medium-to-low SES level, but it may reduce vaccination among lower SES groups.

Shifting from vitamin K antagonists to non-vitamin K antagonist oral anticoagulants in patients with atrial fibrillation: predictors, patterns and temporal trends.

BACKGROUND: Non-Vitamin K antagonist oral anticoagulants (NOACs) emerged as an alternative with comparable or superior efficacy and safety to vitamin K antagonists (VKAs) for stroke prevention in patients with non-valvular atrial fibrillation (AF). OBJECTIVES: The aim of the current study was to investigate the patterns, predictors, timelines and temporal trends of shifting from VKAs to NOACs. METHODS: In this retrospective observational study, the computerized database of a large healthcare provider in Israel, Maccabi Healthcare Services, was searched to identify patients with AF for whom either a VKA or NOAC was prescribed between 2012 and 2015. Time from diagnosis to therapy initiation and to shifting between therapies was evaluated. RESULTS: Out of 6987 eligible AF incident patients, 2338 (33.4%) initiated treatment with a VKA and 2221 (31.7%) with a NOAC. In addition, 5259 prevalent patients were analyzed. During the study period, NOAC prescriptions proportion among the newly diagnosed cases increased from 32 to 68.4% (p for trend <  0.001). The median time from diagnosis to first dispensing was greater in NOAC than VKA and decreased among patients treated with NOAC during the study period (2012: 1.9 and 0.3 months, 2015: 0.7 and 0.2 months, respectively). During follow-up, 3737 (49%) patients (54.3% and 47.1% of the incident and prevalent cases, respectively), shifted from a VKA to a NOAC, after a median of 22 months and 39 months in the incident and prevalent cases, respectively, decreasing throughout the study period. Female gender, younger age, southern district, higher CHADS2 and CHA2DS2-VASC score, non-smoking, and treatment with antiplatelets were associated with a greater likelihood for therapy shift. Shifting from a NOAC to a VKA decreased over time from 8 to 4.5% in 2012 to 0.5% and 0.7% in 2015 in the incident and prevalent groups, p <  0.001 respectively. CONCLUSIONS: Shifting from VKA to NOAC occurred in 50% of the cases, more frequently among incident cases, and younger patients with greater stroke risk. Shifting from a NOAC to a VKA was much less frequent, yet it occurred more often in incident cases and decreased over time. A socially and economically sensitive program to optimize the initiation of OAC therapy upon diagnosis is warranted.

Maternal Thyroid Disorders and Risk of Autism Spectrum Disorder in Progeny.

BACKGROUND: Maternal thyroid dysfunction is suspected of causing adverse neurodevelopmental effects, but current evidence is inconclusive. Epidemiologic investigations generally suggest an association between maternal thyroid dysfunction and neurodevelopment impairments in progeny, but clinical trials of thyroid treatment during pregnancy reported null effects. To better understand these discrepant findings, we evaluated the association between maternal thyroid conditions and autism spectrum disorder (ASD), including examining the role of gestational thyroid-related hormone concentrations and thyroid medications use. METHODS: Analyses considered 437,222 singleton live births occurring in a large Israeli health fund in 1999-2013, followed through 2016. Thyroid conditions and ASD cases were identified through International Classification of Diseases-9 codes with subsequent validation through review of medical records. Laboratory gestational thyroid hormone measurements were also considered. RESULTS: Children of mothers who ever experienced hypothyroidism had a higher risk of ASD compared with children of mothers without hypothyroidism (adjusted odds ratio [aOR] = 1.26, 95% confidence interval [CI] = 1.12, 1.42). The association with hyperthyroidism was less consistent, but elevated in main analyses (aOR = 1.42, 95% CI = 1.04, 1.94). These associations were not explained by maternal gestational thyroid hormones levels nor mitigated by gestational use of thyroid medications. CONCLUSIONS: Results indicate that maternal thyroid conditions are associated with increased ASD risk in progeny, but suggestively not due to direct effects of thyroid hormones. Instead, factors that influence maternal thyroid function could have etiologic roles in ASD through pathways independent of maternal gestational thyroid hormones and thus be unaffected by medication treatment. Factors known to disrupt thyroid function should be examined for possible involvement in ASD etiology.

Prediction of progression from pre-diabetes to diabetes: Development and validation of a machine learning model.

AIMS: Identification, a priori, of those at high risk of progression from pre-diabetes to diabetes may enable targeted delivery of interventional programmes while avoiding the burden of prevention and treatment in those at low risk. We studied whether the use of a machine-learning model can improve the prediction of incident diabetes utilizing patient data from electronic medical records. METHODS: A machine-learning model predicting the progression from pre-diabetes to diabetes was developed using a gradient boosted trees model. The model was trained on data from The Health Improvement Network (THIN) database cohort, internally validated on THIN data not used for training, and externally validated on the Canadian AppleTree and the Israeli Maccabi Health Services (MHS) data sets. The model's predictive ability was compared with that of a logistic-regression model within each data set. RESULTS: A cohort of 852 454 individuals with pre-diabetes (glucose ≥ 100 mg/dL and/or HbA1c ≥ 5.7) was used for model training including 4.9 million time points using 900 features. The full model was eventually implemented using 69 variables, generated from 11 basic signals. The machine-learning model demonstrated superiority over the logistic-regression model, which was maintained at all sensitivity levels - comparing AUC [95% CI] between the models; in the THIN data set (0.865 [0.860,0.869] vs 0.778 [0.773,0.784] P < .05), the AppleTree data set (0.907 [0.896, 0.919] vs 0.880 [0.867, 0.894] P < .05) and the MHS data set (0.925 [0.923, 0.927] vs 0.876 [0.872, 0.879] P < .05). CONCLUSIONS: Machine-learning models preserve their performance across populations in diabetes prediction, and can be integrated into large clinical systems, leading to judicious selection of persons for interventional programmes.

Secular trends in testosterone- findings from a large state-mandate care provider.

BACKGROUND: Several studies from the US and Europe have shown a population-level decline in serum testosterone in men from 1970's to early 2000's. However, to the best of our knowledge, no study examining population-level decline in testosterone has been published in more recent years. The study objective was therefore to examine secular trends in testosterone levels among Israeli men in the first and second decades of the twenty-first century, METHODS: All incident total testosterone performed between1/2006 and 3/2019 among 102,334 male members of a large health organization. RESULTS: A significant (p < 0.001) and prominent trend of age-independent decline in the testosterone levels was recorded during the study period for most age groups. CONCLUSIONS: There was a highly significant age-independent decline in total testosterone in the first and second decades of the twenty-first century. The decline was unlikely to be explained by increasing rates of obesity.

Postnatal steroid therapy is associated with autism spectrum disorder in children and adolescents of very low birth weight infants.

BACKGROUND: This study evaluates the association between major neonatal morbidities and autism spectrum disorder (ASD) in children and adolescents born of very low birth weight (VLBW). METHODS: Historical cohort study using the Israel national VLBW infant database linked with the Maccabi Healthcare Services (MHS) medical records. The study cohort comprised 4963 VLBW subjects born from 1999 to 2012, >1 year of age. Multivariable logistic regression analyses were used to assess factors associated with ASD. RESULTS: The diagnosis of ASD was confirmed in 113 children (2.3%). Infants with major neonatal morbidities had higher rates of ASD; however, in the multivariable analyses these were not significantly associated with ASD: severe intraventricular hemorrhage (OR 1.21 [95% CI 0.60-2.45]), post-hemorrhagic hydrocephalus (OR 1.77 [0.73-4.29]), periventricular leukomalacia (OR 1.02 [0.42-2.51]), severe retinopathy of prematurity (OR 1.91 [0.995-3.67]), and bronchopulmonary dysplasia (OR 1.44 [0.84-2.45]). Postnatal steroid therapy when included separately was associated with an OR of 1.97 [1.18-3.29] for ASD. This association remained significant when postnatal steroid therapy was included with each of the neonatal morbidities (ORs ranging from 1.91 to 2.11). CONCLUSIONS: This study suggests a significant association between postnatal steroid therapy and ASD in VLBW infants. This possible association should be considered in future studies evaluating potential risk factors for ASD in preterm infants.

Very Low Birth Weight Preterm Infants have Decreased Celiac Disease Autoimmunity During Childhood and Adolescence.

Little is known about the effect of prematurity on later development of celiac disease (CD). We conducted a retrospective analysis of real-world data examining the association between very low birth weight (VLBW) prematurity and later development of CD autoimmunity (CDA) in 3580 infants born between years 2000 and 2012 and their matched controls. At a median of 12 years, VLBW prematurity was negatively associated with later development of CDA with a cumulative prevalence of 5.9 per 1000 versus 10.3 per 1000 (P = 0.02), though more former VLBW premature infants were ever tested for CDA (48.5% vs 37.4%, P < 0.001). The odds ratio for developing CDA among children born preterm at VLBW was 0.57 (95% confidence interval (CI) 0.35-0.92) as compared with matched controls. There was no difference in clinical characteristics of CDA between both groups. In conclusion, VLBW preterm infants present a decreased risk for the development of CDA during childhood and adolescence.

Personal and social patterns predict influenza vaccination decision.

BACKGROUND: Seasonal influenza vaccination coverage remains suboptimal in most developed countries, despite longstanding recommendations of public health organizations. The individual's decision regarding vaccination is located at the core of non-adherence. We analyzed large-scale data to identify personal and social behavioral patterns for influenza vaccination uptake, and develop a model to predict vaccination decision of individuals in an upcoming influenza season. METHODS: We analyzed primary data from the electronic medical records of a retrospective cohort of 250,000 individuals between the years 2007 and 2017, collected from 137 clinics. Individuals were randomly sampled from the database of Maccabi Healthcare Services. Maccabi's clients are representative of the Israeli population, reflect all demographic, ethnic, and socioeconomic groups and levels. We used several machine-learning models to predict whether a patient would get vaccinated in the future. Models' performance was evaluated based on the area under the ROC curve. RESULTS: The vaccination decision of an individual can be explained in two dimensions, Personal and social. The personal dimension is strongly shaped by a "default" behavior, such as vaccination timing in previous seasons and general health consumption, but can also be affected by temporal factors such as respiratory illness in the prior year. In the social dimension, a patient is more likely to become vaccinated in a given season if at least one member of his family also became vaccinated in the same season. Vaccination uptake was highly assertive with age, socioeconomic score, and geographic location. An XGBoost-based predictive model achieved an ROC-AUC score of 0.91 with accuracy and recall rates of 90% on the test set. Prediction relied mainly on the patient's individual and household vaccination status in the past, age, number of encounters with the healthcare system, number of prescribed medications, and indicators of chronic illnesses. CONCLUSIONS: Our ability to make an excellent prediction of the patient's decision sets a major step toward personalized influenza vaccination campaigns, and will help shape the next generation of targeted vaccination efforts.

Evaluating Glomerular Filtration Rate Slope as a Surrogate End Point for ESKD in Clinical Trials: An Individual Participant Meta-Analysis of Observational Data.

BACKGROUND: Decline in eGFR is a biologically plausible surrogate end point for the progression of CKD in clinical trials. However, it must first be tested to ensure strong associations with clinical outcomes in diverse populations, including patients with higher eGFR. METHODS: To investigate the association between 1-, 2-, and 3-year changes in eGFR (slope) with clinical outcomes over the long term, we conducted a random effects meta-analysis of 3,758,551 participants with baseline eGFR≥60 ml/min per 1.73 m2 and 122,664 participants with eGFR<60 ml/min per 1.73 m2 from 14 cohorts followed for an average of 4.2 years. RESULTS: Slower eGFR decline by 0.75 ml/min per 1.73 m2 per year over 2 years was associated with lower risk of ESKD in participants with baseline eGFR≥60 ml/min per 1.73 m2 (adjusted hazard ratio, 0.70; 95% CI, 0.68 to 0.72) and eGFR<60 ml/min per 1.73 m2 (0.71; 95% CI, 0.68 to 0.74). The relationship was stronger with 3-year slope. For a rapidly progressing population with predicted 5-year risk of ESKD of 8.3%, an intervention that reduced eGFR decline by 0.75 ml/min per 1.73 m2 per year over 2 years would reduce the ESKD risk by 1.6%. For a hypothetical low-risk population with a predicted 5-year ESKD risk of 0.58%, the same intervention would reduce the risk by only 0.13%. CONCLUSIONS: Slower decline in eGFR was associated with lower risk of subsequent ESKD, even in participants with eGFR≥60 ml/min per 1.73 m2, but those with the highest risk would be expected to benefit the most.

Kidney-Failure Risk Projection for the Living Kidney-Donor Candidate.

BACKGROUND: Evaluation of candidates to serve as living kidney donors relies on screening for individual risk factors for end-stage renal disease (ESRD). To support an empirical approach to donor selection, we developed a tool that simultaneously incorporates multiple health characteristics to estimate a person's probable long-term risk of ESRD if that person does not donate a kidney. METHODS: We used risk associations from a meta-analysis of seven general population cohorts, calibrated to the population-level incidence of ESRD and mortality in the United States, to project the estimated long-term incidence of ESRD among persons who do not donate a kidney, according to 10 demographic and health characteristics. We then compared 15-year projections with the observed risk among 52,998 living kidney donors in the United States. RESULTS: A total of 4,933,314 participants from seven cohorts were followed for a median of 4 to 16 years. For a 40-year-old person with health characteristics that were similar to those of age-matched kidney donors, the 15-year projections of the risk of ESRD in the absence of donation varied according to race and sex; the risk was 0.24% among black men, 0.15% among black women, 0.06% among white men, and 0.04% among white women. Risk projections were higher in the presence of a lower estimated glomerular filtration rate, higher albuminuria, hypertension, current or former smoking, diabetes, and obesity. In the model-based lifetime projections, the risk of ESRD was highest among persons in the youngest age group, particularly among young blacks. The 15-year observed risks after donation among kidney donors in the United States were 3.5 to 5.3 times as high as the projected risks in the absence of donation. CONCLUSIONS: Multiple demographic and health characteristics may be used together to estimate the projected long-term risk of ESRD among living kidney-donor candidates and to inform acceptance criteria for kidney donors. (Funded by the National Institute of Diabetes and Digestive and Kidney Diseases and others.).

Predicting Breast Cancer by Applying Deep Learning to Linked Health Records and Mammograms.

Background Computational models on the basis of deep neural networks are increasingly used to analyze health care data. However, the efficacy of traditional computational models in radiology is a matter of debate. Purpose To evaluate the accuracy and efficiency of a combined machine and deep learning approach for early breast cancer detection applied to a linked set of digital mammography images and electronic health records. Materials and Methods In this retrospective study, 52 936 images were collected in 13 234 women who underwent at least one mammogram between 2013 and 2017, and who had health records for at least 1 year before undergoing mammography. The algorithm was trained on 9611 mammograms and health records of women to make two breast cancer predictions: to predict biopsy malignancy and to differentiate normal from abnormal screening examinations. The study estimated the association of features with outcomes by using t test and Fisher exact test. The model comparisons were performed with a 95% confidence interval (CI) or by using the DeLong test. Results The resulting algorithm was validated in 1055 women and tested in 2548 women (mean age, 55 years ± 10 [standard deviation]). In the test set, the algorithm identified 34 of 71 (48%) false-negative findings on mammograms. For the malignancy prediction objective, the algorithm obtained an area under the receiver operating characteristic curve (AUC) of 0.91 (95% CI: 0.89, 0.93), with specificity of 77.3% (95% CI: 69.2%, 85.4%) at a sensitivity of 87%. When trained on clinical data alone, the model performed significantly better than the Gail model (AUC, 0.78 vs 0.54, respectively; P < .004). Conclusion The algorithm, which combined machine-learning and deep-learning approaches, can be applied to assess breast cancer at a level comparable to radiologists and has the potential to substantially reduce missed diagnoses of breast cancer. © RSNA, 2019 Online supplemental material is available for this article.

Acute renal outcomes with sodium-glucose co-transporter-2 inhibitors: Real-world data analysis.

AIM: To assess the possible risk of acute kidney injury (AKI) with the use of sodium-glucose co-transporter-2 inhibitors (SGLT2-i) as well as changes in estimated glomerular filtration rate (eGFR), hospitalizations and mortality in a real-world setting. MATERIALS AND METHODS: Included in this historical cohort study were patients with type 2 diabetes in a large health organization in Israel who initiated therapy with SGLT2-i or dipeptidyl peptidase-4 inhibitors (DPP-4i) during 1 April 2015 to 30 June 2017. We collected data on serum creatinine measurements taken between 180 days prior to and 24 weeks after therapy initiation. Study endpoints included ≥30% reduction in eGFR, hospitalization with AKI, any hospitalization and all-cause mortality. RESULTS: Overall 6418 and 5604 patients initiated SGLT2-i and DPP-4i, respectively. Baseline mean (SD) eGFR was higher among the SGLT2-i group compared with the DPP-4i group (88.3 [17.4] and 82.8 [23.7], respectively) but were similar when stratifying by chronic kidney disease (CKD) stages. The adjusted odds ratio (OR) (95% confidence interval [CI]) for ≥30% reduction in eGFR with SGLT2-i versus DPP4-i was 0.70 (0.49-1.00) and ORs ranged from 1.97 (0.62-6.26) to 0.45 (0.21-0.99) in patients with baseline eGFR 30 to 45 and ≥90 mL/min/1.73 m2 , respectively. Risks of AKI (OR = 0.47, 95% CI 0.27-0.80), hospitalization (OR = 0.66, 95% CI 0.56-0.78) or all-cause mortality (OR = 0.43, 95% CI 0.20-0.95) were lower in patients initiating SGLT2-i versus DPP-4i. CONCLUSIONS: This real-world data analysis supports reassuring findings from previous randomized clinical trials showing no increased AKI risk among SGLT2-i users. Nevertheless, because of the more prominent decrease in eGFR in patients with moderate CKD, cautious use of SGLT2-i in patients with reduced eGFR is advised.

Helicobacter pylori infection and prevalence of stroke.

BACKGROUND: Helicobacter pylori causes peptic ulcer disease; however, conflicting evidence exists regarding its role in extragastric conditions. We aimed to examine associations of H pylori infection and peptic ulcer disease with stroke. METHODS: A cross-sectional study was undertaken using data of 147 936 individuals aged 25-95 years who underwent the urea breath test during 2002-2012, based on the computerized database of the second largest health maintenance organization in Israel. Logistic regression models were fitted to control for potential confounders. RESULTS: Overall, 1397 (0.9%) patients had stroke and 76 965 (52.0%) had a H pylori positive test. The likelihood of prevalent stroke increased in relation to H pylori infection: adjusted odds ratio (aOR) 1.16 (95% confidence intervals [CI]: 1.04-1.29), gastric ulcer: aOR 1.50 (95% CI: 1.18-1.91), and duodenal ulcer: aOR 1.25 (95% CI: 1.07-1.46). CONCLUSIONS: The results support the premise that stroke may be associated with a history of H pylori infection.

Serum potassium and adverse outcomes across the range of kidney function: a CKD Prognosis Consortium meta-analysis.

Aims: Both hypo- and hyperkalaemia can have immediate deleterious physiological effects, and less is known about long-term risks. The objective was to determine the risks of all-cause mortality, cardiovascular mortality, and end-stage renal disease associated with potassium levels across the range of kidney function and evaluate for consistency across cohorts in a global consortium. Methods and results: We performed an individual-level data meta-analysis of 27 international cohorts [10 general population, 7 high cardiovascular risk, and 10 chronic kidney disease (CKD)] in the CKD Prognosis Consortium. We used Cox regression followed by random-effects meta-analysis to assess the relationship between baseline potassium and adverse outcomes, adjusted for demographic and clinical characteristics, overall and across strata of estimated glomerular filtration rate (eGFR) and albuminuria. We included 1 217 986 participants followed up for a mean of 6.9 years. The average age was 55 ± 16 years, average eGFR was 83 ± 23 mL/min/1.73 m2, and 17% had moderate- to-severe increased albuminuria levels. The mean baseline potassium was 4.2 ± 0.4 mmol/L. The risk of serum potassium of >5.5 mmol/L was related to lower eGFR and higher albuminuria. The risk relationship between potassium levels and adverse outcomes was U-shaped, with the lowest risk at serum potassium of 4-4.5 mmol/L. Compared with a reference of 4.2 mmol/L, the adjusted hazard ratio for all-cause mortality was 1.22 [95% confidence interval (CI) 1.15-1.29] at 5.5 mmol/L and 1.49 (95% CI 1.26-1.76) at 3.0 mmol/L. Risks were similar by eGFR, albuminuria, renin-angiotensin-aldosterone system inhibitor use, and across cohorts. Conclusions: Outpatient potassium levels both above and below the normal range are consistently associated with adverse outcomes, with similar risk relationships across eGFR and albuminuria.

Development of Risk Prediction Equations for Incident Chronic Kidney Disease.

Importance: Early identification of individuals at elevated risk of developing chronic kidney disease (CKD) could improve clinical care through enhanced surveillance and better management of underlying health conditions. Objective: To develop assessment tools to identify individuals at increased risk of CKD, defined by reduced estimated glomerular filtration rate (eGFR). Design, Setting, and Participants: Individual-level data analysis of 34 multinational cohorts from the CKD Prognosis Consortium including 5 222 711 individuals from 28 countries. Data were collected from April 1970 through January 2017. A 2-stage analysis was performed, with each study first analyzed individually and summarized overall using a weighted average. Because clinical variables were often differentially available by diabetes status, models were developed separately for participants with diabetes and without diabetes. Discrimination and calibration were also tested in 9 external cohorts (n = 2 253 540). Exposures: Demographic and clinical factors. Main Outcomes and Measures: Incident eGFR of less than 60 mL/min/1.73 m2. Results: Among 4 441 084 participants without diabetes (mean age, 54 years, 38% women), 660 856 incident cases (14.9%) of reduced eGFR occurred during a mean follow-up of 4.2 years. Of 781 627 participants with diabetes (mean age, 62 years, 13% women), 313 646 incident cases (40%) occurred during a mean follow-up of 3.9 years. Equations for the 5-year risk of reduced eGFR included age, sex, race/ethnicity, eGFR, history of cardiovascular disease, ever smoker, hypertension, body mass index, and albuminuria concentration. For participants with diabetes, the models also included diabetes medications, hemoglobin A1c, and the interaction between the 2. The risk equations had a median C statistic for the 5-year predicted probability of 0.845 (interquartile range [IQR], 0.789-0.890) in the cohorts without diabetes and 0.801 (IQR, 0.750-0.819) in the cohorts with diabetes. Calibration analysis showed that 9 of 13 study populations (69%) had a slope of observed to predicted risk between 0.80 and 1.25. Discrimination was similar in 18 study populations in 9 external validation cohorts; calibration showed that 16 of 18 (89%) had a slope of observed to predicted risk between 0.80 and 1.25. Conclusions and Relevance: Equations for predicting risk of incident chronic kidney disease developed from more than 5 million individuals from 34 multinational cohorts demonstrated high discrimination and variable calibration in diverse populations. Further study is needed to determine whether use of these equations to identify individuals at risk of developing chronic kidney disease will improve clinical care and patient outcomes.