Hydroxychloroquine Dose and Risk for Incident Retinopathy : A Cohort Study.

BACKGROUND: Hydroxychloroquine is recommended for all patients with systemic lupus erythematosus and is often used for other inflammatory conditions, but a critical long-term adverse effect is vision-threatening retinopathy. OBJECTIVE: To characterize the long-term risk for incident hydroxychloroquine retinopathy and examine the degree to which average hydroxychloroquine dose within the first 5 years of treatment predicts this risk. DESIGN: Cohort study. SETTING: U.S. integrated health network. PARTICIPANTS: All patients aged 18 years or older who received hydroxychloroquine for 5 or more years between 2004 and 2020 and had guideline-recommended serial retinopathy screening. MEASUREMENTS: Hydroxychloroquine dose was assessed from pharmacy dispensing records. Incident hydroxychloroquine retinopathy was assessed by central adjudication of spectral domain optical coherence tomography with severity assessment (mild, moderate, or severe). Risk for hydroxychloroquine retinopathy was estimated over 15 years of use according to hydroxychloroquine weight-based dose (>6, 5 to 6, or ≤5 mg/kg per day) using the Kaplan-Meier estimator. RESULTS: Among 3325 patients in the primary study population, 81 developed hydroxychloroquine retinopathy (56 mild, 17 moderate, and 8 severe), with overall cumulative incidences of 2.5% and 8.6% at 10 and 15 years, respectively. The cumulative incidences of retinopathy at 15 years were 21.6% for higher than 6 mg/kg per day, 11.4% for 5 to 6 mg/kg per day, and 2.7% for 5 mg/kg per day or lower. The corresponding risks for moderate to severe retinopathy at 15 years were 5.9%, 2.4%, and 1.1%, respectively. LIMITATION: Possible misclassifications of dose due to nonadherence to filled prescriptions. CONCLUSION: In this large, contemporary cohort with active surveillance retinopathy screening, the overall risk for hydroxychloroquine retinopathy was 8.6% after 15 years, and most cases were mild. Higher hydroxychloroquine dose was associated with progressively greater risk for incident retinopathy. PRIMARY FUNDING SOURCE: National Institutes of Health.

Quantification of risk factors for diabetic retinopathy progression.

AIMS: To determine the importance of blood sugar control, blood pressure, and other key systemic factors on the risk of progression from no retinopathy to various stages of diabetic retinopathy. METHODS: Restrospective cohort analysis of patients (N = 99, 280) in the Kaiser Permanente Northern California healthcare system with a baseline retina photographic screening showing no evidence of retinopathy and a minimum follow-up surveillance period of 3 years from 2008 to 2019. We gathered longitudinal data on diabetic retinopathy progression provided by subsequent screening fundus photographs and data captured in the electronic medical record over a mean surveillance of 7.3 ± 2.2 (mean ± SD) years. Progression from an initial state of no diabetic retinopathy to any of four outcomes was determined: (1) any incident retinopathy, (2) referable (moderate or worse) retinopathy, (3) diabetic macular edema, and (4) proliferative diabetic retinopathy. Multiple predictors, including age, race, gender, glycosylated hemoglobin (HbA1c), systolic blood pressure (SBP), cholesterol, chronic renal disease, and type of diabetes were investigated. RESULTS: Among modifiable risk factors, the average HbA1c had the strongest impact on the progression of diabetic retinopathy, followed by average SBP control and total cholesterol. Patients with an average HbA1c of 10.0% or greater (≥ 97 mmol/mol) had a risk ratio of 5.72 (95% CI 5.44-6.02) for progression to any retinopathy, 18.84 (95% CI 17.25-20.57) for referable retinopathy, 22.85 (95% CI 18.87-27.68) for diabetic macular edema, and 25.96 (95% CI 18.75-36.93) for proliferative diabetic retinopathy compared to those with an average HbA1c of 7.0% (53 mmol/mol) or less. Non-white patients generally had a higher risk of progression to all forms of diabetic retinopathy, while Asian patients were less likely to develop diabetic macular edema (HR 0.76, 95% CI 0.66-0.87). CONCLUSIONS: We confirm the critical importance of glucose control as measured by HbA1c on the risk of development of diabetic retinopathy.

The Association of Chinese Ethnicity and Language Preference with Advance Directive Completion Among Older Patients in an Integrated Health System.

BACKGROUND: Little is known about possible differences in advance directive completion (ADC) based on ethnicity and language preference among Chinese Americans on a regional level. OBJECTIVE: To understand the association of ethnicity and language preference with ADC among Chinese Americans. DESIGN: Retrospective cohort analysis with direct standardization. PARTICIPANTS: A total of 31,498 Chinese and 502,991 non-Hispanic White members enrolled in Kaiser Permanente Northern California during the entire study period between 2013 and 2017 who were 55 or older as of January 1, 2018. MAIN MEASURES: We compared the proportion of ADC among non-Hispanic White and Chinese patients, and also analyzed the rates according to language preference within the Chinese population. We calculated ADC rates with direct standardization using covariates previously found in literature to be significant predictors of ADC such as age and utilization. KEY RESULTS: Among Chinese members, 60% preferred English, 16% preferred another language without needing an interpreter, and 23% needed an interpreter. After standardizing for age and utilization, non-Hispanic Whites were more than twice as likely to have ADC as Chinese members (20.6% (95% confidence interval (CI): 20.5-20.7%) vs. 10.0% (95% CI: 9.6-10.3%), respectively). Among Chinese members, there was an inverse association between preference for a language other than English and ADC (13.3% (95% CI: 12.8-13.8%) if preferring English, 6.1% (95% CI: 5.4-6.7%) if preferring non-English language but not needing an interpreter, and 5.1% (95% CI: 4.6-5.6%) if preferring non-English language and needing an interpreter). CONCLUSIONS: Chinese members are less likely to have ADC relative to non-Hispanic White members, and those preferring a language other than English are most affected. Further studies can assess reasons for lower ADC among Chinese members, differences in other Asian American populations, and interventions to reduce differences among Chinese members especially among those preferring a language other than English.

Advance Directive Completion Among Black Adults Within a Community-Based Integrated Health Care System.

ObjectivesThe study was conducted to estimate the prevalence of advance directive (AD) completion among Black adults vs non-Hispanic White adults within Kaiser Permanente Northern California integrated health system that includes access to outpatient advance care planning (ACP) specialists and to identify medical services utilization patterns and societal factors that could influence ACP engagement. DesignThe study was carried out through retrospective analysis of electronic health record data of active Kaiser Permanente Northern California members from January 1, 2013 to December 31, 2017, who were age 55 and older, and represented 572,466 active members, of which 11.7% were Black adults. The primary objective was AD completion comparing Black adults to non-Hispanic White adults. Demographic data included age, sex, comorbidities (Charlson comorbidity score ≥ 3) and medical services utilization (inpatient, outpatient, and emergency department [ED] use). Sociodemographic data derived from census data that include census block demographics and head of household educational attainment were utilized. ResultsBlack adults were younger, but had a higher burden of comorbidities (Charlson comorbidity score ≥ 3, 25.3% vs 19.3%) and were more likely to have multiple ED visits (6.7% vs 3.3%) compared to non-Hispanic White adults. The crude AD completion rate was lower among Black adults (10.0% vs 20.3%), and after adjusting for age and health system service area, the difference remained largely unchanged (11.7% vs 20.3%) compared to non-Hispanic White adults. ConclusionsAmong Kaiser Permanente Northern California members with access to outpatient ACP specialists, Black adults were only half as likely to complete an AD. This disparity was only slightly attenuated when standardized for age and health system service area. In addition, Black adults were also less likely to use outpatient services and more likely to use ED services.

Food Insecurity Screening in a Pediatric Clinic: Findings on Documentation and Multidimensional Care Needs.

Background The American Academy of Pediatrics recommends routine screening and resource provision for food insecurity (FI). The authors describe documentation of FI, as well as social, developmental, and health care utilization characteristics, among patients with and without FI in a pediatric clinic. Methods This data-only case-control study describes patients with and without FI seen by pediatricians who identified at least 10 patients with FI between January 2019 and January 2020. Controls were seen by the same pediatrician on the same day, matched 2:1 with cases. Chart review identified FI documentation. Bivariate analyses were used to describe demographic, clinical, and health care utilization characteristics. Results The authors identified 74 patients with FI (cases) and 144 controls. Pediatricians documented FI in the medical record for 76% of patients identified with FI, although only 43% had FI in the problem list. There were no differences between cases and controls in the distribution of age or sex. A larger proportion of cases used the emergency department (25.7% vs 9.6%; p < 0.05), were Black (36.5% vs 15.3%; p < 0.05) or Hispanic (44.6% vs 18.8%; p < 0.05), utilized Medi-Cal (41.9% vs 16.7%; p < 0.05), had developmental delay (13.5% vs 4.2%; p < 0.05), received speech therapy (20.3% vs 4.9%; p < 0.05), or were referred to a social worker (37.8% vs 3.5%; p < 0.05). Discussion Our findings show inconsistent documentation of FI in medical records and that FI is associated with increased social, developmental, and health care utilization needs. Conclusion Further research is needed to assess the impact of documenting FI in patient charts.

Future of Teleoncology: Trends and Disparities in Telehealth and Secure Message Utilization in the COVID-19 Era.

PURPOSE: The COVID-19 pandemic created an imperative to re-examine the role of telehealth in oncology. We studied trends and disparities in utilization of telehealth (video and telephone visits) and secure messaging (SM; ie, e-mail via portal/app), before and during the pandemic. METHODS: Retrospective cohort study of hematology/oncology patient visits (telephone/video/office) and SM between January 1, 2019, and September 30, 2020, at Kaiser Permanente Northern California. RESULTS: Among 334,666 visits and 1,161,239 SM, monthly average office visits decreased from 10,562 prepandemic to 1,769 during pandemic, telephone visits increased from 5,114 to 8,663, and video visits increased from 40 to 4,666. Monthly average SM increased from 50,788 to 64,315 since the pandemic began. Video visits were a significantly higher fraction of all visits (P < .01) in (1) younger patients (Generation Z 48%, Millennials 46%; Generation X 40%; Baby Boomers 34.4%; Silent Generation 24.5%); (2) patients with commercial insurance (39%) compared with Medicaid (32.7%) or Medicare (28.1%); (3) English speakers (33.7%) compared with those requiring an interpreter (24.5%); (4) patients who are Asian (35%) and non-Hispanic White (33.7%) compared with Black (30.1%) and Hispanic White (27.5%); (5) married/domestic partner patients (35%) compared with single/divorced/widowed (29.9%); (6) Charlson comorbidity index ≤ 3 (36.2%) compared with > 3 (31.3%); and (7) males (34.6%) compared with females (32.3%). Similar statistically significant SM utilization patterns were also seen. CONCLUSION: In the pandemic era, hematology/oncology telehealth and SM use rapidly increased in a manner that is feasible and sustained. Possible disparities existed in video visit and SM use by age, insurance plan, language, race, ethnicity, marital status, comorbidities, and sex.

Complications of Acute Posterior Vitreous Detachment.

PURPOSE: To evaluate the risk factors for retinal tear (RT) or rhegmatogenous retinal detachment (RRD) associated with acute, symptomatic posterior vitreous detachment (PVD) in a large comprehensive eye care setting. DESIGN: Retrospective cohort study. PARTICIPANTS: A total of 8305 adult patients in the Kaiser Permanente Northern California Healthcare System (KPNC) during calendar year 2018 who met inclusion criteria. METHODS: The KPNC electronic medical record was queried to capture acute, symptomatic PVD events. Each chart was reviewed to confirm diagnoses and capture specific data elements from the patient history and ophthalmic examination. MAIN OUTCOME MEASURES: Presence of RT or RRD at initial presentation or within 1 year thereafter. RESULTS: Of 8305 patients who presented with acute PVD symptoms, 448 (5.4%) were diagnosed with RT and 335 (4.0%) were diagnosed with RRD. When considering variables available before examination, blurred vision (odds ratio [OR], 2.7; confidence interval [CI], 2.2-3.3), male sex (OR, 2.1; CI, 1.8-2.5), age < 60 years (OR, 1.8; CI, 1.5-2.1), prior keratorefractive surgery (OR, 1.6; CI, 1.3-2.0), and prior cataract surgery (OR, 1.4; CI, 1.2-1.8) were associated with higher risk of RT or RRD, whereas symptoms of flashes were mildly protective (OR, 0.8; CI, 0.7-0.9). Examination variables associated with a high risk of RT or RRD included vitreous pigment (OR, 57.0; CI, 39.7-81.7), vitreous hemorrhage (OR, 5.9; CI, 4.6-7.5), lattice degeneration (OR, 6.0; CI, 4.7-7.7), and visual acuity worse than 20/40 (OR, 3.0; CI, 2.5-3.7). Late RTs or RRDs occurred in 12.4% of patients who had vitreous hemorrhage, lattice degeneration, or a history of RT or RRD in the fellow eye at initial presentation but only 0.7% of patients without any of these 3 risk factors. Refractive error had an approximately linear relationship with age at presentation of PVD, with myopic patients presenting at a younger age (r = 0.4). CONCLUSIONS: This study, based in a comprehensive eye care setting, found the rate of RT and RRD associated with acute PVD to be lower than rates previously reported by retina subspecialty practices. Several patient features strongly predicted the presence of initial and late complications of acute PVD.

Heterogeneity in colorectal cancer incidence among people recommended 3-yearly surveillance post-polypectomy: a validation study.

BACKGROUND: Colonoscopy surveillance is recommended for patients at increased risk of colorectal cancer (CRC) following adenoma removal. Low-, intermediate-, and high-risk groups are defined by baseline adenoma characteristics. We previously examined intermediate-risk patients from hospital data and identified a higher-risk subgroup who benefited from surveillance and a lower-risk subgroup who may not require surveillance. This study explored whether these findings apply in individuals undergoing CRC screening. METHODS: This retrospective study used data from the UK Flexible Sigmoidoscopy Screening Trial (UKFSST), English CRC screening pilot (ECP), and US Kaiser Permanente CRC prevention program (KPCP). Screening participants (50 - 74 years) classified as intermediate-risk at baseline colonoscopy were included. CRC data were available through 2006 (KPCP) or 2014 (UKFSST, ECP). Lower- and higher-risk subgroups were defined using our previously identified baseline risk factors: higher-risk participants had incomplete colonoscopies, poor bowel preparation, adenomas ≥ 20 mm or with high-grade dysplasia, or proximal polyps. We compared CRC incidence in these subgroups and in the presence vs. absence of surveillance using Cox regression. RESULTS: Of 2291 intermediate-risk participants, 45 % were classified as higher risk. Median follow-up was 11.8 years. CRC incidence was higher in the higher-risk than lower-risk subgroup (hazard ratio [HR] 2.08, 95 % confidence interval [CI] 1.07 - 4.06). Surveillance reduced CRC incidence in higher-risk participants (HR 0.35, 95 %CI 0.14 - 0.86) but not statistically significantly so in lower-risk participants (HR 0.41, 95 %CI 0.12 - 1.38). CONCLUSION: As previously demonstrated for hospital patients, screening participants classified as intermediate risk comprised two risk subgroups. Surveillance clearly benefited the higher-risk subgroup.

Underdiagnosis of Isolated Systolic and Isolated Diastolic Hypertension.

Systolic and diastolic hypertension independently predict the risk of adverse cardiovascular events. It remains unclear how systolic pressure, diastolic pressure, and other patient characteristics influence the initial diagnosis of hypertension. Here, we use a cohort of 146,816 adults in a large healthcare system to examine how elevated systolic and/or diastolic blood pressure measurements influence initial diagnosis of hypertension and how other patient characteristics influence the diagnosis. Thirty-four percent of the cohort were diagnosed with hypertension within 1 year. In multivariable logistic regression of the diagnosis of hypertension, controlling for covariates, isolated systolic hypertensive measures (odds ratio [OR] 0.42 [95% confidence interval {CI} 0.41 to 0.43]) and isolated diastolic hypertensive measures (OR 0.32 [95% CI 0.31 to 0.33]) were less likely to lead to hypertension diagnosis when compared with combined hypertensive measures. Higher levels of systolic blood pressure had a greater impact on hypertension diagnosis (OR 1.77 [95% CI 1.75 to 1.79] per Z-score) than did higher levels of diastolic blood pressure (OR 1.34 [95% CI 1.32 to 1.36] per Z-score). Older age, non-white race/ethnicity, and medical comorbidities all predicted the establishment of a diagnosis of hypertension. Isolated systolic and isolated diastolic hypertension are underdiagnosed in clinical practice, and several patient-centered factors also strongly influence whether a diagnosis is made. In conclusion, our findings uncover a care gap that can be closed with increased attention to the independent influence of systolic and diastolic hypertension and the various patient-centered factors that may impact hypertension diagnosis.

The Association of Spanish Language Preference with Advance Directive Completion.

BACKGROUND/OBJECTIVES: Hispanics have lower advance directive (AD) completion than non-Hispanic Whites. Few studies have assessed the role of language preference in end-of-life planning. We investigated whether language preference and needing an interpreter affected AD completion among older adults in an integrated health system. DESIGN: Retrospective cohort investigation of electronic medical records. SETTING: Northern California integrated health system. PARTICIPANTS: A total of 620,948 Hispanic and non-Hispanic White patients, aged 55 years and older, between January 1, 2013, and December 31, 2017. MEASUREMENTS: Descriptive statistics and bivariate analysis were performed to compare AD completion among non-Hispanic Whites, Hispanics, and Hispanic subgroups by language preference (English speaking, Spanish speaking, and needed interpreter). We conducted multivariable logistic regression to determine the relationship between language preference and having an AD while controlling for demographic, clinical, and utilization factors. RESULTS: We found 20.3% of non-Hispanic Whites (n = 512,577) and 10.9% of Hispanics (n = 108,371) had completed an AD. Among Hispanics, after controlling for demographic, clinical, and utilization factors, compared with Spanish speakers requiring an interpreter, English speakers had nearly two-fold increased odds of completing an AD (adjusted odds ratio (aOR) = 2.6; 95% confidence interval (CI) = 2.4-2.9), whereas Spanish speakers not requiring an interpreter had 20% increased odds (aOR = 1.2; 95% CI = 1.1-1.3). Additional predictors of successful AD completion were being female, being older, having more comorbidities, having more hospital and emergency department visits, and having higher socioeconomic status. There were no differences associated with primary care provider characteristics. CONCLUSION: These findings indicate the need for a tailored outreach to Hispanics, particularly among those subgroups who require the need of an interpreter, to reduce AD completion disparities.

Diabetic retinopathy screening using a virtual reading center.

AIMS: To summarize the effects of centralization of diabetic fundus photograph interpretation into a virtual reading center. METHODS: In 2016 Kaiser Permanente Northern California, a large, membership-based health plan with an ethnically and racially diverse population, centralized diabetic retinopathy screening into a virtual reading center. Retina screens were based on single field, 45-degree fundus photographs. We compared the accuracy of photography interpretation the year before centralization to the year after using masked reads performed by retina specialists of 1000 randomly selected screens from each time period. RESULTS: In all, 1902 patient screens with adequate quality images were included in the primary analysis. Images from pre-centralization screens were largely read by ophthalmologists (76.2%), while screens post-centralization were mainly read by optometrists (84.6%). Despite being interpreted by readers with lower levels of professional training, the sensitivity of screening increased from 43.9% (95% CI 38.0-49.8%) to 66.0% (95% CI 60.5-71.4%). CONCLUSION: A move to a centralized virtual reading center was associated with improved accuracy of diabetic retinopathy screening.

Effect of Systolic and Diastolic Blood Pressure on Cardiovascular Outcomes.

BACKGROUND: The relationship between outpatient systolic and diastolic blood pressure and cardiovascular outcomes remains unclear and has been complicated by recently revised guidelines with two different thresholds (≥140/90 mm Hg and ≥130/80 mm Hg) for treating hypertension. METHODS: Using data from 1.3 million adults in a general outpatient population, we performed a multivariable Cox survival analysis to determine the effect of the burden of systolic and diastolic hypertension on a composite outcome of myocardial infarction, ischemic stroke, or hemorrhagic stroke over a period of 8 years. The analysis controlled for demographic characteristics and coexisting conditions. RESULTS: The burdens of systolic and diastolic hypertension each independently predicted adverse outcomes. In survival models, a continuous burden of systolic hypertension (≥140 mm Hg; hazard ratio per unit increase in z score, 1.18; 95% confidence interval [CI], 1.17 to 1.18) and diastolic hypertension (≥90 mm Hg; hazard ratio per unit increase in z score, 1.06; 95% CI, 1.06 to 1.07) independently predicted the composite outcome. Similar results were observed with the lower threshold of hypertension (≥130/80 mm Hg) and with systolic and diastolic blood pressures used as predictors without hypertension thresholds. A J-curve relation between diastolic blood pressure and outcomes was seen that was explained at least in part by age and other covariates and by a higher effect of systolic hypertension among persons in the lowest quartile of diastolic blood pressure. CONCLUSIONS: Although systolic blood-pressure elevation had a greater effect on outcomes, both systolic and diastolic hypertension independently influenced the risk of adverse cardiovascular events, regardless of the definition of hypertension (≥140/90 mm Hg or ≥130/80 mm Hg). (Funded by the Kaiser Permanente Northern California Community Benefit Program.).

Association between Inpatient Delirium and Hospital Readmission in Patients ≥ 65 Years of Age: A Retrospective Cohort Study.

BACKGROUND: Delirium affects more than seven million hospitalized adults in the United States annually. However, its impact on postdischarge healthcare utilization remains unclear. OBJECTIVE: To determine the association between delirium and 30-day hospital readmission. DESIGN: A retrospective cohort study. SETTING: A general community medical and surgical hospital. PATIENTS: All adults who were at least 65 years old, without a history of delirium or alcohol-related delirium, and were hospitalized from September 2010 to March 2015. MEASUREMENTS: The patients deemed at risk for or displaying symptoms of delirium were screened by nurses using the Confusion Assessment Method with a followup by a staff psychiatrist for a subset of screen-positive patients. Patients with delirium confirmed by a staff psychiatrist were compared with those without delirium. The primary outcome was the 30-day readmission rate. The secondary outcomes included emergency department (ED) visits 30 days postdischarge, mortality during hospitalization and 30 days postdischarge, and discharge location. RESULTS: The cohort included 718 delirious patients and 7,927 nondelirious patients. Using an unweighted multivariable logistic regression, delirium was determined to be significantly associated with the increased odds of readmission within 30 days of discharge (odds ratio (OR): 2.60; 95% CI, 1.96-3.44; P < .0001). Delirium was also significantly (P < .0001) associated with ED visits within 30 days postdischarge (OR: 2.18; 95% CI: 1.77-2.69) and discharge to a facility (OR: 2.52; 95% CI: 2.09-3.01). CONCLUSIONS: Delirium is a significant predictor of hospital readmission, ED visits, and discharge to a location other than home. Delirious patients should be targeted to reduce postdischarge healthcare utilization.

Reducing risk of Clostridium difficile infection and overall use of antibiotic in the outpatient treatment of urinary tract infection.

BACKGROUND: Risk of community-acquired Clostridium difficile infection (CA-CDI) following antibiotic treatment specifically for urinary tract infection (UTI) has not been evaluated. METHODS: We conducted a nested case-control study at Kaiser Permanente Northern California, 2007-2010, to assess antibiotic prescribing and other factors in relation to risk of CA-CDI in outpatients with uncomplicated UTI. Cases were diagnosed with CA-CDI within 90 days of antibiotic use. We used matched controls and confirmed case-control eligibility through chart review. Antibiotics were classified as ciprofloxacin (most common), or low risk (nitrofurantoin, sulfamethoxazole/trimethoprim), moderate risk, or high risk (e.g. cefpodoxime, ceftriaxone, clindamycin) for CDI. We computed the adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for the relationship of antibiotic treatment for uncomplicated UTI and history of relevant gastrointestinal comorbidity (including gastrointestinal diagnoses, procedures, and gastric acid suppression treatment) with risk of CA-CDI using logistic regression analysis. RESULTS: Despite the large population, only 68 cases were confirmed with CA-CDI for comparison with 112 controls. Female sex [81% of controls, adjusted odds ratio (OR) 6.3, CI 1.7-24), past gastrointestinal comorbidity (prevalence 39%, OR 2.3, CI 1.1-4.8), and nongastrointestinal comorbidity (prevalence 6%, OR 2.8, CI 1.4-5.6) were associated with increased CA-CDI risk. Compared with low-risk antibiotic, the adjusted ORs for antibiotic groups were as follows: ciprofloxacin, 2.7 (CI 1.0-7.2); moderate-risk antibiotics, 3.6 (CI 1.2-11); and high-risk antibiotics, 11.2 (CI 2.4-52). CONCLUSIONS: Lower-risk antibiotics should be used for UTI whenever possible, particularly in patients with a gastrointestinal comorbidity. However, UTI can be managed through alternative approaches. Research into the primary prevention of UTI is urgently needed.

Type 2 Diabetes Mellitus and Glioblastoma Multiforme-Assessing Risk and Survival: Results of a Large Retrospective Study and Systematic Review of the Literature.

OBJECTIVE: Despite studies showing a positive correlation between type 2 diabetes mellitus (DM2), a modifiable risk factor, and various cancer types, the link remains controversial in the setting of glioblastoma multiforme (GBM). In this study, we assessed whether DM2 and DM2-associated factors were associated with a higher risk of developing GBM and also determined if DM2 affected the survival of patients with GBM. METHODS: A cross-sectional case-control study of 1144 GBM cases diagnosed between 2000 and 2013 of which 969 patients matched for age and sex was performed to assess the association between DM2, hyperlipidemia, and obesity with the incidence of GBM. A longitudinal study of the patients with GBM was also performed to assess the association between the effect of DM2 and GBM survival. RESULTS: No association was seen between DM2, hyperlipidemia, obesity, and GBM. DM2 was associated with poorer survival in univariate testing yet not in multivariate testing. Diabetic patients with GBM had good glycemic control. Older patients had poorer survival and overall survival improved over years of study. CONCLUSIONS: DM2, hyperlipidemia, and obesity were not associated with increased risk of developing GBM, and DM2 itself does not seem to influence survival among these patients. This finding might be related to good glycemic control in this cohort. Survey of the literature consistently shows that hyperglycemia is associated with poorer survival. Our findings suggest that rather than the presence or absence of DM2, glycemic control seems to be more important in the survival of patients with GBM, which warrants future investigation.

Statin Adherence Is Associated With Reduced Recurrent Stroke Risk in Patients With or Without Atrial Fibrillation.

BACKGROUND AND PURPOSE: Outpatient statin use reduces the risk of recurrent ischemic stroke among patients with stroke of atherothrombotic cause. It is not known whether statins have similar effects in ischemic stroke caused by atrial fibrillation (AFib). METHODS: We studied outpatient statin adherence, measured by percentage of days covered, and the risk of recurrent ischemic stroke in patients with or without AFib in a 21-hospital integrated healthcare delivery system. RESULTS: Among 6116 patients with ischemic stroke discharged on a statin over a 5-year period, 1446 (23.6%) had a diagnosis of AFib at discharge. The mean statin adherence rate (percentage of days covered) was 85, and higher levels of percentage of days covered correlated with greater degrees of low-density lipoprotein suppression. In multivariable survival models of recurrent ischemic stroke over 3 years, after controlling for age, sex, race/ethnicity, medical comorbidities, and hospital center, higher statin adherence predicted reduced stroke risk both in patients without AFib (hazard ratio, 0.78; 95% confidence interval, 0.63-0.97) and in patients with AFib (hazard ratio, 0.59; 95% confidence interval, 0.43-0.81). This association was robust to adjustment for the time in the therapeutic range for international normalized ratio among AFib subjects taking warfarin (hazard ratio, 0.61; 95% confidence interval, 0.41-0.89). CONCLUSIONS: The relationship between statin adherence and reduced recurrent stroke risk is as strong among patients with AFib as it is among patients without AFib, suggesting that AFib status should not be a reason to exclude patients from secondary stroke prevention with a statin.

Does Surgical Teaching Take Time? Resident Participation in Minimally Invasive Hysterectomy for Endometrial Cancer.

STUDY OBJECTIVE: To determine the association between resident involvement and operative time for minimally invasive surgery (MIS) for endometrial cancer. DESIGN: A retrospective cohort study (Canadian Task Force classification II-2). SETTING: An integrated health care system in Northern California. PATIENTS: A total of 1433 women who underwent MIS for endometrial cancer and endometrial intraepithelial neoplasia from January 2009 to January 2014. INTERVENTIONS: Resident participation in 430 of 688 laparoscopic cases (62%) and 341 of 745 robotic cases (46%). MEASUREMENTS AND MAIN RESULTS: The primary outcome was the impact of resident involvement on surgical time. When residents were involved in laparoscopic and robotic surgery, there was an increase of 61 minutes (median operative time, 186 vs 125 minutes; p < .001) and 31 minutes (median operative time, 165 vs 134 minutes; p < .001), respectively. Resident participation was associated with increased operative times in all levels of surgical complexity from hysterectomy alone to hysterectomy with pelvic and para-aortic lymph node dissection. Resident participation was also associated with increased major intraoperative complications (3.4% vs 1.8%, p = .02) as well as major postoperative complications (6.4% vs 3.8%, p = .003). CONCLUSION: The presence of a resident was associated with a 32% increase in operative time for minimally invasive cases in gynecologic oncology for endometrial cancer. Because of the retrospective nature, we cannot infer causality of operative outcomes because residents were also involved in more high-risk patients and complex cases. For health care systems using surgical metrics, there may be a need to allocate more time for resident involvement.

Classification of Postoperative Complications in Robotic-assisted Compared With Laparoscopic Hysterectomy for Endometrial Cancer.

STUDY OBJECTIVE: To compare intraoperative and postoperative surgical complications and outcomes between robotic-assisted and laparoscopic surgical management of endometrial cancer using a standardized classification system. DESIGN: A retrospective cohort study (Canadian Task Force classification II-2). SETTING: An integrated health care system in Northern California. PATIENTS: One thousand four hundred thirty-three women with a diagnosis of complex atypical hyperplasia and endometrial cancer managed by minimally invasive hysterectomy and surgical staging from January 2009 to January 2014. INTERVENTIONS: Seven hundred forty-five robotic-assisted and 688 laparoscopic hysterectomies were evaluated. MEASUREMENTS AND MAIN RESULTS: The primary outcome was intraoperative and postoperative complications within 30 days. All complications were categorized using the Clavien-Dindo classification system. Secondary outcomes included total operative time, estimated blood loss, transfusion rates, length of stay, conversion to laparotomy, and number of pelvic and para-aortic lymph nodes retrieved. The modality of hysterectomy was not associated with either overall intraoperative complications or major postoperative complications (p > .1). However, there were significantly fewer minor postoperative complications with robotic surgery (16.6% vs 25.6%, p < .01). Statistically significant differences were also noted in the following outcomes: decreased median operative time, length of stay, estimated blood loss, conversion to laparotomy, and median number of lymph nodes retrieved in the robotic group when compared with the laparoscopic group. CONCLUSION: There was no difference in the rate of major complication between robotic and laparoscopic surgery using the Clavien-Dindo system of categorizing surgical complications; however, there were clinically significant differences favoring the robotic approach, including a lower rate of minor complications and conversion rate to laparotomy.