Modular design, application architecture, and usage of a self-service model for enterprise data delivery: the Duke Enterprise Data Unified Content Explorer (DEDUCE)

Continuation of Oral Antidiabetic Medications Was Associated With Better Early Postoperative Blood Glucose Control Compared to Sliding Scale Insulin After Total Knee Arthroplasty

BACKGROUND

This study evaluated blood glucose (BG), creatinine levels, metabolic issues, length of stay (LOS), and early postoperative complications in diabetic primary total knee arthroplasty (TKA) patients. It examined those who continued home oral antidiabetic medications and those who switched to insulin postoperatively. The hypothesis was that continuing home medications would lead to lower BG levels without metabolic abnormalities.

METHODS

Patients who had diabetes who underwent primary TKA from 2013 to 2022 were evaluated retrospectively. Diabetic patients who were not on home oral antidiabetic medications or who were not managed as an inpatient postoperatively were excluded. Patient demographics and laboratory tests collected preoperatively and postoperatively as well as 90-day emergency department visits and 90-day readmissions, were pulled from electronic records. Patients were grouped based on inpatient diabetes management: continuation of home medications versus new insulin coverage. Acute postoperative BG control, creatinine levels, metabolic abnormalities, LOS, and early postoperative complications were compared between groups. Multivariable regression analyses were performed to measure associations.

RESULTS

A total of 867 primary TKAs were assessed; 703 (81.1%) patients continued their home oral antidiabetic medications. Continuing home antidiabetic medications demonstrated lower median maximum inpatient BG (180.0 mg/dL versus 250.0 mg/dL; P < .001) and median average inpatient BG (136.7 mg/dL versus 173.7 mg/dL; P < .001). Logistic regression analyses supported the presence of an association (odds ratio = 17.88 [8.66, 43.43]; P < .001). Proportions of acute kidney injury (13.5 versus 26.7%; P < .001) were also lower. There was no difference in relative proportions of metabolic acidosis (4.4 versus 3.7%; P = .831), LOS (2.0 versus 2.0 days; P = .259), or early postoperative complications.

CONCLUSIONS

Continuing home oral antidiabetic medications after primary TKA was associated with lower BG levels without an associated worsening creatinine or increase in metabolic acidosis.

LEVEL III EVIDENCE

Retrospective Cohort Study.

Continuing Home Oral Hypoglycemic Medications Was Associated With Superior Postoperative Glycemic Control Versus Initiating Sliding Scale Insulin After Total Hip Arthroplasty

BACKGROUND

This retrospective study investigated the impact of continuing vs discontinuing home oral hypoglycemic medications for patients with diabetes undergoing total hip arthroplasty.

MATERIALS AND METHODS

Patients who were not exclusively receiving home oral hypoglycemic regimens were excluded. Additionally, patients whose diabetes was not managed inpatient postoperatively were excluded. Included patients were retrospectively evaluated for early postoperative glycemic control, renal function, and metabolic abnormalities. Patients were then compared based on whether their home oral hypoglycemic regimen was continued vs discontinued in favor of initiating insulin while inpatient and analyzed using multivariable regression analysis.

RESULTS

A total of 532 patients undergoing total hip arthroplasty met inclusion criteria, with 78.6% continuing their home oral hypoglycemic regimen. Those who continued showed significantly lower median maximum inpatient blood glucose (178.5 mg/dL vs 249.5 mg/dL; P<.001) and median average inpatient blood glucose (138.4 mg/dL vs 178.6 mg/dL; P<.001). Linear regression analysis, adjusting for various potential confounding factors, revealed a positive correlation between discontinuation of home hypoglycemic medications and higher maximum in-patient blood glucose (β=70.15 [95% CI, 59.27-81.03]; P<.001). Patients in the continuation group had lower proportions of acute kidney injury (18.7% vs 41.2%; P<.001) and metabolic acidosis (4.3% vs 17.5%; P<.001), along with a shorter length of stay (1.0 d vs 2.0 d; P<.001).

CONCLUSION

These findings suggest that continuing a home oral hypoglycemic regimen for patients with diabetes after total hip arthroplasty is associated with superior glycemic control without exacerbating renal abnormalities or increasing metabolic complications. [Orthopedics. 202x;4x(x):xx-xx.].

Clinical outcomes in patients with piperacillin/tazobactam-non-susceptible but ceftriaxone-susceptible E. coli or K. pneumoniae bloodstream infection

BACKGROUND

A small proportion of Escherichia coli and Klebsiella pneumoniae demonstrate in vitro non-susceptibility to piperacillin/tazobactam but retain susceptibility to ceftriaxone. Uncertainty remains regarding how best to treat these isolates.

OBJECTIVES

We sought to compare clinical outcomes between patients with piperacillin/tazobactam-non-susceptible but ceftriaxone-susceptible E. coli or K. pneumoniae bloodstream infection receiving definitive therapy with ceftriaxone versus an alternative effective antibiotic.

METHODS

We retrospectively identified patients with a positive blood culture for piperacillin/tazobactam-non-susceptible but ceftriaxone-susceptible E. coli or K. pneumoniae between 1 January 2013 and 31 December 2022. Patients were divided into one of two definitive treatment groups: ceftriaxone or alternative effective antibiotic. Our primary outcome was a composite of 90 day all-cause mortality, hospital readmission, or recurrence of infection. We used Cox proportional hazards models to compare time with the composite outcome between groups.

RESULTS

Sixty-two patients were included in our analysis. Overall, median age was 63 years (IQR 49.5-71.0), the most common source of infection was intra-abdominal (25/62; 40.3%) and the median total duration of therapy was 12.0 days (IQR 9.0-16.8). A total of 9/22 (40.9%) patients in the ceftriaxone treatment group and 18/40 (45.0%) patients in the alternative effective antibiotic group met the composite endpoint. In an adjusted time-to-event analysis, there was no difference in the composite endpoint between groups (HR 0.67, 95% CI 0.30-1.50). The adjusted Bayesian posterior probability that the HR was less than or equal to 1 (i.e. ceftriaxone is as good or better than alternative therapy) was 85%.

CONCLUSIONS

These findings suggest that ceftriaxone can be used to effectively treat bloodstream infections with E. coli or K. pneumoniae that are non-susceptible to piperacillin/tazobactam but susceptible to ceftriaxone.

Tobacco exposures are associated with health care utilization (HCU) and health care costs in pregnant persons and their newborn babies

INTRODUCTION

Identifying health care utilization and costs associated with active and passive smoking during pregnancy could help improve health management strategies.

METHODS

Data are from the Newborn Epigenetics STudy (NEST), a birth cohort enrolled from 2005-2011 in Durham and adjacent counties in North Carolina, United States. Participants included those for whom prenatal serum samples were assayed and for whom administrative data were obtainable (N=1,045). Zero-inflated Poisson (ZIP) regression models were used to assess associations between cotinine, adjusted for covariates (e.g., race and ethnicity, age at delivery, cohabitation status, education), and health care utilization outcomes. Generalized linear regression models were used to estimate average total charges. Simulation models were conducted to determine the economic benefits of reducing SHS and smoking during pregnancy.

RESULTS

Increasing levels of cotinine were positively associated with parent's number of ED visits (coefficient(b)=0.0012, standard error (SE)=0.0002; P<.001), the number of ICU hours (b=0.0079, SE=0.0025; P=.002)), time spent in the ICU (b=0.0238, SE=0.0020, P<.001), and the number of OP visits (b=0.0003, SE=0.0001; P<.001). For infants, higher cotinine levels were associated with higher number of ED (b=0.0012, SE=0.0004; P=.005), ICU (b=0.0050, SE=0.001; P<.001), and OP (b=0.0006, SE=0.0002; P<.001) visits and longer time spent in the ED (b=0.0025, SE=0.0003; P<.001), ICU (b=0.0005, SE=0.0001; P<.001), and IP (b=0.0020, SE=0.0002; P<.001). Simulation results showed that a 5% reduction in smoking would correspond to a potential median cost savings of $150,533 from ED visits of parents and infants.

CONCLUSION

Our findings highlight the importance of smoke exposure cessation during pregnancy to reduce health care utilization and costs for both parents and infants.

IMPLICATIONS

This study reinforces the importance of reducing smoking and secondhand smoke exposure during pregnancy. Focusing on expanding cessation services to this group could help reduce morbidities observed within this population. Furthermore, there is the potential for health care costs savings to health care systems, especially to those with high delivery numbers. These cost savings are represented by potential reductions in ED, OP, and ICU hours and visits.

Implications of trial eligibility in patients with heart failure with mildly reduced or preserved ejection fraction

AIMS

Clinical trials in heart failure with mildly reduced or preserved ejection fraction (HFmrEF/HFpEF) commonly have detailed eligibility criteria. This may contribute to challenges with efficient enrolment and questions regarding the generalizability of trial findings.

METHODS AND RESULTS

Patients with HFmrEF/HFpEF from a large US healthcare system were identified through a computable phenotype applied in linked imaging and electronic health record databases. We evaluated shared eligibility criteria from five recent/ongoing HFmrEF/HFpEF trials (PARAGON-HF, EMPEROR-Preserved, DELIVER, FINE-ARTS, and SPIRRIT-HFpEF) and compared clinical and echocardiographic features as well as outcomes between trial-eligible and trial-ineligible patients. Among 5552 patients with HFpEF/HFmrEF, 792 (14%) were eligible for trial consideration, having met all criteria assessed. Causes of ineligibility included lack of recent loop diuretics (37%), significant pulmonary disease (24%), reduced estimated glomerular filtration rate (17%), recent stroke/transient ischaemic attack (13%), or low natriuretic peptides (12%); 53% of ineligible patients had >1 reason for exclusion. Compared with eligible patients, ineligible patients were younger (age 71 vs. 75 years, P < 0.001) with higher rates of coronary artery disease (66% vs. 59%, P < 0.001) and peripheral vascular disease (40% vs. 33%, P < 0.001), but less mitral regurgitation, lower E/e' ratio, and smaller left atrial sizes. Both eligible and ineligible patients demonstrated high rates of structural heart disease consistent with HFpEF [elevated left atrial size or left ventricular (LV) hypertrophy/increased LV mass], although this was slightly higher among eligible patients (95% vs. 92%, P = 0.001). The two cohorts demonstrated similar LV global longitudinal strain along with a similar prevalence of atrial fibrillation/flutter, hypertension, and obesity. Ineligible patients had similar all-cause mortality (33% vs. 33% at 3 years) to those eligible but lower rates of heart failure hospitalization (20% vs. 28% at 3 years, P < 0.001).

CONCLUSIONS

Among patients with HFmrEF/HFpEF from a large health system, approximately one in seven were eligible for major trials based on key criteria applied through a clinical computable phenotype. These findings highlight the large proportion of patients with HFmrEF/HFpEF ineligible for contemporary trials for whom the generalizability of trial findings may be questioned and further investigation would be beneficial.

Population Pharmacokinetics of Caffeine in Neonates with Congenital Heart Disease and Associations with Acute Kidney Injury

Cardiac surgery-associated acute kidney injury (CS-AKI) occurs in approximately 65% of neonates undergoing cardiac surgery on cardiopulmonary bypass and contributes to morbidity and mortality. Caffeine may reduce CS-AKI by counteracting adenosine receptor upregulation after bypass, but pharmacokinetics (PK) in this population are unknown. The goal of our analysis is to address knowledge gaps in age-, disease-, and bypass-related effects on caffeine disposition and explore preliminary associations between caffeine exposure and CS-AKI using population PK modeling techniques and an opportunistic, electronic health record-integrated trial design. We prospectively enrolled neonates receiving preoperative caffeine per standard of care and collected PK samples. We retrospectively identified neonates without caffeine exposure undergoing surgery on bypass as a control cohort. We followed US Food and Drug Administration guidance for population PK model development using NONMEM. Effects of clinical covariates on PK parameters were evaluated. We simulated perioperative exposures and used multivariable logistic regression to evaluate the association between caffeine exposure and CS-AKI. Twenty-seven neonates were included in model development. A 1-compartment model with bypass time as a covariate on clearance and volume of distribution best fit the data. Twenty-three neonates with caffeine exposure and 109 controls were included in the exposure-response analysis. Over half of neonates developed CS-AKI. On multivariable analysis, there were no significant differences between CS-AKI with and without caffeine exposure. Neonates with single-ventricle heart disease without CS-AKI had consistently higher simulated caffeine exposures. Our results highlight areas for further study to better understand disease- and bypass-specific effects on drug disposition and identify populations where caffeine may be beneficial.

Concurrent Sexually Transmitted Infection Testing Among Patients Tested for Mpox at a Tertiary Healthcare System

Coinfection with sexually transmitted infections (STIs) and mpox is common. We evaluated concurrent STI testing among Duke Health patients tested for mpox. We found that most patients tested for mpox were not comprehensively tested for STIs, despite concurrent STIs being diagnosed in 15% of patients when testing was performed.

Noninvasive Ventilation or CPAP for Postextubation Support in Small Infants

BACKGROUND

Infants with a high risk of extubation failure are often treated with noninvasive ventilation (NIV) or CPAP, but data on the role of these support modalities following extubation are sparse. This report describes our experience using NIV or CPAP to support infants following extubation in our pediatric ICUs (PICUs).

METHODS

We performed a retrospective study of children < 10 kg receiving postextubation NIV or CPAP in our PICUs. Data on demographics, medical history, type of support, vital signs, pulse oximetry, near-infrared spectroscopy (NIRS), gas exchange, support settings, and re-intubation were extracted from the electronic medical record. Support was classified as prophylactic if planned before extubation and rescue if initiated within 24 h of extubation. We compared successfully extubated and re-intubated subjects using chi-square test for categorical variables and Mann-Whitney test for continuous variables.

RESULTS

We studied 51 subjects, median age 44 (interquartile range 0.5-242) d and weight 3.7 (3-4.9) kg. There were no demographic differences between groups, except those re-intubated were more likely to have had cardiac surgery prior to admission (0% vs 14%, P = .040). NIV was used in 31 (61%) and CPAP in 20 (39%) subjects. Prophylactic support was initiated in 25 subjects (49%), whereas rescue support was needed in 26 subjects (51%). Twenty-two subjects (43%) required re-intubation. Re-intubation rate was higher for rescue support (58% vs 28%, P = .032). Subjects with a pH < 7.35 (4.3% vs 42.0%, P = .003) and lower somatic NIRS (39 [24-56] vs 62 [46-72], P = .02) were more likely to be re-intubated. The inspiratory positive airway pressure, expiratory positive airway pressure, and FIO2 were higher in subjects who required re-intubation.

CONCLUSIONS

NIV or CPAP use was associated with a re-intubation rate of 43% in a heterogeneous sample of high-risk infants. Acidosis, cardiac surgery, higher FIO2 , lower somatic NIRS, higher support settings, and application of rescue support were associated with the need for re-intubation.

Relevance of cleft palatal width in short-term otologic outcomes

OBJECTIVE

Children with cleft palate are at increased risk for Eustachian tube dysfunction (ETD) and conductive hearing loss from chronic otitis media. While it has been proposed that the severity of ETD is related to the severity of cleft palate, data are lacking to support this hypothesis. An improved understanding of the relevance of cleft width may have prognostic value that could inform decisions on the timing of tympanostomy tube placement and choice of tympanostomy tube design. The objective of this study was to assess severity of ETD in children with narrow, moderate, and wide cleft palate, with examination of hearing outcomes, number of tympanostomy procedures, and incidence of otologic complications.

METHODS

Retrospective chart review was conducted on 58 patients with primary palatoplasty performed at a single academic medical center from January 1, 2016-December 31, 2019. The primary outcome was the number of otologic procedures performed after the initial palatoplasty. Secondary outcomes included audiometric findings, number of tympanostomy tube placements, presence of effusion at the time of myringotomy, and occurrence of any postoperative otologic complication. Outcomes were compared for patients with narrow (<10 mm), moderate (10-15 mm), and wide (>15 mm) cleft palate. Analysis included consideration of cleft palatal morphology (Veau I - IV), presence of Robin sequence or syndromes, and risk factors for otitis media.

RESULTS

Patients with moderate and wide cleft palate underwent higher mean numbers of otologic procedures [narrow: 1.3 (95% confidence interval [CI] 0.9, 1.7); moderate: 1.6 (95% CI 1.1, 2.1); wide: 1.8 (95% CI 1.2, 2.4)]. Moderate and wide cleft palate were less likely to have normal hearing after their first tympanostomy (narrow: 50%, 10/20; moderate: 25%, 6/24; wide: 36%, 5/14). Patients with a wide cleft palate had a shorter median time between first and second tympanostomy procedures (median, IQR; narrow: 27.0, 20.8-35.7; moderate 20.4, 16.3-25.9; wide 17.3, 11.5-23.4).

CONCLUSION

Our findings suggest that patients with wider cleft palate may be more susceptible to severe ETD. Further large-scale study may help to allow for more informed and personalized clinical decision making for management of cleft palate, incorporating cleft width for prognosis of risks for persistent middle ear dysfunction.

Association of Neighborhood Deprivation With Healthcare Utilization Among Persons With Human Immunodeficiency Virus: A Latent Class Analysis

Background

We previously identified 3 latent classes of healthcare utilization among people with human immunodeficiency virus (PWH): adherent, nonadherent, and sick. Although membership in the "nonadherent" group was associated with subsequent disengagement from human immunodeficiency virus (HIV) care, socioeconomic predictors of class membership remain unexplored.

Methods

We validated our healthcare utilization-based latent class model of PWH receiving care at Duke University (Durham, North Carolina) using patient-level data from 2015 to 2018. SDI scores were assigned to cohort members based on residential addresses. Associations of patient-level covariates with class membership were estimated using multivariable logistic regression and movement between classes was estimated using latent transition analysis.

Results

A total of 1443 unique patients (median age of 50 years, 28% female sex at birth, 57% Black) were included in the analysis. PWH in the most disadvantaged (highest) SDI decile were more likely to be in the "nonadherent" class than the remainder of the cohort (odds ratio [OR], 1.58 [95% confidence interval {CI}, .95-2.63]) and were significantly more likely to be in the "sick" class (OR, 2.65 [95% CI, 2.13-3.30]). PWH in the highest SDI decile were also more likely to transition into and less likely to transition out of the "sick" class.

Conclusions

PWH who resided in neighborhoods with high levels of social deprivation were more likely to have latent class membership in suboptimal healthcare utilization groupings, and membership persisted over time. Risk stratification models based on healthcare utilization may be useful tools in the early identification of persons at risk for suboptimal HIV care engagement.

Using Pharmacokinetic Modeling and Electronic Health Record Data to Predict Clinical and Safety Outcomes after Methylprednisolone Exposure during Cardiopulmonary Bypass in Neonates

Background

Infants undergoing cardiac surgery with cardiopulmonary bypass (CPB) frequently receive intraoperative methylprednisolone (MP) to suppress CPB-related inflammation; however, the optimal dosing strategy and efficacy of MP remain unclear.

Methods

We retrospectively analyzed all infants under 90 days-old who received intra-operative MP for cardiac surgery with CPB from 2014-2017 at our institution. We combined real-world dosing data from the electronic health record (EHR) and two previously developed population pharmacokinetic/pharmacodynamic models to simulate peak concentration (Cmax) and area under the concentration-time curve for 24 h (AUC24) for MP and the inflammatory cytokines interleukin-6 (IL-6) and interleukin-10 (IL-10). We evaluated the relationships between post-operative, safety, and other clinical outcomes obtained from the EHR with each predicted exposure using non-parametric tests.

Results

A total of 142 infants with median post-natal age 8 (interquartile range [IQR]: 5, 37) days received a total dose of 30 (19, 49) mg/kg of MP. Twelve (8%) died, 37 (26%) met the composite post-operative outcome, 114 (80%) met the composite safety outcome, and 23 (16%) had a major complication. Predicted median Cmax and AUC24 IL-6 exposure was significantly higher for infants meeting the composite post-operative outcome and those with major complications. Predicted median Cmax and AUC24 MP exposure was significantly higher for infants requiring insulin. No exposure was associated with death or other safety outcomes.

Conclusions

Pro-inflammatory IL-6, but not MP exposure, was associated with post-operative organ dysfunction, suggesting current MP dosing may not adequately suppress IL-6 or increase IL-10 to impact clinical outcomes. Prospective study will be required to define the optimal exposure-efficacy and exposure-safety profiles in these infants.

Methicillin-resistant Staphylococcus aureus nasal colonization in children with cerebral palsy

A retrospective cohort of children admitted to the pediatric intensive care unit (PICU) with cerebral palsy was matched 1:3 by age and admission year to determine odds of methicillin-resistant Staphylococcus aureus (MRSA) nasal colonization. Adjusted odds of MRSA nasal colonization at PICU admission were 2.6-fold higher among children with cerebral palsy.

Echocardiographic Features Beyond Ejection Fraction and Associated Outcomes in Patients With Heart Failure With Mildly Reduced or Preserved Ejection Fraction

BACKGROUND

Heart failure (HF) guidelines recommend assessment of left ventricular ejection fraction (LVEF) to classify patients and guide therapy implementation. However, LVEF alone may be insufficient to adequately characterize patients with HF, especially those with mildly reduced or preserved LVEF. Recommendations on additional testing are lacking, and there are limited data on use of echocardiographic features beyond LVEF in patients with heart failure with mildly reduced or preserved LVEF.

METHODS

In patients with HF with mildly reduced or preserved LVEF identified in a large US health care system, the association of the following metrics with mortality was evaluated: LV global longitudinal strain (LV GLS>-16), left atrial volume index (>28 mL/m2), left ventricular hypertrophy (LVH), and E/e´>13 and e´<9. A multivariable model for mortality was constructed including age, sex, and key comorbidities followed by stepwise selection of echocardiographic features. Characteristics and outcomes of subgroups with normal versus abnormal LV GLS and LVEF were evaluated.

RESULTS

Among 2337 patients with complete echocardiographic data assessed between 2017 and 2020, the following features were associated with all-cause mortality on univariate analysis over 3 years of follow-up: E/e´+e´, LV GLS, left atrial volume index (all P<0.01). In the multivariable model (C-index=0.65), only abnormal LV GLS was independently associated with all-cause mortality (HR, 1.35 [95% CI, 1.11-1.63]; P=0.002). Among patients with LVEF>55%, 498/1255 (40%) demonstrated abnormal LV GLS. Regardless of specific LVEF, patients with abnormal LV GLS demonstrated a higher burden of multiple comorbidities and higher event rates compared with patients with normal LV GLS.

CONCLUSIONS

In a large, real-world HF with mildly reduced or preserved LVEF population, echocardiographic features, led by LV GLS, were associated with adverse outcomes irrespective of LVEF. A large proportion of patients demonstrate adverse myocardial function by LV GLS despite preserved LVEF and may represent a key cohort of interest for HF medical therapies and future clinical studies.

Utility of a risk assessment model in predicting 30 day unplanned hospital readmission in adult patients receiving outpatient parenteral antimicrobial therapy

Objectives

Outpatient parenteral antimicrobial therapy (OPAT) is associated with high hospital readmission rates. A 30 day unplanned readmission risk prediction model for OPAT patients has been developed in the UK. Given significant differences in patient mix and methods of OPAT delivery, we explored the model for its utility in Duke University Health System (DUHS) patients receiving OPAT.

Methods

We analysed OPAT episodes of adult patients from two hospitals between 1 July 2019 and 1 February 2020. The discriminative ability of the model to predict 30 day unplanned all-cause and OPAT-related admission was examined. An updated model was created by logistic regression with the UK risk factors and additional risk factors, OPAT delivery in a skilled nursing facility, vancomycin use and IV drug abuse.

Results

Compared with patients of the UK cohort, our study patients were of higher acuity, treated for more invasive infections, and received OPAT through different modes. The 30 day unplanned readmission rate in our cohort was 20% (94/470), with 59.5% (56/94) of those being OPAT-related. The original model was unable to discriminate for all-cause readmission with a C-statistic of 0.52 (95% CI 0.46-0.59) and for OPAT-related readmission with a C-statistic of 0.55 (95% CI 0.47-0.64). The updated model with additional risk factors did not have improved performance, with a C-statistic of 0.55 (95% CI 0.49-0.62).

Conclusions

The UK 30 day unplanned hospital readmission model performed poorly in predicting readmission for the OPAT population at a US academic medical centre.

Association of chronic non-cancer pain status and buprenorphine treatment retention among individuals with opioid use disorder: Results from electronic health record data

BACKGROUND

Although chronic non-cancer pain (CNCP) is common among individuals with opioid use disorder (OUD), its impact on buprenorphine treatment retention is unclear. The goal of this study was to use electronic health record (EHR) data to examine the association of CNCP status and 6-month buprenorphine retention among patients with OUD.

METHODS

We analyzed EHR data of patients with OUD who received buprenorphine treatment in an academic healthcare system between 2010 and 2020 (N = 676). We used Kaplan-Meier curves and Cox proportional hazards regression to estimate risk of buprenorphine treatment discontinuation (≥90 days between subsequent prescriptions). We used Poisson regression to estimate the association of CNCP and the number of buprenorphine prescriptions over 6 months.

RESULTS

Compared to those without CNCP, a higher proportion of patients with CNCP were of older age and had comorbid diagnoses for psychiatric and substance use disorders. There were no differences in the probability of buprenorphine treatment continuation over 6 months by CNCP status (p = 0.15). In the adjusted cox regression model, the presence of CNCP was not associated with time to buprenorphine treatment discontinuation (HR = 0.90, p = 0.28). CNCP status was associated with a higher number of prescriptions over 6 months (IRR = 1.20, p < 0.01).

CONCLUSIONS

These findings suggest that the presence of CNCP alone cannot be reliably associated with buprenorphine retention in patients with OUD. Nonetheless, providers should be aware of the association between CNCP and greater psychiatric comorbidity among patients with OUD when developing treatment plans. Research on the influence of additional characteristics of CNCP on treatment retention is needed.

Practice Patterns of Fundoscopic Examination for Diabetic Retinopathy Screening in Primary Care

IMPORTANCE

Primary care professionals (PCPs) have a central role in screening for diabetic retinopathy (DR), especially in settings where access to specialty eye care is limited. Data on current DR screening practice patterns in primary care are needed to inform screening strategies.

OBJECTIVES

To assess the practice patterns of fundoscopic examination for DR screening in a large primary care network and to evaluate the sensitivity and accuracy of PCP fundoscopy for detecting DR.

DESIGN, SETTING, AND PARTICIPANTS

A retrospective cohort study was performed using random sampling and manual review of electronic health records of PCP fundoscopic examination documentation compared with documentation of an examination performed by an eye care professional (ophthalmologist or optometrist) within 2 years before or after primary care encounters. From a single-institution primary care network of 28 clinics, 7449 adult patients with diabetes seen at least once in the primary care network in 2019 were eligible for this study. Data from 2001 encounters were abstracted from the electronic health record for a random sample of 767 patients. Data analysis was performed from January 2021 to May 2022.

EXPOSURE

Fundoscopic examination by PCPs.

MAIN OUTCOMES AND MEASURES

The frequency of PCPs performing fundoscopy at least once in the calendar year for patients with diabetes. Univariate and multivariable logistic regression analyses were performed to identify patient, clinician, and clinic factors associated with PCPs performing fundoscopy at least once in the calendar year. The PCP examination results were compared with diagnoses made by eye care professionals to assess the sensitivity and accuracy of the findings from PCP examinations.

RESULTS

Among the 767 adult patients with diabetes included in the analysis, 387 (50.5%) were female, and the median age was 64 years (IQR, 54-71 years). Primary care professionals documented a fundoscopic examination for 93 patients (12.1%); all results were documented as normal. When eye care professional examination results were used as the reference standard, the accuracy of PCP fundoscopic examination was 62.7% (95% CI, 50.0%-73.9%) and sensitivity for detecting disease was 0.0% (95% CI, 0.0%-14.9%). No patient demographic or clinical characteristics were associated with PCPs performing fundoscopy. In multivariable logistic regression, the number of PCP years in practice was associated with greater odds of patients receiving fundoscopy at least once in the year (adjusted odds ratio per 10 years in practice, 1.26; 95% CI, 1.01-1.59; P = .04); having nurse practitioner credentials was associated with lower odds of receiving fundoscopy (adjusted odds ratio, 0.23; 95% CI, 0.04-0.79; P = .049; compared with having physician credentials); after adjusting for rural clinic location, clinic location in a primary care shortage area, and documentation of an up-to-date eye care professional examination by a PCP in the study year.

CONCLUSIONS AND RELEVANCE

In this cohort study, fundoscopic examination was rarely performed and was not sensitive for detecting DR in primary care practice. Because the rate of DR screening by eye care professionals remains low, research to explore and break down barriers to the implementation of effective primary care-based DR screening strategies, such as teleretinal imaging, is needed to prevent vision loss from undiagnosed DR.

Underutilization of Guideline-based Abdominal Aortic Aneurysm Screening in an Academic Health System

OBJECTIVES

The US Preventive Services Task Force (USPSTF) recommends a 1-time screening for AAA with ultrasonography in men aged 65-75 who have ever smoked. Our objectives were to identify the AAA screening rates in a large academic health system and assess factors associated with receipt of screening.

METHODS

Data were extracted from electronic health record data from the Duke University Health System and the US Census Bureau. Index screening eligibility date was defined as the 65^th birthdate for male patients with a history of smoking. Patients with an index screening eligibility date between 1/1/2016 and 12/31/2018 were included in the study population and followed through 12/31/2019. Screened patients were identified by procedure codes for ultrasonography, CT or MRI.

RESULTS

Among 6,682 eligible patients who turned 65 years old between 1/1/2016 and 12/31/2018 with at least 1 year of follow-up, only 463 (6.9%) received AAA screening during the study period. The odds of receiving AAA screening within 1 year of index eligibility were 27% lower for Black patients compared to whites [OR=0.73, 95% CI (0.58,0.93)]. Patients who visited a PCP or had hypertension had 75% and 41% greater odds of receiving screening, respectively [OR 1.75, 95% CI (1.36,2.25)] and [OR 1.41 95% CI (1.11,1.80)] compared with patients who did not. Among 4,580 men with 2 years of follow-up, AAA screening rate increased to 13.0%. Patients who visited a PCP had 64% greater odds of receiving screening within 2 years of index eligibility compared to those who did not [OR=1.64, 95% CI (1.30,2.06)].

CONCLUSION

Screening for AAA per USPSTF guidelines is underutilized with evidence of a racial disparity. Although PCP visit is the most consistent predictor of screening, provider screening rates are low.

Leveraging Clinical Informatics Tools to Extract Cumulative Anthracycline Exposure, Measure Cardiovascular Outcomes, and Assess Guideline Adherence for Children With Cancer

PURPOSE

Cardiovascular disease is a significant cause of late morbidity and mortality in survivors of childhood cancer. Clinical informatics tools could enhance provider adherence to echocardiogram guidelines for early detection of late-onset cardiomyopathy.

METHODS

Cancer registry data were linked to electronic health record data. Structured query language facilitated the construction of anthracycline-exposed cohorts at a single institution. Primary outcomes included the data quality from automatic anthracycline extraction, sensitivity of International Classification of Disease coding for heart failure, and adherence to echocardiogram guideline recommendations.

RESULTS

The final analytic cohort included 385 pediatric oncology patients diagnosed between July 1, 2013, and December 31, 2018, among whom 194 were classified as no anthracycline exposure, 143 had low anthracycline exposure (< 250 mg/m²), and 48 had high anthracycline exposure (≥ 250 mg/m²). Manual review of anthracycline exposure was highly concordant (95%) with the automatic extraction. Among the unexposed group, 15% had an anthracycline administered at an outside institution not captured by standard query language coding. Manual review of echocardiogram parameters and clinic notes yielded a sensitivity of 75%, specificity of 98%, and positive predictive value of 68% for International Classification of Disease coding of heart failure. For patients with anthracycline exposure, 78.5% (n = 62) were adherent to guideline recommendations for echocardiogram surveillance. There were significant association with provider adherence and race and ethnicity (P = .047), and 50% of patients with Spanish as their primary language were adherent compared with 90% of patients with English as their primary language (P = .003).

CONCLUSION

Extraction of treatment exposures from the electronic health record through clinical informatics and integration with cancer registry data represents a feasible approach to assess cardiovascular disease outcomes and adherence to guideline recommendations for survivors.

An atomic approach to the design and implementation of a research data warehouse

Objective

As a long-standing Clinical and Translational Science Awards (CTSA) Program hub, the University of Pittsburgh and the University of Pittsburgh Medical Center (UPMC) developed and implemented a modern research data warehouse (RDW) to efficiently provision electronic patient data for clinical and translational research.

Materials and Methods

We designed and implemented an RDW named Neptune to serve the specific needs of our CTSA. Neptune uses an atomic design where data are stored at a high level of granularity as represented in source systems. Neptune contains robust patient identity management tailored for research; integrates patient data from multiple sources, including electronic health records (EHRs), health plans, and research studies; and includes knowledge for mapping to standard terminologies.

Results

Neptune contains data for more than 5 million patients longitudinally organized as Health Insurance Portability and Accountability Act (HIPAA) Limited Data with dates and includes structured EHR data, clinical documents, health insurance claims, and research data. Neptune is used as a source for patient data for hundreds of institutional review board-approved research projects by local investigators and for national projects.

Discussion

The design of Neptune was heavily influenced by the large size of UPMC, the varied data sources, and the rich partnership between the University and the healthcare system. It includes several unique aspects, including the physical warehouse straddling the University and UPMC networks and management under an HIPAA Business Associates Agreement.

Conclusion

We describe the design and implementation of an RDW at a large academic healthcare system that uses a distinctive atomic design where data are stored at a high level of granularity.

Research Integrated Network of Systems (RINS): a virtual data warehouse for the acceleration of translational research

OBJECTIVE

Integrated, real-time data are crucial to evaluate translational efforts to accelerate innovation into care. Too often, however, needed data are fragmented in disparate systems. The South Carolina Clinical & Translational Research Institute at the Medical University of South Carolina (MUSC) developed and implemented a universal study identifier-the Research Master Identifier (RMID)-for tracking research studies across disparate systems and a data warehouse-inspired model-the Research Integrated Network of Systems (RINS)-for integrating data from those systems.

MATERIALS AND METHODS

In 2017, MUSC began requiring the use of RMIDs in informatics systems that support human subject studies. We developed a web-based tool to create RMIDs and application programming interfaces to synchronize research records and visualize linkages to protocols across systems. Selected data from these disparate systems were extracted and merged nightly into an enterprise data mart, and performance dashboards were created to monitor key translational processes.

RESULTS

Within 4 years, 5513 RMIDs were created. Among these were 726 (13%) bridged systems needed to evaluate research study performance, and 982 (18%) linked to the electronic health records, enabling patient-level reporting.

DISCUSSION

Barriers posed by data fragmentation to assessment of program impact have largely been eliminated at MUSC through the requirement for an RMID, its distribution via RINS to disparate systems, and mapping of system-level data to a single integrated data mart.

CONCLUSION

By applying data warehousing principles to federate data at the "study" level, the RINS project reduced data fragmentation and promoted research systems integration.

ACE: the Advanced Cohort Engine for searching longitudinal patient records

OBJECTIVE

To propose a paradigm for a scalable time-aware clinical data search, and to describe the design, implementation and use of a search engine realizing this paradigm.

MATERIALS AND METHODS

The Advanced Cohort Engine (ACE) uses a temporal query language and in-memory datastore of patient objects to provide a fast, scalable, and expressive time-aware search. ACE accepts data in the Observational Medicine Outcomes Partnership Common Data Model, and is configurable to balance performance with compute cost. ACE's temporal query language supports automatic query expansion using clinical knowledge graphs. The ACE API can be used with R, Python, Java, HTTP, and a Web UI.

RESULTS

ACE offers an expressive query language for complex temporal search across many clinical data types with multiple output options. ACE enables electronic phenotyping and cohort-building with subsecond response times in searching the data of millions of patients for a variety of use cases.

DISCUSSION

ACE enables fast, time-aware search using a patient object-centric datastore, thereby overcoming many technical and design shortcomings of relational algebra-based querying. Integrating electronic phenotype development with cohort-building enables a variety of high-value uses for a learning health system. Tradeoffs include the need to learn a new query language and the technical setup burden.

CONCLUSION

ACE is a tool that combines a unique query language for time-aware search of longitudinal patient records with a patient object datastore for rapid electronic phenotyping, cohort extraction, and exploratory data analyses.

Reduction in Expected Survival Associated With Nontuberculous Mycobacterial Pulmonary Disease

BACKGROUND

Nontuberculous mycobacteria (NTM) are opportunistically pathogenic bacteria that are found abundantly in the soil and water. Susceptible individuals exposed to NTM-containing aerosols from environmental sources may develop NTM pulmonary disease (NTM-PD). Reported survival after NTM-PD diagnosis varies widely among existing studies. Prior work has suggested that mortality among persons with NTM-PD is primarily driven by comorbidities rather than NTM-PD.

METHODS

We retrospectively identified a cohort of patients in the Duke University Health System who were diagnosed with NTM-PD between 1996 and 2015. Hospitalizations and survival were compared among patients with NTM-PD with and without other comorbidities. Additionally, survival among patients with NTM-PD was compared with standardized mortality data for a similar cohort of the general population.

RESULTS

Patients with NTM-PD without other comorbidities had 0.65 hospitalizations/1000 patient-days compared with 1.37 hospitalizations/1000 patient-days for patients with other comorbidities. Compared with a cohort of the general population, expected survival decreased by approximately 4 years for a diagnosis of NTM-PD without comorbidities and 8.6 years for a diagnosis of NTM-PD with comorbidities. Mortality 5 years after diagnosis was 25.0% and 44.9% among NTM patients without and with comorbidities, respectively, compared with 5.7% in the general-population cohort.

CONCLUSIONS

NTM-PD was associated with significant morbidity that was worse in patients with comorbidities. Patients with NTM-PD, even without comorbidities, had worse survival than expected.

FIBER: enabling flexible retrieval of electronic health records data for clinical predictive modeling

OBJECTIVES

The development of clinical predictive models hinges upon the availability of comprehensive clinical data. Tapping into such resources requires considerable effort from clinicians, data scientists, and engineers. Specifically, these efforts are focused on data extraction and preprocessing steps required prior to modeling, including complex database queries. A handful of software libraries exist that can reduce this complexity by building upon data standards. However, a gap remains concerning electronic health records (EHRs) stored in star schema clinical data warehouses, an approach often adopted in practice. In this article, we introduce the FlexIBle EHR Retrieval (FIBER) tool: a Python library built on top of a star schema (i2b2) clinical data warehouse that enables flexible generation of modeling-ready cohorts as data frames.

MATERIALS AND METHODS

FIBER was developed on top of a large-scale star schema EHR database which contains data from 8 million patients and over 120 million encounters. To illustrate FIBER's capabilities, we present its application by building a heart surgery patient cohort with subsequent prediction of acute kidney injury (AKI) with various machine learning models.

RESULTS

Using FIBER, we were able to build the heart surgery cohort (n = 12 061), identify the patients that developed AKI (n = 1005), and automatically extract relevant features (n = 774). Finally, we trained machine learning models that achieved area under the curve values of up to 0.77 for this exemplary use case.

CONCLUSION

FIBER is an open-source Python library developed for extracting information from star schema clinical data warehouses and reduces time-to-modeling, helping to streamline the clinical modeling process.

The Determinants for Food Safety Push Notifications on Continuance Intention in an E-Appointment System for Public Health Medical Services: The Perspectives of UTAUT and Information System Quality

Compared to other appointment methods in public hospitals, registering through the Internet or utilizing e-appointments, or registering online as an outpatient, can provide more information to the user. This research investigated the integration of unified theory of the acceptance and use of technology and information system quality in determining factors that influence the adoption of e-appointments by patients, based on the requirements of food safety consultation in Taiwan. Empirical data from 369 valid samples were assessed using Partial Least Squares (PLS). The key findings of this study indicated that patients’ acceptance of e-appointments was influenced by users’ perceptions (i.e., performance expectancy and facilitating conditions), along with information quality and service quality. The practical and academic implications are provided for future practitioners and scholars, and to enhance patients’ usage of e-appointments in their healthcare activities.

Recurrent Tube Erosions with Anti-Vascular Endothelial Growth Factor Therapy in Patients with Age-Related Macular Degeneration

PURPOSE

To determine the rates of primary and recurrent glaucoma tube shunt erosions in patients with age-related macular degeneration (AMD) receiving anti-vascular endothelial growth factor (VEGF) intravitreal injections.

DESIGN

Retrospective case series.

PARTICIPANTS

Patients with AMD who underwent tube revision for erosion at the Duke Eye Center from January 1, 1999, to January 1, 2019, were identified. Patients with and without anti-VEGF injections were compared.

METHODS

Patient demographics, ocular diagnoses, glaucoma tube shunt types and locations, and dates of glaucoma surgeries and anti-VEGF injections were collected. Statistical analyses were performed with P < 0.05 as significant.

MAIN OUTCOME MEASURES

Outcome measures included the number of tube erosions, time from anti-VEGF injection to tube erosion, and secondary complications after tube revisions.

RESULTS

A total of 150 patients with AMD with anti-VEGF (309 tubes) and 262 patients with AMD without anti-VEGF (459 tubes) were identified. There was no statistically significant difference in the number of tube erosions in patients with anti-VEGF (15 tubes, 4.8%) versus without anti-VEGF (12 tubes, 2.6%) (P = 0.10). However, patients receiving anti-VEGF had on average a greater number of tube erosion events (2.1±0.7 events) compared with patients without anti-VEGF (1.3±0.7, P < 0.01). Ten patients (91%) received concurrent anti-VEGF injections at the time of tube erosion, and the average duration of prior anti-VEGF therapy was approximately 2 years. Tube erosion was noted 46.5±60.7 days from the preceding anti-VEGF injection. Secondary complications after tube revision in the anti-VEGF group included 5 explanted tubes for recurrent erosions.

CONCLUSIONS

Our results suggest intravitreal anti-VEGF injections are linked to higher rates of recurrent glaucoma tube erosions in patients with AMD. The majority of patients received chronic and serial anti-VEGF injections. Thus, additional consideration should be given to glaucoma surgical planning in patients receiving anti-VEGF injections, especially in those with a primary tube erosion.

Development of an electronic health records datamart to support clinical and population health research

Introduction:

Electronic health record (EHR) data have emerged as an important resource for population health and clinical research. There have been significant efforts to leverage EHR data for research; however, given data security concerns and the complexity of the data, EHR data are frequently difficult to access and use for clinical studies. We describe the development of a Clinical Research Datamart (CRDM) that was developed to provide well-curated and easily accessible EHR data to Duke University investigators.

Methods:

The CRDM was designed to (1) contain most of the patient-level data elements needed for research studies; (2) be directly accessible by individuals conducting statistical analyses (including Biostatistics, Epidemiology, and Research Design (BERD) core members); (3) be queried via a code-based system to promote reproducibility and consistency across studies; and (4) utilize a secure protected analytic workspace in which sensitive EHR data can be stored and analyzed. The CRDM utilizes data transformed for the PCORnet data network, and was augmented with additional data tables containing site-specific data elements to provide additional contextual information.

Results:

We provide descriptions of ideal use cases and discuss dissemination and evaluation methods, including future work to expand the user base and track the use and impact of this data resource.

Conclusions:

The CRDM utilizes resources developed as part of the Clinical and Translational Science Awards (CTSAs) program and could be replicated by other institutions with CTSAs.

PROSPeCT: A Predictive Research Online System for Prostate Cancer Tasks

PURPOSE

An online clinical information system, called Predictive Research Online System Prostate Cancer Tasks (PROSPeCT), was developed to enable users to query the Alberta Prostate Cancer Registry database hosted by the Alberta Prostate Cancer Research Initiative. To deliver high-quality patient treatment, prostate cancer clinicians and researchers require a user-friendly system that offers an easy and efficient way to obtain relevant and accurate information about patients from a robust and expanding database.

METHODS

PROSPeCT was designed and implemented to make it easy for users to query the prostate cancer patient database by creating, saving, and reusing simple and complex definitions. We describe its intuitive nature by exemplifying the creation and use of a complex definition to identify a "high-risk" patient cohort.

RESULTS

PROSPeCT was made to minimize user error and to maximize efficiency without requiring the user to have programming skills. Thus, it provides tools that allow both novice and expert users to easily identify patient cohorts, manage individual patient care, perform Kaplan Meier estimates, plot aggregate PSA views, compute PSA-doubling time, and visualize results.

CONCLUSION

This report provides an overview of PROSPeCT, a system that helps clinicians to identify appropriate patient treatments and researchers to develop prostate cancer hypotheses, with the overarching goal of improving the quality of life of patients with prostate cancer. We have made available the code for the PROSPeCT implementation at https://github.com/max-uhlich/e-PROSPeCT .

Chronic non-cancer pain among adults with substance use disorders: Prevalence, characteristics, and association with opioid overdose and healthcare utilization

BACKGROUND

Chronic non-cancer pain (CNCP) among patients with substance use disorder (SUD) poses a risk for worse treatment outcomes. Understanding the association of CNCP with SUD is important for informing the need and potential benefits of pain assessment/management among those with SUDs.

METHODS

We analyzed electronic health record data from 2013 to 2018 among adults aged ≥18 years (N = 951,533; mean age: 48.4 years; 57.4 % female) in a large academic healthcare system. Adjusted logistic regression models were conducted to estimate the association of CNCP conditions with opioid overdose, emergency department utilization, and inpatient hospitalization stratified by different SUD diagnoses and by gender.

RESULTS

Among the total sample, the prevalence of CNCP was 46.6 % and any SUD was 11.2 %. The majority of patients with a SUD had CNCP (opioid: 74.7 %; sedative: 72.3 %; cannabis: 64.3 %; alcohol: 58.7 %; tobacco: 59.5 %). The prevalence of CNCP was greater in females vs. males for most SUD diagnoses. The presence of CNCP was associated with more mental health disorders and chronic medical conditions among each SUD group. CNCP was associated with significantly decreased odds of overdose among those with opioid use disorder but increased odds of overdose and healthcare utilization among other SUDs. CNCP was positively associated with overdose in females, but not males, with alcohol or non-opioid drug use disorders.

CONCLUSIONS

The direction and magnitude of the association between CNCP and negative health indicators differed as a function of SUD type and gender, respectively. Greater awareness of potential unmet pain treatment need may have implications for improving SUD outcomes.

Length of Stay Predicts Risk of Early Infection for Hospitalized Patients Undergoing Central Venous Port Placement

PURPOSE

To compare early totally implantable central venous port catheter-related infection rates after inpatient vs outpatient placement and to determine whether the risk associated with inpatient placement is influenced by length of hospital stay.

MATERIALS AND METHODS

In this single-institution retrospective study, 5,301 patients (3,618 women; mean age 57 y) underwent port placement by interventional radiologists between October 2004 and January 2018. The 30-day infection rate was compared between inpatients and outpatients using survival analysis. Among inpatients, the effect of time from admission to port placement and from placement to discharge was analyzed using a survival regression tree.

RESULTS

The 30-day infection rate was 3.6% (95% confidence interval [CI] = 1.9%-6.1%) among 386 inpatients and 1.0% (95% CI = 0.7%-1.3%) among 4,915 outpatients (hazard ratio [HR] = 3.6, 95% CI = 2.0-6.6, P < .001). Inpatient placement was a significant risk factor after accounting for covariates in multivariate analysis (HR = 2.2, 95% CI = 1.0-4.7, P = .05) and controlling for demographic differences by propensity score matching (HR = 2.8, 95% CI = 1.0-7.8, P = .04). Infection rate was 11% (95% CI = 4.7%-22%) among 65 inpatients in whom time from admission to placement was ≥ 7 days, 5.1% (95% CI = 1.9%-11%) among 129 inpatients in whom admission to placement was < 7 days and time to discharge was > 3 days, and 0% (95% CI = 0%-2.1%) among 192 inpatients in whom admission to placement was < 7 days and time to discharge was ≤ 3 days (P < .001).

CONCLUSIONS

Inpatient port placement was associated with a higher risk of early infection. However, a clinical decision tree based on shorter length of stay before and after placement may identify a subset of hospitalized patients not at increased risk for infection.

Age matters: older age as a risk factor for CMV reactivation in the CMV serostatus-positive kidney transplant recipient

Evaluate risk factors for cytomegalovirus (CMV) reactivation during the first year after kidney transplantation in the CMV-seropositive older recipient. Retrospective single-center study. Between 2011 and 2015, 91 patients ≥ 65 years received a kidney transplant; these were matched with 91 controls, aged 40-60. Risk of CMV reactivation in the CMV-seropositive recipients was analyzed. Sixty-three older and 54 younger recipients were included; 50% had received CMV-directed prophylaxis. CMV reactivation was significantly more frequent in the older group (71.4% vs 44.4%, p = 0.003) and occurred earlier (p = 0.003). A multivariate model showed that only age was associated with CMV reactivation (OR 2.48, p = 0.03). After excluding patients that received thymoglobulin, older age group remained the only risk factor of CMV reactivation (OR 3.81, p = 0.014). Recurrent event analysis showed that the older cohort had an HR of 1.94 (p = 0.01) of CMV viremia; there was significant episode-cohort interaction (p < 0.01). While the older group had a higher risk of infection (HR = 2.43), after the initial episode the relative hazards were approximately equal (HR = 1.08, at period 2). This suggests that it is key to specifically avoid the first episode of reactivation. Universal prophylaxis or a hybrid prophylaxis model should be considered in the CMV-seropositive kidney transplant recipient aged ≥ 65 years.

Pulmonary Hypertension Subtypes and Mortality in CKD

RATIONALE & OBJECTIVE

Pulmonary hypertension (PH) contributes to cardiovascular disease and mortality in patients with chronic kidney disease (CKD), but the pathophysiology is mostly unknown. This study sought to estimate the prevalence and consequences of PH subtypes in the setting of CKD.

STUDY DESIGN

Observational retrospective cohort study.

SETTING & PARTICIPANTS

We examined 12,618 patients with a right heart catheterization in the Duke Databank for Cardiovascular Disease from January 1, 2000, to December 31, 2014.

EXPOSURES

Baseline kidney function stratified by CKD glomerular filtration rate category and PH subtype.

OUTCOMES

All-cause mortality.

ANALYTICAL APPROACH

Multivariable Cox proportional hazards analysis.

RESULTS

In this cohort, 73.4% of patients with CKD had PH, compared with 56.9% of patients without CKD. Isolated postcapillary PH (39.0%) and combined pre- and postcapillary PH (38.3%) were the most common PH subtypes in CKD. Conversely, precapillary PH was the most common subtype in the non-CKD cohort (35.9%). The relationships between mean pulmonary artery pressure, pulmonary capillary wedge pressure, and right atrial pressure with mortality were similar in both the CKD and non-CKD cohorts. Compared with those without PH, precapillary PH conferred the highest mortality risk among patients without CKD (HR, 2.27; 95% CI, 2.00-2.57). By contrast, in those with CKD, combined pre- and postcapillary PH was associated with the highest risk for mortality in CKD in adjusted analyses (compared with no PH, HRs of 1.89 [95% CI, 1.57-2.28], 1.87 [95% CI, 1.52-2.31], 2.13 [95% CI, 1.52-2.97], and 1.63 [95% CI, 1.12-2.36] for glomerular filtration rate categories G3a, G3b, G4, and G5/G5D).

LIMITATIONS

The cohort referred for right heart catheterization may not be generalizable to the general population. Serum creatinine data in the 6 months preceding catheterization may not reflect true baseline CKD. Observational design precludes assumptions of causality.

CONCLUSIONS

In patients with CKD referred for right heart catheterization, PH is common and associated with poor survival. Combined pre- and postcapillary PH was common and portended the worst survival for patients with CKD.

Where Did They Go? Tracking Young Adult Follow-up During the Transition From Pediatric to Adult-Oriented Care

Continuity of care during health care transition is critical. We sought to evaluate electronic medical record clinic attendance data to examine this outcome. We identified 1623 patients (ages 18-27 years) with 1 of 6 childhood-onset chronic conditions and tracked clinic utilization from January 2002 to July 2016. Patients were classified as active in pediatric care; lost from pediatric care; successfully transferred; or lost from adult care. Using random effects logistic regression, we compared the number of days between last pediatric and first adult visit to each clinic's self-reported transitional care quality score. In this cohort, >44% remained active in care at the end of the study. Clinics with higher proportions of successfully transferred patients had lower median numbers of days between last pediatric and first adult visit and higher transitional care quality scores. Characterizing utilization patterns with electronic medical record data allows health systems to track transitional care outcomes and target improvement efforts.

Design and Implementation of a Comprehensive Surveillance System for Venous Thromboembolism in a Defined Region Using Electronic and Manual Approaches

BACKGROUND

Systematic surveillance for venous thromboembolism (VTE) in the United States has been recommended by several organizations. Despite adoption of electronic medical records (EMRs) by most health care providers and facilities, however, systematic surveillance for VTE is not available.

OBJECTIVES

This article develops a comprehensive, population-based surveillance strategy for VTE in a defined geographical region.

METHODS

The primary surveillance strategy combined computerized searches of the EMR with a manual review of imaging data at the Duke University Health System in Durham County, North Carolina, United States. Different strategies of searching the EMR were explored. Consolidation of results with autopsy reports (nonsearchable in the EMR) and with results from the Durham Veterans' Administration Medical Center was performed to provide a comprehensive report of new VTE from the defined region over a 2-year timeframe.

RESULTS

Monthly searches of the primary EMR missed a significant number of patients with new VTE who were identified by a separate manual search of radiology records, apparently related to delays in data entry and coding into the EMR. Comprehensive searches incorporating a location-restricted strategy were incomplete due to the assigned residence reflecting the current address and not the address at the time of event. The most comprehensive strategy omitted the geographic restriction step and identified all patients with VTE followed by manual review of individual records to remove incorrect entries (e.g., outside the surveillance time period or geographic location; no evidence for VTE). Consolidation of results from the EMR searches with results from autopsy reports and the separate facility identified additional patients not diagnosed within the Duke system.

CONCLUSION

We identified several challenges with implementing a comprehensive VTE surveillance program that could limit accuracy of the results. Improved electronic strategies are needed to cross-reference patients across multiple health systems and to minimize the need for manual review and confirmation of results.

Infections after kidney transplantation. Does age matter?

Infections threaten successful outcomes after kidney transplantation. Our aim was to determine if the number, types of infections and the risk factors for common infections differed between older compared to younger kidney transplant (KT) recipients in the first year after surgery. We performed a single-center retrospective cohort study. Between 2011 and 2015, 91 KTs were performed in patients ≥65 years of age; these were matched 1:1 (by year of transplantation, sex and race) to controls aged 40-60 years. Over 90% of both groups had an infectious complication. Urinary tract infections (UTIs) and cytomegalovirus (CMV) viremia were significantly more frequent in older recipients. Older adults had more late onset UTIs, including after stent removal. CMV viremia was more frequent in older adults in the 1-6 months post-transplant period. Due to our center-specific protocol utilizing pre-emptive monitoring in the CMV recipient-seropositive population, the higher CMV incidence in the aged recipient was driven by this subpopulation of older adults. No difference in pneumonias or bloodstream infections were found, nor in surgical complications, rejection or graft loss. Mortality was higher at 1-year post-transplant in the older recipients (9.9% vs 1.1%; P = 0.018). Prophylactic and immunosuppressive strategies may need to be altered for older KT recipients.

Group B Streptococcus (GBS) Colonization and Disease among Pregnant Women: A Historical Cohort Study

Background

Maternal GBS colonization is associated with early-onset neonatal sepsis and extensive efforts are directed to preventing this complication. Less is known about maternal risks of GBS colonization. We seek to provide a modern estimate of the incidence and impact of maternal GBS colonization and invasive GBS disease.

Methods

A single center historical cohort study of all births between 2003 and 2015 was performed. Data was collected via electronic health record abstraction using an institutional specific tool. Descriptive statistics were performed regarding GBS status. Inferential statistics were performed comparing risk of adverse pregnancy outcomes in cohorts with and without GBS colonization as well as cohorts with GBS colonization and invasive GBS disease.

Results

A total of 60,029 deliveries were included for analysis. Overall, 21.6% of the population was GBS colonized and 0.1% had invasive GBS disease. GBS colonization was associated with younger maternal age, Black race, non-Hispanic ethnicity, chronic hypertension, preexisting diabetes, and tobacco use (p<0.01). In the adjusted analyses, there was an increased risk of gestational diabetes (aRR 1.21, 95% CI 1.11-1.32) in colonized pregnancies and a decreased incidence of short cervix (aRR 0.64, 95% CI 0.52-0.79), chorioamnionitis (aRR 0.76, 95% CI 0.66-0.87), wound infection (aRR 0.75, 95% CI 0.64-0.88), and operative delivery (aRR 0.85, 95% CI 0.83-0.88).

Conclusions

This modern-day large cohort of all births over a 12-year period demonstrates a GBS colonization rate of 21.6%. This data reflects a need to assess maternal and perinatal outcomes in addition to neonatal GBS sepsis rates to inform decisions regarding the utility of maternal vaccination.

No Increased Mortality After TIPS Compared with Serial Large Volume Paracenteses in Patients with Higher Model for End-Stage Liver Disease Score and Refractory Ascites

PURPOSE

To compare survival after transjugular intrahepatic portosystemic shunt (TIPS) creation versus serial large volume paracenteses (LVP) in patients with refractory ascites and higher Model for End-Stage Liver Disease (MELD) scores.

MATERIALS AND METHODS

In this retrospective study, from 1/1/2013 to 10/1/2018, 478 patients (294 male; mean age 58, range 23-89) underwent serial LVP (n = 386) or TIPS (n = 92) for ascites. Propensity-matched cohorts were constructed based on age, MELD, Charlson comorbidity index, varices, and hepatic encephalopathy. Survival was analyzed using a Cox proportional hazards model in which MELD score and TIPS were treated as time-dependent covariates. An interaction term was used to assess the impact of TIPS versus serial LVP on survival as a function of increasing MELD.

RESULTS

In the overall patient sample, higher MELD score predicted worse survival after either serial LVP or TIPS [hazard ratio (HR) = 1.13; p < 0.001], but there was no significant interaction between TIPS and higher MELD score conferring worse survival (HR = 1.01; p = 0.55). In 92 propensity-matched serial LVP and 92 TIPS patients, higher MELD score predicted worse survival after either serial LVP or TIPS (HR = 1.19; p < 0.001), but there was no significant survival interaction between TIPS and higher MELD (HR = 0.97; p = 0.22). In 30 propensity-matched serial LVP patients and 30 TIPS patients with baseline MELD greater than 18, TIPS did not predict worse survival (HR = 0.97; p = 0.94).

CONCLUSION

Higher MELD predicts poorer survival after either serial LVP or TIPS, but TIPS creation is not associated with worse survival compared to serial LVP in patients with higher MELD scores LEVEL OF EVIDENCE: Level 4, case series.

Design and Implementation of a Secure Computing Environment for Analysis of Sensitive Data at an Academic Medical Center

Academic medical centers need to make sensitive data from electronic health records, payer claims, genomic pipelines, and other sources available for analytical and educational purposes while ensuring privacy and security. Although many studies have described warehouses for collecting biomedical data, few studies have described secure computing environments for analysis of sensitive data. This case report describes the Weill Cornell Medicine Data Core with respect to user access, data controls, hardware, software, audit, and financial considerations. In the 2.5 years since launch, the Data Core has supported more than 200 faculty, staff, and students across nearly 60 research and education projects. Other institutions may benefit from adopting elements of the approach, including tools available on Github, for balancing access with privacy and security.

Multicomorbidity of chronic diseases and substance use disorders and their association with hospitalization: Results from electronic health records data

BACKGROUND

Chronic diseases are prevalent and the leading causes of mortality. Comorbidity of substance use disorders (SUDs) and chronic diseases is understudied to inform behavioral healthcare integration.

OBJECTIVES

This study leveraged electronic health record (EHR) data of 211,880 adults from a large health system to examine prevalence and correlates of comorbidity of SUDs and nine chronic disease groups and to determine their association with hospitalization.

METHODS

Logistic regression analyses were conducted to estimate associations between chronic diseases and SUDs. To control for severity of diagnosis, analyses of associations between SUD and hospitalization were stratified by the number of chronic conditions.

RESULTS

In the sample, 48.3% had ≥1 chronic condition (hypertension 33.7%, arthritis 16.2%, diabetes 13.7%, chronic kidney disease 9.9%, asthma 9.1%, chronic obstructive pulmonary disease 8.9%, ischemic heart disease 8.3%, cancer 4.6%, and hepatitis 1.3%). Prevalence of SUD (overall 13.3%) among patients increased with multiple chronic conditions (14.3% having SUD among patients with one condition; 21.2% having SUD among patients with two to three conditions; and 32.5% having SUD among patients with 4-9 conditions). Chronic conditions were associated with increased odds of SUDs. For all SUD groups, hospitalization was more prevalent among patients with SUD than those without it; prevalence of hospitalization increased with the number of comorbid chronic conditions.

CONCLUSIONS

Findings reveal a striking pattern of multicomorbidity of SUD and chronic diseases and its positive association with hospitalization. Behavioral healthcare integration should consider efforts to assess and treat comorbid SUD and chronic diseases, especially among adults with multiple chronic conditions.

Single-center analysis of infectious complications in older adults during the first year after kidney transplantation

Infections are among the top three causes of death of older adults in the first year after kidney transplantation (KT). Our aim was to describe infectious complications among KT recipients aged ≥ 65 during the first 12 months post-transplant. Single-center retrospective cohort study. Ninety-one KTs had been performed in patients ≥ 65 years of age between 2011 and 2015. 92.3% of the patients developed at least one infection. Infectious episodes increased the risk of future infection by 10% (p = 0.0018) with each infection portending a greater risk. At a patient level, viral (71.4%) and bacterial (70.2%) infections predominated. Urinary tract infections were the most frequent complication (30.3%), followed by cytomegalovirus infections (22.7%). Infections were the main reason for readmission. 7.7% of the patients developed rejection; and overall 3.3% lost their graft. Mortality at 1 year was 9.9%. Older KT recipients have a high incidence of infectious complications the first year after KT. Infections were the number one reason for readmission, and an infection episode predicted future infections for the individual patient. Despite these complications, the majority of older KT recipients were alive with a functioning graft at 1 year.

Ranolazine reduces atrial fibrillatory wave frequency

Aims

Antiarrhythmic medications for the treatment of atrial fibrillation (AF) have limited efficacy and rare but potentially life-threatening side effects. Ranolazine is an antianginal agent that may have antiarrhythmic activity in AF.

Methods and results

Using the Duke Enterprise Data Unified Content Explorer database, we analysed a cohort of AF patients on ranolazine. Patients served as their own historic control. Electrocardiograms (ECGs) were analysed before and after ranolazine initiation to determine the effect of ranolazine on dominant frequency (DF), f-wave amplitude, and organizational index (OI). We identified 15 patients with ECGs in AF before and after ranolazine. Ranolazine was associated with lower DF by an average of 10% (5.10 ± 0.74 vs. 5.79 ± 0.96 Hz, P = 0.04) but not with changes in OI (0.47 ± 0.11 vs. 0.50 ± 0.12, P = 0.71) or amplitude (0.47 ± 0.43 vs. 0.41 ± 0.40 mV, P = 0.82). Ranolazine was also associated with lower DF in patients (n = 10) not on concomitant antiarrhythmic therapy (5.25 ± 0.78 vs. 6.03 ± 0.79 Hz, P = 0.04).

Conclusion

Ranolazine is associated with lower AF DF but no change in OI or fibrillatory wave amplitude. Prospective trials are needed to evaluate ranolazine's potential as a novel antiarrhythmic drug for AF.

Usability Evaluation of an Unstructured Clinical Document Query Tool for Researchers

Natural Language Processing - Patient Information Extraction for Researchers (NLP-PIER) was developed for clinical researchers for self-service Natural Language Processing (NLP) queries with clinical notes. This study was to conduct a user-centered analysis with clinical researchers to gain insight into NLP-PIER's usability and to gain an understanding of the needs of clinical researchers when using an application for searching clinical notes. Clinical researcher participants (n=11) completed tasks using the system's two existing search interfaces and completed a set of surveys and an exit interview. Quantitative data including time on task, task completion rate, and survey responses were collected. Interviews were analyzed qualitatively. Survey scores, time on task and task completion proportions varied widely. Qualitative analysis indicated that participants found the system to be useful and usable in specific projects. This study identified several usability challenges and our findings will guide the improvement of NLP-PIER 's interfaces.

Nocardia infections in the transplanted host

BACKGROUND

Nocardia are uncommon pathogens that disproportionately afflict the immunocompromised host. Epidemiology and outcome data of Nocardia infections in transplant recipients are limited.

METHODS

We performed a retrospective chart review of all patients at Duke University Hospital with a history of solid organ transplant (SOT) or hematopoietic cell transplant (HCT) and at least one positive culture for Nocardia between 1996 and 2013. Our aim was to describe the epidemiology and outcomes of Nocardia infections in the transplanted host.

RESULTS

During the 18-year study period, 51 patients (14 HCT and 37 SOT recipients) had Nocardia infection. Nocardia incidence was stable during the study period in all populations except heart transplants, whose incidence declined. Infection occurred earlier in the HCT group than the SOT group (median time to diagnosis of 153 and 370 days, respectively). In both groups, the most common site involved was the lung. Outcomes were overall poor, especially in the HCT group with a cure rate of 29%. Heart transplant recipients had significantly better overall survival (P < .05) than other patients. Trimethoprim-sulfamethoxazole (TMP-SMX) prophylaxis did not provide complete protection from Nocardia infections, nor did it appear to select for resistant Nocardia isolates.

CONCLUSIONS

Infections with Nocardia are typically a late post-transplant complication. The use of TMP-SMX prophylaxis was not associated with TMP-SMX-resistant Nocardia. Overall outcomes remain poor.

Substance use disorders and medical comorbidities among high-need, high-risk patients with diabetes

BACKGROUND

The majority of the U.S. healthcare resources are utilized by a small population characterized as high-risk, high-need persons with complex care needs (e.g., adults with multiple chronic conditions). Substance use disorders (SUDs) and mental health disorders (MHDs) are a driver of poor health and additional healthcare costs, but they are understudied among high-need patients.

OBJECTIVE

We examine the prevalence and correlates of SUDs and MHDs among adults with high-risk diabetes, who are patients at the top 10% risk score for developing poor outcomes (hospital admission or death).

METHODS

A risk algorithm developed from Duke University Health System electronic health records (EHRs) data was used to identify patients with high-risk diabetes for targeting home-based primary care. The EHR data of the 263 patients with high-risk diabetes were analyzed to understand patterns of SUDs and MHDs to inform care-coordinating efforts.

RESULTS

Both SUDs (any SUD 48.3%, alcohol 12.5%, tobacco 38.8%, drug 23.2%) and MHDs (any MHD 74.9%, mood 53.2%, sleep 37.3%, anxiety 32.7%, schizophrenia/psychotics/delusional 14.8%, dementia/delirium/amnestic/cognitive 14.4%, adjustment 9.1%) were prevalent. Overall, 81.7% of the sample had SUD or MHD. Elevated odds of SUD were noted among men (tobacco, alcohol) and those who were never-married (alcohol, cannabis). African-American race (vs. other race/ethnicity) was associated with lower odds of anxiety disorders.

CONCLUSION

While data are limited to one large academic health system, they provide clinical evidence revealing that 82% of patients with high-risk diabetes had SUD and/or MHD recorded in their EHRs, highlighting a need for developing service models to optimize high-risk care.

Health Care Utilization Behaviors Predict Disengagement From HIV Care: A Latent Class Analysis

Background

The traditional definition of engagement in HIV care in terms of only clinic attendance and viral suppression provides a limited understanding of how persons living with HIV (PLWH) interact with the health care system.

Methods

We conducted a retrospective analysis of patients with ≥1 HIV clinic visits at the Duke Adult Infectious Diseases Clinic between 2008 and 2013. Health care utilization was characterized by 4 indicators: clinic attendance in each half of the year (yes/no), number of emergency department (ED) visits/year (0, 1, or 2+), inpatient admissions/year (0, 1, 2+), and viral suppression (never, intermittent, always). Health care engagement patterns were modeled using latent class/latent transition analysis.

Results 

A total of 2288 patients (median age, 46.4 years; 59% black, 71% male) were included in the analysis. Three care engagement classes were derived from the latent class model: "adherent" "nonadherent," and "sick." Patients age ≤40 years were more likely to be in the nonadherent class (odds ratio, 2.64; 95% confidence interval, 1.38-5.04) than other cohort members. Whites and males were more likely to transition from nonadherent to adherent the following year. Nonadherent patients were significantly more likely to disengage from care the subsequent year than adherent patients (23.6 vs 0.2%, P < .001).

Conclusions

A broader definition of health care engagement revealed distinct and dynamic patterns among PLWH that would have been hidden had only previous HIV clinic attendance had been considered. These patterns may be useful for designing engagement-targeted interventions.

Increasing Nocardia Incidence Associated with Bronchiectasis at a Tertiary Care Center

RATIONALE

Nocardia is a genus of pathogens that most commonly afflict immunocompromised hosts but may be an emerging infection among persons with bronchiectasis.

OBJECTIVES

To examine the epidemiology and clinical presentation of adult patients with Nocardia and bronchiectasis relative to other patient groups.

METHODS

We examined a retrospectively assembled cohort of adults at Duke University Hospital in Durham, North Carolina with at least one positive culture from a bodily fluid or tissue specimen for Nocardia between January 1996 and December 2013. Denominator data for key populations (e.g., bronchiectasis, transplant) were obtained using International Classification of Diseases, Ninth Revision codes. In addition, we performed a case-control analysis to examine the relationship between inhaled corticosteroid use and Nocardia lung infection among otherwise immunocompetent patients with bronchiectasis.

MEASUREMENTS AND MAIN RESULTS

We identified 183 patients with one or more cultures positive for Nocardia: 44 from 1996 to 2001, 64 from 2002 to 2007, and 75 from 2008 to 2013. Immune compromise was common (56%), particularly solid organ or hematopoietic cell transplant (30%). Infection usually was confined to the lungs (62%), followed by skin (10%), other sites (6%), brain (2%), and multiple sites (17%). Non-cystic fibrosis bronchiectasis was common among both immunocompetent (38%) and immunocompromised (10%) patients. Nocardia incidence in patients with bronchiectasis increased significantly over time, but there was no significant change in Nocardia incidence in hematopoietic cell or solid organ transplant recipients (our largest immunocompromised population). Among patients with bronchiectasis, Nocardia was positively but nonsignificantly associated with use of inhaled corticosteroids (odds ratio, 1.8; 95% confidence interval, 0.7-4.4).

CONCLUSIONS

The increasing incidence of Nocardia infections at our medical center appears to be driven by increased incidence in patients with bronchiectasis rather than increases in immunocompromised populations. It is unclear whether increased environmental exposures, microbiologic surveillance, or other factors account for the increased incidence of Nocardia in our patients with bronchiectasis.

Direct-Acting Antivirals Improve Access to Care and Cure for Patients With HIV and Chronic HCV Infection

Background

Direct-acting antivirals (DAA) as curative therapy for hepatitis C virus (HCV) infection offer >95% sustained virologic response (SVR), including in patients with human immunodeficiency virus (HIV) infection. Despite improved safety and efficacy of HCV treatment, challenges remain, including drug-drug interactions between DAA and antiretroviral therapy (ART) and restrictions on access by payers.

Methods

We performed a retrospective cohort study of all HIV/HCV co-infected and HCV mono-infected patients captured in care at our institution from 2011-2015, reflecting the DAA era, to determine treatment uptake and SVR, and to elucidate barriers to accessing DAA for co-infected patients.

Results

We identified 9290 patients with HCV mono-infection and 507 with HIV/HCV co-infection. Compared to mono-infected patients, co-infected patients were younger and more likely to be male and African-American. For both groups, treatment uptake improved from the DAA/pegylated interferon (PEGIFN)-ribavirin to IFN-free DAA era. One-third of co-infected patients in the IFN-free DAA era required ART switch and nearly all remained virologically suppressed after 6 months. We observed SVR >95% for most patient subgroups including those with co-infection, prior treatment-experience, and cirrhosis. Predictors of access to DAA for co-infected patients included Caucasian race, CD4 count ≥200 cells/mm³, HIV virologic suppression and cirrhosis. Time to approval of DAA was longest for patients insured by Medicaid, followed by private insurance and Medicare.

Conclusions

DAA therapy has significantly improved access to HCV treatment and high SVR is independent of HIV status. However, in order to realize cure for all, barriers and disparities in access need to be urgently addressed.

Incidence, Long-Term Outcomes, and Healthcare Utilization of Patients With Human Immunodeficiency Virus/Acquired Immune Deficiency Syndrome and Disseminated Mycobacterium avium Complex From 1992-2015

Background

Despite the advent of combination antiretroviral therapy (cART), patients with human immunodeficiency virus (HIV) continue to develop late-stage complications including acquired immune deficiency syndrome (AIDS), disseminated Mycobacterium avium complex (DMAC), and death.

Methods

We performed an observational retrospective cohort study of HIV-infected adults who developed DMAC in the Duke University Health System from 1992 to 2015 to determine the incidence, long-term outcomes, and healthcare utilization of this population at high risk for poor outcomes. Findings were stratified by the “pre-cART” era (before January 1, 1996) and “post-cART” thereafter.

Results

We identified 330 adult HIV-infected patients newly diagnosed with DMAC, the majority (75.2%) of whom were male and non-Hispanic black (69.1%), with median age of 37 years. Incidence of DMAC declined significantly from 65.3/1000 in 1992 to 2.0/1000 in 2015, and the proportion of females and non-Hispanic blacks was significantly higher in the post-cART era. The standardized mortality ratios for DMAC patients who received cART were 69, 58, 27, 5.9, and 6.8 at years 1–5, respectively, after DMAC diagnosis. For patients diagnosed with DMAC in 2000 or later (n = 135), 20% were newly diagnosed with HIV in the 3 months preceding presentation with DMAC. Those with established HIV had a median time from HIV diagnosis to DMAC diagnosis of 7 years and were more likely to be black, rehospitalized in the 6 months after DMAC diagnosis, and die in the long term.

Conclusions

Disseminated Mycobacterium avium complex continues to be a lethal diagnosis in the cART era, disproportionately afflicts minority populations, and reflects both delayed entry into care and failure to consistently engage care.

Clinical code set engineering for reusing EHR data for research: A review

INTRODUCTION

The construction of reliable, reusable clinical code sets is essential when re-using Electronic Health Record (EHR) data for research. Yet code set definitions are rarely transparent and their sharing is almost non-existent. There is a lack of methodological standards for the management (construction, sharing, revision and reuse) of clinical code sets which needs to be addressed to ensure the reliability and credibility of studies which use code sets.

OBJECTIVE

To review methodological literature on the management of sets of clinical codes used in research on clinical databases and to provide a list of best practice recommendations for future studies and software tools.

METHODS

We performed an exhaustive search for methodological papers about clinical code set engineering for re-using EHR data in research. This was supplemented with papers identified by snowball sampling. In addition, a list of e-phenotyping systems was constructed by merging references from several systematic reviews on this topic, and the processes adopted by those systems for code set management was reviewed.

RESULTS

Thirty methodological papers were reviewed. Common approaches included: creating an initial list of synonyms for the condition of interest (n=20); making use of the hierarchical nature of coding terminologies during searching (n=23); reviewing sets with clinician input (n=20); and reusing and updating an existing code set (n=20). Several open source software tools (n=3) were discovered.

DISCUSSION

There is a need for software tools that enable users to easily and quickly create, revise, extend, review and share code sets and we provide a list of recommendations for their design and implementation.

CONCLUSION

Research re-using EHR data could be improved through the further development, more widespread use and routine reporting of the methods by which clinical codes were selected.

Consensus-Driven Development of a Terminology for Biobanking, the Duke Experience

Biobanking at Duke University has existed for decades and has grown over time in silos and based on specialized needs, as is true with most biomedical research centers. These silos developed informatics systems to support their own individual requirements, with no regard for semantic or syntactic interoperability. Duke undertook an initiative to implement an enterprise-wide biobanking information system to serve its many diverse biobanking entities. A significant part of this initiative was the development of a common terminology for use in the commercial software platform. Common terminology provides the foundation for interoperability across biobanks for data and information sharing. We engaged experts in research, informatics, and biobanking through a consensus-driven process to agree on 361 terms and their definitions that encompass the lifecycle of a biospecimen. Existing standards, common terms, and data elements from published articles provided a foundation on which to build the biobanking terminology; a broader set of stakeholders then provided additional input and feedback in a secondary vetting process. The resulting standardized biobanking terminology is now available for sharing with the biobanking community to serve as a foundation for other institutions who are considering a similar initiative.