Using aggregated, de-identified electronic health record data for multivariate pharmacosurveillance: a case study of azathioprine

Clinical Research Informatics: a Decade-in-Review

BACKGROUND

Clinical Research Informatics (CRI) is a subspeciality of biomedical informatics that has substantially matured during the last decade. Advances in CRI have transformed the way clinical research is conducted. In recent years, there has been growing interest in CRI, as reflected by a vast and expanding scientific literature focused on the topic. The main objectives of this review are: 1) to provide an overview of the evolving definition and scope of this biomedical informatics subspecialty over the past 10 years; 2) to highlight major contributions to the field during the past decade; and 3) to provide insights about more recent CRI research trends and perspectives.

METHODS

We adopted a modified thematic review approach focused on understanding the evolution and current status of the CRI field based on literature sources identified through two complementary review processes (AMIA CRI year-in-review/IMIA Yearbook of Medical Informatics) conducted annually during the last decade.

RESULTS

More than 1,500 potentially relevant publications were considered, and 205 sources were included in the final review. The review identified key publications defining the scope of CRI and/or capturing its evolution over time as illustrated by impactful tools and methods in different categories of CRI focus. The review also revealed current topics of interest in CRI and prevailing research trends.

CONCLUSION

This scoping review provides an overview of a decade of research in CRI, highlighting major changes in the core CRI discoveries as well as increasingly impactful methods and tools that have bridged the principles-to-practice gap. Practical CRI solutions as well as examples of CRI-enabled large-scale, multi-organizational and/or multi-national research projects demonstrate the maturity of the field. Despite the progress demonstrated, some topics remain challenging, highlighting the need for ongoing CRI development and research, including the need of more rigorous evaluations of CRI solutions and further formalization and maturation of CRI services and capabilities across the research enterprise.

Incidence of Kidney Stones After Bariatric Surgeries: Comparing Roux-en-Y Gastric Bypass and Sleeve Gastrectomy

INTRODUCTION/PURPOSE

The two most common procedures performed to treat obesity are Roux-En-Y gastric bypass (RNYGB) and laparoscopic sleeve gastrectomy (LSG). Due to changes in enteric absorption, bariatric surgery increases rates of nephrolithiasis. As population-based data are limited, we aimed to compare the incidence of kidney stones after RNYGB and LSG.

MATERIALS AND METHODS

We queried Explorys (Cleveland, OH), a database that aggregated data from 26 healthcare systems. We identified patients who were newly diagnosed with nephrolithiasis 3, 6, and 12 months after their RNYGB or LSG. Additionally, a multivariate analysis was conducted to investigate the association of nephrolithiasis with RNYGB as compared to LSG. This analysis adjusted for other risk factors, including age above 65, male gender, Caucasian race, diabetes mellitus, hypertension, primary hyperparathyroidism, gout, and obesity.

RESULTS

From 1999 to 2019, there were 11,480 patients who underwent RNYGB and 22,770 patients who underwent LSG. The incidence of nephrolithiasis in the RNYGB cohort at all three time points was higher than in the LSG cohort (3 months, 7.1% vs. 2.4%; 6 months, 6.6% vs. 2.0%; 1 year, 5.8% vs. 1.4%; P < 0.001). After the multivariate analysis, it was found that, though both RNYGB and LSG were independently associated with the development of nephrolithiasis, the risk of nephrolithiasis was higher in those who underwent RNYGB compared to those who underwent LSG (OR 1.594, 95% CI 1.494 to 1.701, P < 0.001).

CONCLUSION

RNYGB is associated with a higher risk of nephrolithiasis when compared to LSG.

Use of Electronic Health Record Data for Drug Safety Signal Identification: A Scoping Review

INTRODUCTION

Pharmacovigilance programs protect patient health and safety by identifying adverse event signals through postmarketing surveillance of claims data and spontaneous reports. Electronic health records (EHRs) provide new opportunities to address limitations of traditional approaches and promote discovery-oriented pharmacovigilance.

METHODS

To evaluate the current state of EHR-based medication safety signal identification, we conducted a scoping literature review of studies aimed at identifying safety signals from routinely collected patient-level EHR data. We extracted information on study design, EHR data elements utilized, analytic methods employed, drugs and outcomes evaluated, and key statistical and data analysis choices.

RESULTS

We identified 81 eligible studies. Disproportionality methods were the predominant analytic approach, followed by data mining and regression. Variability in study design makes direct comparisons difficult. Studies varied widely in terms of data, confounding adjustment, and statistical considerations.

CONCLUSION

Despite broad interest in utilizing EHRs for safety signal identification, current efforts fail to leverage the full breadth and depth of available data or to rigorously control for confounding. The development of best practices and application of common data models would promote the expansion of EHR-based pharmacovigilance.

Digital health: trends, opportunities and challenges in medical devices, pharma and bio-technology

Digital health interventions refer to the use of digital technology and connected devices to improve health outcomes and healthcare delivery. This includes telemedicine, electronic health records, wearable devices, mobile health applications, and other forms of digital health technology. To this end, several research and developmental activities in various fields are gaining momentum. For instance, in the medical devices sector, several smart biomedical materials and medical devices that are digitally enabled are rapidly being developed and introduced into clinical settings. In the pharma and allied sectors, digital health-focused technologies are widely being used through various stages of drug development, viz. computer-aided drug design, computational modeling for predictive toxicology, and big data analytics for clinical trial management. In the biotechnology and bioengineering fields, investigations are rapidly growing focus on digital health, such as omics biology, synthetic biology, systems biology, big data and personalized medicine. Though digital health-focused innovations are expanding the horizons of health in diverse ways, here the development in the fields of medical devices, pharmaceutical technologies and biotech sectors, with emphasis on trends, opportunities and challenges are reviewed. A perspective on the use of digital health in the Indian context is also included.

Mining comorbidities of opioid use disorder from FDA adverse event reporting system and patient electronic health records

BACKGROUND

Opioid use disorder (OUD) has become an urgent health problem. People with OUD often experience comorbid medical conditions. Systematical approaches to identifying co-occurring conditions of OUD can facilitate a deeper understanding of OUD mechanisms and drug discovery. This study presents an integrated approach combining data mining, network construction and ranking, and hypothesis-driven case-control studies using patient electronic health records (EHRs).

METHODS

First, we mined comorbidities from the US Food and Drug Administration Adverse Event Reporting System (FAERS) of 12 million unique case reports using frequent pattern-growth algorithm. The performance of OUD comorbidity mining was measured by precision and recall using manually curated known OUD comorbidities. We then constructed a disease comorbidity network using mined association rules and further prioritized OUD comorbidities. Last, novel OUD comorbidities were independently tested using EHRs of 75 million unique patients.

RESULTS

The OUD comorbidities from association rules mining achieves a precision of 38.7% and a recall of 78.2 Based on the mined rules, the global DCN was constructed with 1916 nodes and 32,175 edges. The network-based OUD ranking result shows that 43 of 55 known OUD comorbidities were in the first decile with a precision of 78.2%. Hypothyroidism and type 2 diabetes were two top-ranked novel OUD comorbidities identified by data mining and network ranking algorithms. Based on EHR-based case-control studies, we showed that patients with OUD had significantly increased risk for hyperthyroidism (AOR = 1.46, 95% CI 1.43-1.49, p value < 0.001), hypothyroidism (AOR = 1.45, 95% CI 1.42-1.48, p value < 0.001), type 2-diabetes (AOR = 1.28, 95% CI 1.26-1.29, p value < 0.001), compared with individuals without OUD.

CONCLUSION

Our study developed an integrated approach for identifying and validating novel OUD comorbidities from health records of 87 million unique patients (12 million for discovery and 75 million for validation), which can offer new opportunities for OUD mechanism understanding, drug discovery, and multi-component service delivery for co-occurring medical conditions among patients with OUD.

Combining phenome-driven drug-target interaction prediction with patients' electronic health records-based clinical corroboration toward drug discovery

Motivation

Predicting drug–target interactions (DTIs) using human phenotypic data have the potential in eliminating the translational gap between animal experiments and clinical outcomes in humans. One challenge in human phenome-driven DTI predictions is integrating and modeling diverse drug and disease phenotypic relationships. Leveraging large amounts of clinical observed phenotypes of drugs and diseases and electronic health records (EHRs) of 72 million patients, we developed a novel integrated computational drug discovery approach by seamlessly combining DTI prediction and clinical corroboration.

Results

We developed a network-based DTI prediction system (TargetPredict) by modeling 855 904 phenotypic and genetic relationships among 1430 drugs, 4251 side effects, 1059 diseases and 17 860 genes. We systematically evaluated TargetPredict in de novo cross-validation and compared it to a state-of-the-art phenome-driven DTI prediction approach. We applied TargetPredict in identifying novel repositioned candidate drugs for Alzheimer’s disease (AD), a disease affecting over 5.8 million people in the United States. We evaluated the clinical efficiency of top repositioned drug candidates using EHRs of over 72 million patients. The area under the receiver operating characteristic (ROC) curve was 0.97 in the de novo cross-validation when evaluated using 910 drugs. TargetPredict outperformed a state-of-the-art phenome-driven DTI prediction system as measured by precision–recall curves [measured by average precision (MAP): 0.28 versus 0.23, P-value < 0.0001]. The EHR-based case–control studies identified that the prescriptions top-ranked repositioned drugs are significantly associated with lower odds of AD diagnosis. For example, we showed that the prescription of liraglutide, a type 2 diabetes drug, is significantly associated with decreased risk of AD diagnosis [adjusted odds ratios (AORs): 0.76; 95% confidence intervals (CI) (0.70, 0.82), P-value < 0.0001]. In summary, our integrated approach that seamlessly combines computational DTI prediction and large-scale patients’ EHRs-based clinical corroboration has high potential in rapidly identifying novel drug targets and drug candidates for complex diseases.

Availability and implementation

nlp.case.edu/public/data/TargetPredict.

COVID-19 risk and outcomes in patients with substance use disorders: analyses from electronic health records in the United States

The global pandemic of COVID-19 is colliding with the epidemic of opioid use disorders (OUD) and other substance use disorders (SUD) in the United States (US). Currently, there is limited data on risks, disparity, and outcomes for COVID-19 in individuals suffering from SUD. This is a retrospective case-control study of electronic health records (EHRs) data of 73,099,850 unique patients, of whom 12,030 had a diagnosis of COVID-19. Patients with a recent diagnosis of SUD (within past year) were at significantly increased risk for COVID-19 (adjusted odds ratio or AOR = 8.699 [8.411-8.997], P < 10-30), an effect that was strongest for individuals with OUD (AOR = 10.244 [9.107-11.524], P < 10-30), followed by individuals with tobacco use disorder (TUD) (AOR = 8.222 ([7.925-8.530], P < 10-30). Compared to patients without SUD, patients with SUD had significantly higher prevalence of chronic kidney, liver, lung diseases, cardiovascular diseases, type 2 diabetes, obesity and cancer. Among patients with recent diagnosis of SUD, African Americans had significantly higher risk of COVID-19 than Caucasians (AOR = 2.173 [2.01-2.349], P < 10-30), with strongest effect for OUD (AOR = 4.162 [3.13-5.533], P < 10-25). COVID-19 patients with SUD had significantly worse outcomes (death: 9.6%, hospitalization: 41.0%) than general COVID-19 patients (death: 6.6%, hospitalization: 30.1%) and African Americans with COVID-19 and SUD had worse outcomes (death: 13.0%, hospitalization: 50.7%) than Caucasians (death: 8.6%, hospitalization: 35.2%). These findings identify individuals with SUD, especially individuals with OUD and African Americans, as having increased risk for COVID-19 and its adverse outcomes, highlighting the need to screen and treat individuals with SUD as part of the strategy to control the pandemic while ensuring no disparities in access to healthcare support.

The prevalence of hypertension in multiple sclerosis based on 37 million electronic health records from the United States

BACKGROUND AND PURPOSE

Hypertension (HTN) is a common comorbidity in multiple sclerosis (MS), and it significantly contributes to adverse outcomes. Unfortunately, the distribution of HTN in persons with MS has not been well characterized, and prior estimates have primarily relied on modest sample sizes. The objective of this study was to robustly describe the distribution of HTN in the MS population in comparison to the non-MS population with considerations for age, sex, and race. To date, this is the largest investigation of its kind.

METHODS

We conducted a cross-sectional study of 37 million unique electronic health records available in the IBM Explorys Enterprise Performance Management: Explore database (Explorys) spanning the United States. This resource has previously been validated for use in MS. We evaluated the prevalence of HTN in MS (N = 122 660) and non-MS (N = 37 075 350) cohorts, stratifying by age, sex, and race.

RESULTS

The prevalence of HTN was significantly greater among those with MS than among those without MS across age, sex, and race subpopulations, even after adjusting for age and sex. HTN was 25% more common in MS. In both MS and non-MS cohorts, the prevalence of HTN progressively increased with age and was higher in Black Americans and in males.

DISCUSSION

This study demonstrated that HTN is significantly more common in the MS population compared to the non-MS population, irrespective of sex and race. Because HTN is the leading global risk factor for disability and death, these results emphasize the need for aggressive screening for, and management of, HTN in the MS population.

Tumor Necrosis Factor (TNF) blocking agents are associated with lower risk for Alzheimer's disease in patients with rheumatoid arthritis and psoriasis

This large, retrospective case-control study of electronic health records from 56 million unique adult patients examined whether or not treatment with a Tumor Necrosis Factor (TNF) blocking agent is associated with lower risk for Alzheimer’s disease (AD) in patients with rheumatoid arthritis (RA), psoriasis, and other inflammatory diseases which are mediated in part by TNF and for which a TNF blocker is an approved treatment. The analysis compared the diagnosis of AD as an outcome measure in patients receiving at least one prescription for a TNF blocking agent (etanercept, adalimumab, and infliximab) or for methotrexate. Adjusted odds ratios (AORs) were estimated using the Cochran-Mantel-Haenszel (CMH) method and presented with 95% confidence intervals (CIs) and p-values. RA was associated with a higher risk for AD (Adjusted Odds Ratio (AOR) = 2.06, 95% Confidence Interval: (2.02–2.10), P-value <0.0001) as did psoriasis (AOR = 1.37 (1.31–1.42), P <0.0001), ankylosing spondylitis (AOR = 1.57 (1.39–1.77), P <0.0001), inflammatory bowel disease (AOR = 2.46 (2.33–2.59), P < 0.0001), ulcerative colitis (AOR = 1.82 (1.74–1.91), P <0.0001), and Crohn’s disease (AOR = 2.33 (2.22–2.43), P <0.0001). The risk for AD in patients with RA was lower among patients treated with etanercept (AOR = 0.34 (0.25–0.47), P <0.0001), adalimumab (AOR = 0.28 (0.19–0.39), P < 0.0001), or infliximab (AOR = 0.52 (0.39–0.69), P <0.0001). Methotrexate was also associated with a lower risk for AD (AOR = 0.64 (0.61–0.68), P <0.0001), while lower risk was found in patients with a prescription history for both a TNF blocker and methotrexate. Etanercept and adalimumab also were associated with lower risk for AD in patients with psoriasis: AOR = 0.47 (0.30–0.73 and 0.41 (0.20–0.76), respectively. There was no effect of gender or race, while younger patients showed greater benefit from a TNF blocker than did older patients. This study identifies a subset of patients in whom systemic inflammation contributes to risk for AD through a pathological mechanism involving TNF and who therefore may benefit from treatment with a TNF blocking agent.

Racial Differences in Clinical Features and Comorbidities in Ankylosing Spondylitis in the United States

OBJECTIVE

To examine racial differences of clinical features, medication usage, and comorbidities of patients with ankylosing spondylitis (AS) in the United States.

METHODS

In the Explorys database, 28,520 patients with AS were identified. Data were stratified by 2 rheumatology visits, race, sex, clinical characteristics, medication use, and comorbidities. Datasets were recorded as proportions, which were compared using chi-square test (p < 0.05).

RESULTS

Of the 10,990 patients with AS, 8% were African Americans and had elevated erythrocyte sedimentation rate and C-reactive protein, and high frequency of anterior uveitis, hypertension, diabetes, depression, and heart disease.

CONCLUSION

African Americans with AS in the United States have high disease activity and comorbidities compared to whites.

Prevalence of asthma in multiple sclerosis: A United States population-based study

BACKGROUND

Multiple sclerosis (MS) and asthma are complex multifactorial diseases which adversely impact daily functioning. However, the prevalence of asthma in those with MS is not clear. The objective of this study is to characterize the prevalence of asthma in those with MS, with considerations for age, gender, and race.

METHODS

We conducted a U.S. population-based, cross-sectional study of electronic health record information for 56.6 million Americans available in the IBM® Explorys EPM: Explore database. We evaluated the prevalence of asthma in MS (N = 141,880) and non-MS (N = 56,416,790) cohorts, stratifying by age, gender, and race (All, White Americans, and African Americans).

RESULTS

The prevalence of asthma was significantly greater among those with MS than the general population across age, gender, and racial subpopulations. Adjusting for age and gender, asthma was three times more common in MS. In the MS cohort, the prevalence of asthma had a U-shaped distribution with respect to age, with the greatest asthma prevalence among the young and the elderly (> 20% prevalence among those <30 or ≥80 years; prevalence range: 15 to 30%); this significantly differed from the fairly uniform distribution observed in the non-MS cohort (prevalence range: 4 to 9%). These patterns were relatively consistent when stratifying by gender and race.

CONCLUSION

Asthma is significantly more common in those with MS than in the general population - particularly in the young and elderly - irrespective of gender and race. The results add to the growing MS comorbidity literature, and emphasizes the need for comorbidity management as a part of comprehensive MS patient care.

New approaches to pharmacosurveillance for monitoring prescription frequency, diversity, and co-prescription in a large sentinel network of companion animal veterinary practices in the United Kingdom, 2014-2016

Pharmaceutical agents (PAs) are commonly prescribed in companion animal practice in the United Kingdom. However, little is known about PA prescription on a population-level, particularly with respect to PAs authorised for human use alone prescribed via the veterinary cascade; this raises important questions regarding the efficacy and safety of PAs prescribed to companion animals. This study explored new approaches for describing PA prescription, diversity and co-prescription in dogs, cats and rabbits utilising electronic health records (EHRs) from a sentinel network of 457 companion animal-treating veterinary sites throughout the UK over a 2-year period (2014-2016). A novel text mining-based identification and classification methodology was utilised to semi-automatically map practitioner-defined product descriptions recorded in 918,333 EHRs from 413,870 dogs encompassing 1,242,270 prescriptions; 352,730 EHRs from 200,541 cats encompassing 491,554 prescriptions, and 22,526 EHRS from 13,398 rabbits encompassing 18,490 prescriptions respectively. PA prescription as a percentage of booked consultations was 65.4% (95% confidence interval, CI, 64.6-66.3) in dogs; in cats it was 69.1% (95% CI, 67.9-70.2) and in rabbits, 56.3% (95% CI, 54.7-57.8). Vaccines were the most commonly prescribed PAs in all three species, with antibiotics, antimycotics, and parasiticides also commonly prescribed. PA prescription utilising products authorised for human use only (hence, 'human-authorised') comprised 5.1% (95% CI, 4.7-5.5) of total canine prescription events; in cats it was 2.8% (95% CI, 2.6-3.0), and in rabbits, 7.8% (95% CI, 6.5-9.0). The most commonly prescribed human-authorised PA in dogs was metronidazole (antibiotic); in cats and rabbits it was ranitidine (H2 histamine receptor antagonist). Using a new approach utilising the Simpson's Diversity Index (an ecological measure of relative animal, plant etc. species abundance), we identified differences in prescription based on presenting complaint and species, with rabbits generally exposed to a less diverse range of PAs than dogs or cats, potentially reflecting the paucity of authorised PAs for use in rabbits. Finally, through a novel application of network analysis, we demonstrated the existence of three major co-prescription groups (preventive health; treatment of disease, and euthanasia); a trend commonly observed in practice. This study represents the first time PA prescription has been described across all pharmaceutical families in a large population of companion animals, encompassing PAs authorised for both veterinary and human-only use. These data form a baseline against which future studies could be compared, and provides some useful tools for understanding PA comparative efficacy and risks when prescribed in the varied setting of clinical practice.

Applications of network analysis to routinely collected health care data: a systematic review

Objective

To survey network analyses of datasets collected in the course of routine operations in health care settings and identify driving questions, methods, needs, and potential for future research.

Materials and Methods

A search strategy was designed to find studies that applied network analysis to routinely collected health care datasets and was adapted to 3 bibliographic databases. The results were grouped according to a thematic analysis of their settings, objectives, data, and methods. Each group received a methodological synthesis.

Results

The search found 189 distinct studies reported before August 2016. We manually partitioned the sample into 4 groups, which investigated institutional exchange, physician collaboration, clinical co-occurrence, and workplace interaction networks. Several robust and ongoing research programs were discerned within (and sometimes across) the groups. Little interaction was observed between these programs, despite conceptual and methodological similarities.

Discussion

We use the literature sample to inform a discussion of good practice at this methodological interface, including the concordance of motivations, study design, data, and tools and the validation and standardization of techniques. We then highlight instances of positive feedback between methodological development and knowledge domains and assess the overall cohesion of the sample.

Real world big data for clinical research and drug development

The objective of this paper is to identify the extent to which real world data (RWD) is being utilized, or could be utilized, at scale in drug development. Through screening peer-reviewed literature, we have cited specific examples where RWD can be used for biomarker discovery or validation, gaining a new understanding of a disease or disease associations, discovering new markers for patient stratification and targeted therapies, new markers for identifying persons with a disease, and pharmacovigilance. None of the papers meeting our criteria was specifically geared toward novel targets or indications in the biopharmaceutical sector; the majority were focused on the area of public health, often sponsored by universities, insurance providers or in combination with public health bodies such as national insurers. The field is still in an early phase of practical application, and is being harnessed broadly where it serves the most direct need in public health applications in early, rare and novel disease incidents. However, these exemplars provide a valuable contribution to insights on the use of RWD to create novel, faster and less invasive approaches to advance disease understanding and biomarker discovery. We believe that pharma needs to invest in making better use of Electronic Health Records and the need for more precompetitive collaboration to grow the scale of this 'big denominator' capability, especially given the needs of precision medicine research.

Does pregnancy increase the risk of abdominal hernia recurrence after prepregnancy surgical repair?

BACKGROUND

By increasing intraabdominal pressure, pregnancy may increase the risk of abdominal hernia recurrence. Current data are limited to studies with small sample size and thus the impact of pregnancy on recurrence is unclear.

OBJECTIVE(S)

The objective of this analysis was to evaluate the impact of pregnancy on clinically significant abdominal hernia recurrence in a large multicenter cohort.

STUDY DESIGN

A multiinstitution deidentified electronic health record database, EPM: Explore (Explorys Inc, Cleveland, OH) was utilized to perform a retrospective cohort study of women aged 18-45 years with a history of an abdominal hernia repair from 1999 through 2013. Abdominal hernia was defined to include ventral and incisional hernias, and other types were excluded. The presence or absence of a pregnancy following primary hernia repair was elucidated from the database. Subjects were excluded if a hernia repair occurred during pregnancy. The rate of hernia recurrence, defined as reoperation, was calculated. The association between pregnancy and hernia recurrence was evaluated with logistic regression, both unadjusted and adjusted for diabetes, obesity (body mass index >30 kg/m(2)), tobacco abuse, and wound complication at the time of initial hernia repair.

RESULTS

A total of 11,020 women with a history of hernia repair were identified, of whom 840 had a subsequent pregnancy. Overall, 915 women in the cohort had a hernia recurrence (8.3%). Women with a history of pregnancy following primary hernia repair were more likely to have a body mass index >30 kg/m(2), a history of tobacco abuse, and a wound complication at the time of primary repair. In an unadjusted analysis, pregnancy was associated with an increase in the risk of hernia recurrence (13.1% vs 7.1%, odds ratio, 1.96, 95% confidence interval, 1.60-2.42). The association between pregnancy and hernia recurrence was attenuated but persisted after adjusting for confounding factors (adjusted odds ratio, 1.73, 95% confidence interval, 1.40-2.14).

CONCLUSION

Pregnancy is associated with an increased risk of abdominal hernia recurrence after adjusting for confounding factors. The magnitude of this association is likely underestimated, given that the risk of recurrence was defined as reoperation, which captures only the most clinically significant group of recurrences. This information will facilitate counseling for reproductive-aged women planning elective ventral or incisional hernia repair. The risk of recurrence and subsequent reoperation should be balanced against the risk of incarceration and emergent surgery during pregnancy. As such, the desire for future pregnancy and/or contraception should be considered when planning asymptomatic hernia repair for women of reproductive age.

Postoperative wound dehiscence: Predictors and associations

The Agency for Healthcare Research and Quality patient safety indicators (PSI) were developed as a metric of hospital complication rates. PSI-14 measures postoperative wound dehiscence and specifically how often a surgical wound in the abdominal or pelvic area fails to heal after abdominopelvic surgery. Wound dehiscence is estimated to occur in 0.5-3.4% of abdominopelvic surgeries, and carries a mortality of up to 40%. Postoperative wound dehiscence has been adopted as a surrogate safety outcome measure as it impacts morbidity, length of stay, healthcare costs and readmission rates. Postoperative wound dehiscence cases from the Nationwide Inpatient Sample demonstrate 9.6% excess mortality, 9.4 days of excess hospitalization and $40,323 in excess hospital charges relative to matched controls. The purpose of the current study was to investigate the associations between PSI-14 and measurable medical and surgical comorbidities using the Explorys technology platform to query electronic health record data from a large hospital system serving a diverse patient population in the Washington, DC and Baltimore, MD metropolitan areas. The study population included 25,636 eligible patients who had undergone abdominopelvic surgery between January 1, 2008 and December 31, 2012. Of these cases, 786 (2.97%) had postoperative wound dehiscence. Patient-associated comorbidities were strongly associated with PSI-14, suggesting that this indicator may not solely be an indicator of hospital safety. There was a strong association between PSI-14 and opioid use after surgery and this finding merits further investigation.

Electronic Health Record-Enabled Research in Children Using the Electronic Health Record for Clinical Discovery

Initially described more than 50 years ago, electronic health records (EHRs) are now becoming ubiquitous throughout pediatric health care settings. The confluence of increased EHR implementation and the exponential growth of digital data within them, the development of clinical informatics tools and techniques, and the growing workforce of experienced EHR users presents new opportunities to use EHRs to augment clinical discovery and improve pediatric patient care. This article reviews the basic concepts surrounding EHR-enabled research and clinical discovery, including the types and fidelity of EHR data elements, EHR data validation/corroboration, and the steps involved in analytical interrogation.

Desiderata for computable representations of electronic health records-driven phenotype algorithms

BACKGROUND

Electronic health records (EHRs) are increasingly used for clinical and translational research through the creation of phenotype algorithms. Currently, phenotype algorithms are most commonly represented as noncomputable descriptive documents and knowledge artifacts that detail the protocols for querying diagnoses, symptoms, procedures, medications, and/or text-driven medical concepts, and are primarily meant for human comprehension. We present desiderata for developing a computable phenotype representation model (PheRM).

METHODS

A team of clinicians and informaticians reviewed common features for multisite phenotype algorithms published in PheKB.org and existing phenotype representation platforms. We also evaluated well-known diagnostic criteria and clinical decision-making guidelines to encompass a broader category of algorithms.

RESULTS

We propose 10 desired characteristics for a flexible, computable PheRM: (1) structure clinical data into queryable forms; (2) recommend use of a common data model, but also support customization for the variability and availability of EHR data among sites; (3) support both human-readable and computable representations of phenotype algorithms; (4) implement set operations and relational algebra for modeling phenotype algorithms; (5) represent phenotype criteria with structured rules; (6) support defining temporal relations between events; (7) use standardized terminologies and ontologies, and facilitate reuse of value sets; (8) define representations for text searching and natural language processing; (9) provide interfaces for external software algorithms; and (10) maintain backward compatibility.

CONCLUSION

A computable PheRM is needed for true phenotype portability and reliability across different EHR products and healthcare systems. These desiderata are a guide to inform the establishment and evolution of EHR phenotype algorithm authoring platforms and languages.

Information technology for clinical, translational and comparative effectiveness research. Findings from the section clinical research informatics

OBJECTIVE

To select and summarize key contributions to current research in the field of Clinical Research Informatics (CRI).

METHOD

A bibliographic search using a combination of MeSH and free terms search over PubMed was performed followed by a blinded review.

RESULTS

The review process resulted in the selection of four papers illustrating various aspects of current research efforts in the area of CRI. The first paper tackles the challenge of extracting accurate phenotypes from Electronic Healthcare Records (EHRs). Privacy protection within shared de-identified, patient-level research databases is the focus of the second selected paper. Two other papers exemplify the growing role of formal representation of clinical data - in metadata repositories - and knowledge - in ontologies - for supporting the process of reusing data for clinical research.

CONCLUSIONS

The selected articles demonstrate how concrete platforms are currently achieving interoperability across clinical research and care domains and have reached the evaluation phase. When EHRs linked to genetic data have the potential to shift the research focus from research driven patient recruitment to phenotyping in large population, a key issue is to lower patient re-identification risks for biomedical research databases. Current research illustrates the potential of knowledge engineering to support, in the coming years, the scientific lifecycle of clinical research.