Development of Phenotyping Algorithms for the Identification of Organ Transplant Recipients: Cohort Study

Abstract

Background Studies involving organ transplant recipients (OTRs) are often limited to the variables collected in the national Scientific Registry of Transplant Recipients database. Electronic health records contain additional variables that can augment this data source if OTRs can be identified accurately. Objective The aim of this study was to develop phenotyping algorithms to identify OTRs from electronic health records. Methods We used Vanderbilt’s deidentified version of its electronic health record database, which contains nearly 3 million subjects, to develop algorithms to identify OTRs. We identified all 19,817 individuals with at least one International Classification of Diseases (ICD) or Current Procedural Terminology (CPT) code for organ transplantation. We performed a chart review on 1350 randomly selected individuals to determine the transplant status. We constructed machine learning models to calculate positive predictive values and sensitivity for combinations of codes by using classification and regression trees, random forest, and extreme gradient boosting algorithms. Results Of the 1350 reviewed patient charts, 827 were organ transplant recipients while 511 had no record of a transplant, and 12 were equivocal. Most patients with only 1 or 2 transplant codes did not have a transplant. The most common reasons for being labeled a nontransplant patient were the lack of data (229/511, 44.8%) or the patient being evaluated for an organ transplant (174/511, 34.1%). All 3 machine learning algorithms identified OTRs with overall >90% positive predictive value and >88% sensitivity. Conclusions Electronic health records linked to biobanks are increasingly used to conduct large-scale studies but have not been well-utilized in organ transplantation research. We present rigorously evaluated methods for phenotyping OTRs from electronic health records that will enable the use of the full spectrum of clinical data in transplant research. Using several different machine learning algorithms, we were able to identify transplant cases with high accuracy by using only ICD and CPT codes.

Differences in Skin Cancer Rates by Transplanted Organ Type and Patient Age After Organ Transplant in White Patients

“…Both our population and outcome definitions were based on International Classification of Diseases and Current Procedural Terminology codes in the electronic health record rather than patient transplant registries. We have validated both of these measures within this cohort, and others have shown that registration in the Organ Procurement Transplant Network and Scientific Registry of Transplant Recipients database poorly captures skin cancer counts . Because of the nature of our research database, we were unable to ascertain which patients had been seen by a dermatologist, either locally or outside of Vanderbilt University Medical Center.…”

“…The cohort of patients in this study has been previously described . This cohort study was approved by the Vanderbilt University Medical Center Institutional Review Board, which waived the requirement for written informed consent because only deidentified data were used.…”

Risks of Multiple Skin Cancers in Organ Transplant Recipients

“…The strengths of our study include the use of established EHR and claims data sets, distinct from more frequently used registries and transplant databases, and an externally validated approach to identify OTRs . We report similar overall risks of skin cancer (5%-13%) in OTRs to those reported by the Transplant Skin Cancer Network (8%) .…”

“…Inclusion criteria were the following: aged 18 years or older at data set entry and classification as an OTR, defined by at least 4 organ transplant diagnosis codes ( International Classification of Diseases [ ICD ], 9th edition or 10th edition; see eAppendix 1 in the Supplement for full list) on different dates. This definition has shown a positive predictive value of greater than 94% in an external validation . We excluded patients with HIV, defined as any ICD-9 or ICD-10 code for HIV.…”

Voriconazole Metabolism is Associated with the Number of Skin Cancers Per Patient

“…Currently, the Clinical Pharmacogenetics Implementation Consortium (CPIC) guidelines recommend that rapid and ultrarapid metabolizers using an antifungal that is not metabolized by CYP2C19 when antifungal treatment is necessary [ 7 ] . As we observed in our study, data from the UK Biobank showed that rapid and ultrarapid metabolizers for CYP2C19 make up roughly one in every three individuals across nearly all genetic populations, indicating that clinical genotyping is very likely to impact treatment decisions and risk profiles [ 21 ] . While there is hope that less voriconazole exposure among this high-risk group will lead to decreased skin cancer risks, dermatologists should be aware of the interaction with metabolizer phenotypes when making decisions on follow-up intervals and prophylaxis.…”