Validating use of diagnostic codes in Canadian administrative data for identification of adverse drug events.

AIMS: While diagnostic codes from administrative health data might be a valuable source to identify adverse drug events (ADEs), their ability to identify unintended harms remains unclear. We validated claims-based diagnosis codes for ADEs based on events identified in a prospective cohort study and assessed whether key attributes predicted their documentation in administrative data. METHODS: This was a retrospective analysis of 3 prospective cohorts in British Columbia, from 2008 to 2015 (n = 13 969). We linked prospectively identified ADEs to administrative insurance data to examine the sensitivity and specificity of different diagnostic code schemes. We used logistic regression to assess which key attributes (e.g., type of event, symptoms and culprit medications) were associated with better documentation of ADEs in administrative data. RESULTS: Among 1178 diagnosed events, the sensitivity of the diagnostic codes in administrative data ranged from 3.4 to 52.6%, depending on the database and codes used. We found that documentation was worse for certain types of ADEs (dose-related: odds ratio [OR]: 0.32, 95% confidence interval [CI]: 0.15, 0.69; nonadherence events (OR: 0.35, 95% CI: 0.20, 0.62), and better for those experiencing arrhythmias (OR: 4.19, 95% CI: 0.96, 18.28). CONCLUSION: ADEs were not well documented in administrative data. Alternative methods should be explored to capture ADEs for health research.

Cause of death coding in asthma.

BACKGROUND: While clinical coding is intended to be an objective and standardized practice, it is important to recognize that it is not entirely the case. The clinical and bureaucratic practices from event of death to a case being entered into a research dataset are important context for analysing and interpreting this data. Variation in practices can influence the accuracy of the final coded record in two different stages: the reporting of the death certificate, and the International Classification of Diseases (Version 10; ICD-10) coding of that certificate. METHODS: This study investigated 91,022 deaths recorded in the Scottish Asthma Learning Healthcare System dataset between 2000 and 2017. Asthma-related deaths were identified by the presence of any of ICD-10 codes J45 or J46, in any position. These codes were categorized either as relating to asthma attacks specifically (status asthmatic; J46) or generally to asthma diagnosis (J45). RESULTS: We found that one in every 200 deaths in this were coded as being asthma related. Less than 1% of asthma-related mortality records used both J45 and J46 ICD-10 codes as causes. Infection (predominantly pneumonia) was more commonly reported as a contributing cause of death when J45 was the primary coded cause, compared to J46, which specifically denotes asthma attacks. CONCLUSION: Further inspection of patient history can be essential to validate deaths recorded as caused by asthma, and to identify potentially mis-recorded non-asthma deaths, particularly in those with complex comorbidities.

Model correction of diagnostic coding-based RSV incidence for children 0-4 years in the US.

BACKGROUND: Although administrative claims data have a high degree of completeness, not all medically attended Respiratory Syncytial Virus-associated lower respiratory tract infections (RSV-LRTIs) are tested or coded for their causative agent. We sought to determine the attribution of RSV to LRTI in claims data via modeling of temporal changes in LRTI rates against surveillance data. METHODS: We estimated the weekly incidence of LRTI (inpatient, outpatient, and total) for children 0-4 years using 2011-2019 commercial insurance claims, stratified by HHS region, matched to the corresponding weekly NREVSS RSV and influenza positivity data for each region, and modelled against RSV, influenza positivity rates, and harmonic functions of time assuming negative binomial distribution. LRTI events attributable to RSV were estimated as predicted events from the full model minus predicted events with RSV positivity rate set to 0. RESULTS: Approximately 42% of predicted RSV cases were coded in claims data. Across all regions, the percentage of LRTI attributable to RSV were 15-43%, 10-31%, and 10-31% of inpatient, outpatient, and combined settings, respectively. However, when compared to coded inpatient RSV-LRTI, 9 of 10 regions had improbable corrected inpatient LRTI estimates (predicted RSV/coded RSV ratio < 1). Sensitivity analysis based on separate models for PCR and antigen-based positivity showed similar results. CONCLUSIONS: Underestimation based on coding in claims data may be addressed by NREVSS-based adjustment of claims-based RSV incidence. However, where setting-specific positivity rates is unavailable, we recommend modeling across settings to mirror NREVSS's positivity rates which are similarly aggregated, to avoid inaccurate adjustments.

Creating a computer assisted ICD coding system: Performance metric choice and use of the ICD hierarchy.

OBJECTIVE: Machine learning methods hold the promise of leveraging available data and generating higher-quality data while alleviating the data collection burden on healthcare professionals. International Classification of Diseases (ICD) diagnoses data, collected globally for billing and epidemiological purposes, represents a valuable source of structured information. However, ICD coding is a challenging task. While numerous previous studies reported promising results in automatic ICD classification, they often describe input data specific model architectures, that are heterogeneously evaluated with different performance metrics and ICD code subsets. This study aims to explore the evaluation and construction of more effective Computer Assisted Coding (CAC) systems using generic approaches, focusing on the use of ICD hierarchy, medication data and a feed forward neural network architecture. METHODS: We conduct comprehensive experiments using the MIMIC-III clinical database, mapped to the OMOP data model. Our evaluations encompass various performance metrics, alongside investigations into multitask, hierarchical, and imbalanced learning for neural networks. RESULTS: We introduce a novel metric, , tailored to the ICD coding task, which offers interpretable insights for healthcare informatics practitioners, aiding them in assessing the quality of assisted coding systems. Our findings highlight that selectively cherry-picking ICD codes diminish retrieval performance without performance improvement over the selected subset. We show that optimizing for metrics such as NDCG and AUPRC outperforms traditional F1-based metrics in ranking performance. We observe that Neural Network training on different ICD levels simultaneously offers minor benefits for ranking and significant runtime gains. However, our models do not derive benefits from hierarchical or class imbalance correction techniques for ICD code retrieval. CONCLUSION: This study offers valuable insights for researchers and healthcare practitioners interested in developing and evaluating CAC systems. Using a straightforward sequential neural network model, we confirm that medical prescriptions are a rich data source for CAC systems, providing competitive retrieval capabilities for a fraction of the computational load compared to text-based models. Our study underscores the importance of metric selection and challenges existing practices related to ICD code sub-setting for model training and evaluation.

Undercounting diagnoses in Australian general practice: a data quality study with implications for population health reporting.

BACKGROUND: Diagnosis can often be recorded in electronic medical records (EMRs) as free-text or using a term with a diagnosis code. Researchers, governments, and agencies, including organisations that deliver incentivised primary care quality improvement programs, frequently utilise coded data only and often ignore free-text entries. Diagnosis data are reported for population healthcare planning including resource allocation for patient care. This study sought to determine if diagnosis counts based on coded diagnosis data only, led to under-reporting of disease prevalence and if so, to what extent for six common or important chronic diseases. METHODS: This cross-sectional data quality study used de-identified EMR data from 84 general practices in Victoria, Australia. Data represented 456,125 patients who attended one of the general practices three or more times in two years between January 2021 and December 2022. We reviewed the percentage and proportional difference between patient counts of coded diagnosis entries alone and patient counts of clinically validated free-text entries for asthma, chronic kidney disease, chronic obstructive pulmonary disease, dementia, type 1 diabetes and type 2 diabetes. RESULTS: Undercounts were evident in all six diagnoses when using coded diagnoses alone (2.57-36.72% undercount), of these, five were statistically significant. Overall, 26.4% of all patient diagnoses had not been coded. There was high variation between practices in recording of coded diagnoses, but coding for type 2 diabetes was well captured by most practices. CONCLUSION: In Australia clinical decision support and the reporting of aggregated patient diagnosis data to government that relies on coded diagnoses can lead to significant underreporting of diagnoses compared to counts that also incorporate clinically validated free-text diagnoses. Diagnosis underreporting can impact on population health, healthcare planning, resource allocation, and patient care. We propose the use of phenotypes derived from clinically validated text entries to enhance the accuracy of diagnosis and disease reporting. There are existing technologies and collaborations from which to build trusted mechanisms to provide greater reliability of general practice EMR data used for secondary purposes.

We Are Not Speaking the Same Language: Current Procedural Terminology Coding and Provision of Care in Lymphatic Reconstruction.

INTRODUCTION: Current Procedural Terminology (CPT) codes provide a uniform language for medical billing, but specific codes have not been assigned for lymphatic reconstruction techniques. The authors hypothesized that inadequate codes would contribute to heterogeneous coding practices and reimbursement challenges, ultimately limiting surgeons' ability to treat patients. METHODS: A 22-item virtual questionnaire was offered to 959 members of the American Society of Reconstructive Microsurgeons to assess the volume of lymphatic reconstruction procedures performed, CPT codes used for each procedure, and challenges related to coding and providing care. RESULTS: The survey was completed by 66 board-certified/board-eligible plastic surgeons (6.9%), who unanimously agreed that lymphatic surgery is integral to cancer care, with 86.4% indicating that immediate lymphatic reconstruction should be offered after lymphadenectomy. Most performed lymphovenous bypass, immediate lymphatic reconstruction, liposuction, and vascularized lymph node transfer.Respondents reported that available CPT codes failed to reflect procedural scope. A wide variety of CPT codes was used to report each type of procedure. Insurance coverage problems led to 69.7% of respondents forgoing operations and 32% reducing treatment offerings. Insurance coverage and CPT codes were identified as significant barriers to care by 98.5% and 95.5% of respondents, respectively. CONCLUSIONS: Respondents unanimously agreed on the importance of lymphatic reconstruction in cancer care, and most identified inadequate CPT codes as causing billing issues, which hindered their ability to offer surgical treatment. Appropriate and specific CPT codes are necessary to ensure accuracy and consistency of reporting and ultimately to improve patient access to care.

Observational Behavioral Coding in the Pediatric Emergency Department: Development of the Emergency Department Child Behavior Coding System.

BACKGROUND: Despite improvements over the past decade, children continue to experience significant pain and distress surrounding invasive procedures in the emergency department (ED). To assess the impact of newly developed interventions, we must create more reliable and valid behavioral assessment tools that have been validated for the unique settings of pediatric EDs. OBJECTIVE: This study aimed to create and test the Emergency Department Child Behavior Coding System (ED-CBCS) for the assessment of child distress and nondistress behaviors surrounding pediatric ED procedures. METHODS: Via an iterative process, a multidisciplinary expert panel developed the ED-CBCS, an advanced time-based behavioral coding measure. Inter-rater reliability and concurrent validity were examined using 38 videos of children aged from 2 to 12 years undergoing laceration procedures. Face, Legs, Activity, Cry, Consolability (FLACC) scale scores were used to examine concurrent validity. RESULTS: The final ED-CBCS included 27 child distress and nondistress behaviors. Time-unit κ values from 0.64 to 0.98 and event alignment κ values from 0.62 to 1.00 indicated good to excellent inter-rater reliability for all but one of the individual codes. ED-CBCS distress (B = 1.26; p < 0.001) and nondistress behaviors (B = -0.69, p = 0.025) were independently significantly associated with FLACC scores, indicating concurrent validity. CONCLUSIONS: We developed a psychometrically sound tool tailored for pediatric ED procedures. Future work could use this measure to better identify behavioral targets and test the effects of interventions to relieve pediatric ED pain and distress.

Toward Efficient, Sustainable, and Scalable Methods of Treatment Characterization: An Investigation of Coding Clinical Practice from Chart Notes.

PURPOSE: Chart notes provide a low-cost data source that could help characterize what occurs in treatment with sufficient precision to improve management of care. This study assessed the interrater reliability of treatment content coded from chart notes and evaluated its concordance with content coded from transcribed treatment sessions. METHOD: Fifty randomly selected and digitally recorded treatment events were transcribed and coded for practice content. Independent coders then applied the same code system to chart notes for these same treatment events. ANALYSIS: We measured reliability and concordance of practice occurrence and extensiveness at two levels of specificity: practices (full procedures) and steps (subcomponents of those procedures). RESULTS: For chart notes, practices had moderate interrater reliability (M k = 0.50, M ICC = 0.56) and steps had moderate (M ICC = 0.74) to substantial interrater reliability (M k = 0.78). On average, 2.54 practices and 5.64 steps were coded per chart note and 4.53 practices and 13.10 steps per transcript. Across sources, ratings for 64% of practices and 41% of steps correlated significantly, with those with significant correlations generally demonstrating moderate concordance (practice M r = 0.48; step M r = 0.47). Forty one percent of practices and 34% of steps from transcripts were also identified in the corresponding chart notes. CONCLUSION: Chart notes provide an accessible data source for evaluating treatment content, with different levels of specificity posing tradeoffs for validity and reliability, which in turn may have implications for chart note interfaces, training, and new metrics to support accurate, reliable, and efficient measurement of clinical practice.

[Evaluation of a coding guide on social determinants of health in primary care consultations: A mixed study]. / Evaluación de una guía de codificación de determinantes sociales de la salud en las consultas de atención primaria: Estudio mixto.

OBJECTIVE: To evaluate a coding guide for social determinants of health in primary care consultations as an effective tool in the professional's daily workflow. DESIGN: Mixed sequential explanatory study. Formed by a quantitative part (experimental) and a qualitative part (descriptive-evaluative). LOCATION: All the primary care teams of the Central Catalonia Management (32 teams). PARTICIPANTS AND SETTING: All nursing, social work and medical professionals working in the 32 primary care teams of the Catalan Institute of Health in Central Catalonia from February 2023 to July 2023. METHODS: A social determinants of health coding guide was developed. This guide was created in a multidisciplinary and multicenter manner. Purposive sampling. Quantitatively, the number of diagnoses recorded by the experimental group versus the control group was counted. Qualitatively, a thematic analysis was carried out from a socio-constructivist perspective. RESULTS: The results were significant and satisfactory. Using a quantitative methodology, the effectiveness of the use of the guide was assessed. A significant increase in the use of the social determinants was observed in the intervention group vs. the control group, with a percentage of post-intervention use of 19.53% in the control group and 32.26% in the intervention group (P < .001). The number of diagnoses recorded increased from 312 to 1322 in the intervention group, while it remained the same in the control group. The main factors identified through qualitative methodology that may explain the effectiveness of the guideline were: 1) the effectiveness of the guideline among primary care professionals, 2) the appropriateness of the guideline by assessing its usefulness and practicality, 3) feasibility and 4) specific contributions to the improvement of the guideline. CONCLUSIONS: The social determinants of health coding guide is effective, appropriate and can be implemented in the workflow of primary health care professionals for good recording of the social determinants of health.

Methodologic Approaches for Using Electronic Medical Records to Identify Experiences of Violence in Transgender and Cisgender People: Closing the Gap Between Diagnostic Coding and Lived Experiences.

BACKGROUND: Transgender people experience extreme rates of violence and the electronic medical record (EMR) remains a mostly untapped resource to study the medical sequelae of such experiences. OBJECTIVES: To develop and test a method for identifying experiences of violence using EMR data. RESEARCH DESIGN: Cross-sectional study utilizing EMR data. PEOPLE: Transgender and cisgender people seen at a regional referral center in Upstate New York. MEASURES: We tested the utility of keyword searches and structured data queries to identify specific types of violence at various ages and in various contexts among cohorts of transgender and cisgender people. We compared the effectiveness of keyword searches to diagnosis codes and a screening question, "Are you safe at home?" using McNemar's test. We compared the prevalence of various types of violence between transgender and cisgender cohorts using the χ 2 test of independence. RESULTS: Of the transgender cohort, 47% had experienced some type of violence versus 14% of the cisgender cohort (χ 2P value <0.001). Keywords were significantly more effective than structured data at identifying violence among both cohorts (McNemar P values all <0.05). CONCLUSIONS: Transgender people experience extreme amounts of violence throughout their lives, which is better identified and studied using keyword searches than structured EMR data. Policies are urgently needed to stop violence against transgender people. Interventions are also needed to ensure safe documentation of violence in EMRs to improve care across settings and aid research to develop and implement effective interventions.

DRCNNTLe: A deep recurrent convolutional neural network with transfer learning through pre-trained embeddings for automated ICD coding.

The International Classification of Diseases (ICD), which is endorsed by the World Health Organization, is a diagnostic classification standard. ICD codes store, retrieve, and analyze health information to make clinical decisions. Currently, ICD coding has been adopted by more than 137 countries. However, in Pakistan, very few hospitals have implemented ICD coding and conducted different epidemiological studies. Moreover, none of them have reported the spectrum of liver disease burden based on ICD coding, nor implemented automated ICD coding. In this study, we annotated ICD codes for the database of the liver transplant unit of the Pir Abdul Qadir Shah Jeelani Institute of Medical Sciences. We named this database Medical Information Mart for Liver Transplantation (MIMLT). The results revealed that the database contains 34 ICD codes, of which V70.8 is the most frequent code. Furthermore, we determined the spectrum of liver disease burden in liver recipients based on ICD coding. We found that chronic hepatitis C (070.54) is the most frequent indication for liver transplantation. Additionally, we implemented automated ICD coding utilizing the MIMLT database and proposed a novel Deep Recurrent Convolutional Neural Network with Transfer Learning through pre-trained Embeddings (DRCNNTLe) model, which is an extended version of our DRCNN-HP model. DRCNNTLe extracts robust text representations from its pre-trained embedding layer, which is trained on a large domain-specific MIMIC III database corpus. The results indicate that utilizing pre-trained word embeddings, which are trained on large domain-specific corpora can significantly improve the performance of the DRCNNTLe model and provide state-of-the-art results when the target database is small.

Explainable clinical coding with in-domain adapted transformers.

BACKGROUND AND OBJECTIVE: Automatic clinical coding is a crucial task in the process of extracting relevant information from unstructured medical documents contained in Electronic Health Records (EHR). However, most of the existing computer-based methods for clinical coding act as "black boxes", without giving a detailed description of the reasons for the clinical-coding assignments, which greatly limits their applicability to real-world medical scenarios. The objective of this study is to use transformer-based models to effectively tackle explainable clinical-coding. In this way, we require the models to perform the assignments of clinical codes to medical cases, but also to provide the reference in the text that justifies each coding assignment. METHODS: We examine the performance of 3 transformer-based architectures on 3 different explainable clinical-coding tasks. For each transformer, we compare the performance of the original general-domain version with an in-domain version of the model adapted to the specificities of the medical domain. We address the explainable clinical-coding problem as a dual medical named entity recognition (MER) and medical named entity normalization (MEN) task. For this purpose, we have developed two different approaches, namely a multi-task and a hierarchical-task strategy. RESULTS: For each analyzed transformer, the clinical-domain version significantly outperforms the corresponding general domain model across the 3 explainable clinical-coding tasks analyzed in this study. Furthermore, the hierarchical-task approach yields a significantly superior performance than the multi-task strategy. Specifically, the combination of the hierarchical-task strategy with an ensemble approach leveraging the predictive capabilities of the 3 distinct clinical-domain transformers, yields the best obtained results, with f1-score, precision and recall of 0.852, 0.847 and 0.849 on the Cantemist-Norm task and 0.718, 0.566 and 0.633 on the CodiEsp-X task, respectively. CONCLUSIONS: By separately addressing the MER and MEN tasks, as well as by following a context-aware text-classification approach to tackle the MEN task, the hierarchical-task approach effectively reduces the intrinsic complexity of explainable clinical-coding, leading the transformers to establish new SOTA performances for the predictive tasks considered in this study. In addition, the proposed methodology has the potential to be applied to other clinical tasks that require both the recognition and normalization of medical entities.

Validation of diagnostic coding for bronchiectasis in an electronic health record system in Hong Kong.

BACKGROUND AND OBJECTIVE: Electronic medical record (EMR) databases can facilitate epidemiology research in various diseases including bronchiectasis. Given the diagnostic challenges of bronchiectasis, the validity of the coding in EMR requires clarification. We aimed to assess the validity of International Classification of Diseases, 9th Revision (ICD-9) code algorithms for identifying bronchiectasis in the territory-wide electronic medical health record system of Clinical Data Analysis and Reporting System (CDARS) in Hong Kong. MATERIALS AND METHODS: Adult patients who had the diagnosis of bronchiectasis input from Queen Mary Hospital in 2011-2020 were identified using the ICD-9 code of 494 by CDARS. All patients who had high resolution computed tomography (HRCT) were reviewed by respiratory specialists to confirm the presence of bronchiectasis on HRCT. RESULTS: A total of 19 617 patients who had the diagnostic code of bronchiectasis among all public hospitals in Hong Kong and 1866 in Queen Mary Hospital in the same period. Six hundred and forty-eight cases were randomly selected and validated using medical record and HRCT review by a respiratory specialist. The overall positive predictive value (PPV) was 92.7% (95% CI 90.7-94.7). CONCLUSIONS: This was the first ICD-9 coding validation for bronchiectasis in Hong Kong CDARS. Our study demonstrated that using ICD-9 code of 494 was reliable to support utility of CDARS database for further clinical research on bronchiectasis.

Validation Study of Diabetes Definitions Using Japanese Diagnosis Procedure Combination Data Among Hospitalized Patients.

BACKGROUND: Validation studies of diabetes definitions using nationwide healthcare databases are scarce. We evaluated the validity of diabetes definitions using disease codes and antidiabetic drug prescriptions in the Japanese Diagnosis Procedure Combination (DPC) data via medical chart review. METHODS: We randomly selected 500 records among 15,334 patients who participated in the Japan Public Health Center-Based Prospective Study for the Next Generation in Yokote City and who had visited a general hospital in Akita between October 2011 and August 2018. Of the 500 patients, 98 were linked to DPC data; however, only 72 had sufficient information in the medical chart. Gold standard confirmation was performed by board-certified diabetologists. DPC-based diabetes definitions were based on the International Classification of Diseases, 10th Revision codes and antidiabetic prescriptions. Sensitivity, specificity, and the positive and negative predictive values (PPV and NPV, respectively) of DPC-based diabetes definitions were evaluated. RESULTS: Of 72 patients, 23 were diagnosed with diabetes using chart review; 19 had a diabetes code, and 13 had both a diabetes code and antidiabetic prescriptions. The sensitivity, specificity, PPV, and NPV were 89.5% (95% confidence interval [CI], 66.9-98.7%), 96.2% (95% CI, 87.0-99.5%), 89.5% (95% CI, 66.9-98.7%), and 96.2% (95% CI, 87.0-99.5%), respectively, for (i) diabetes codes alone; 89.5% (95% CI, 66.9-98.7%), 94.3% (95% CI, 84.3-98.8%), 85.0% (95% CI, 62.1-96.8%), and 96.2% (95% CI, 86.8-99.5%) for (ii) diabetes codes and/or prescriptions; 68.4% (95% CI, 43.4-87.4%), 100% (95% CI, 93.3-100%), 100% (95% CI, 75.3-100%), and 89.8% (95% CI, 79.2-96.2%) for (iii) both diabetes codes and prescriptions. CONCLUSION: Our results suggest that DPC data can accurately identify diabetes among inpatients using (i) diabetes codes alone or (ii) diabetes codes and/or prescriptions.

Interpreting and coding causal relationships for quality and safety using ICD-11.

Many circumstances necessitate judgments regarding causation in health information systems, but these can be tricky in medicine and epidemiology. In this article, we reflect on what the ICD-11 Reference Guide provides on coding for causation and judging when relationships between clinical concepts are causal. Based on the use of different types of codes and the development of a new mechanism for coding potential causal relationships, the ICD-11 provides an in-depth transformation of coding expectations as compared to ICD-10. An essential part of the causal relationship interpretation relies on the presence of "connecting terms," key elements in assessing the level of certainty regarding a potential relationship and how to proceed in coding a causal relationship using the new ICD-11 coding convention of postcoordination (i.e., clustering of codes). In addition, determining causation involves using documentation from healthcare providers, which is the foundation for coding health information. The coding guidelines and examples (taken from the quality and patient safety domain) presented in this article underline how new ICD-11 features and coding rules will enhance future health information systems and healthcare.

Outpatient neurology diagnostic coding: a proposed scheme for standardised implementation.

Clinical coding uses a classification system to assign standard codes to clinical terms and so facilitates good clinical practice through audit, service design and research. However, despite clinical coding being mandatory for inpatient activity, this is often not so for outpatient services, where most neurological care is delivered. Recent reports by the UK National Neurosciences Advisory Group and NHS England's 'Getting It Right First Time' initiative recommend implementing outpatient coding. The UK currently has no standardised system for outpatient neurology diagnostic coding. However, most new attendances at general neurology clinics appear to be classifiable with a limited number of diagnostic terms. We present the rationale for diagnostic coding and its benefits, and the need for clinical engagement to develop a system that is pragmatic, quick and easy to use. We outline a scheme developed in the UK that could be used elsewhere.

Use of open-text responses to recode categorical survey data on postpartum contraception use among women in the United States: A mixed-methods inquiry of Pregnancy Risk Assessment Monitoring System data.

BACKGROUND: Postpartum contraception prevents unintended pregnancies and short interpregnancy intervals. The Pregnancy Risk Assessment Monitoring System (PRAMS) collects population-based data on postpartum contraception nonuse and reasons for not using postpartum contraception. In addition to quantitative questions, PRAMS collects open-text responses that are typically left unused by secondary quantitative analyses. However, abundant preexisting open-text data can serve as a resource for improving quantitative measurement accuracy and qualitatively uncovering unexpected responses. We used PRAMS survey questions to explore unprompted reasons for not using postpartum contraception and offer insight into the validity of categorical responses. METHODS AND FINDINGS: We used 31,208 categorical 2012 PRAMS survey responses from postpartum women in the US to calculate original prevalences of postpartum contraception use and nonuse and reasons for contraception nonuse. A content analysis of open-text responses systematically recoded data to mitigate survey bias and ensure consistency, resulting in adjusted prevalence calculations and identification of other nonuse themes. Recoded contraception nonuse slightly differed from original reports (21.5% versus 19.4%). Both calculations showed that many respondents reporting nonuse may be at a low risk for pregnancy due to factors like tubal ligation or abstinence. Most frequent nonuse reasons were not wanting to use birth control (27.1%) and side effect concerns (25.0%). Other open-text responses showed common themes of infertility, and breastfeeding as contraception. Comparing quantitative and qualitative responses revealed contradicting information, suggesting respondent misinterpretation and confusion surrounding the term "pregnancy prevention." Though this analysis may be limited by manual coding error and researcher biases, we avoided coding exhaustion via 1-hour coding periods and validated reliability through intercoder kappa scores. CONCLUSIONS: In this study, we observed that respondents reporting contraception nonuse often described other methods of pregnancy prevention and contraception barriers that were not included in categorical response options. Open-text responses shed light on a more comprehensive list of pregnancy prevention methods and nonuse options. Our findings contribute to survey questions that can lead to more accurate depiction of postpartum contraceptive behavior. Additionally, future use of these qualitative methods may be used to improve other health behavior survey development and resulting data.

Administrative Coding in Electronic Health Care Record-Based Research of NAFLD: An Expert Panel Consensus Statement.

BACKGROUND AND AIMS: Electronic health record (EHR)-based research allows the capture of large amounts of data, which is necessary in NAFLD, where the risk of clinical liver outcomes is generally low. The lack of consensus on which International Classification of Diseases (ICD) codes should be used as exposures and outcomes limits comparability and generalizability of results across studies. We aimed to establish consensus among a panel of experts on ICD codes that could become the reference standard and provide guidance around common methodological issues. APPROACH AND RESULTS: Researchers with an interest in EHR-based NAFLD research were invited to collectively define which administrative codes are most appropriate for documenting exposures and outcomes. We used a modified Delphi approach to reach consensus on several commonly encountered methodological challenges in the field. After two rounds of revision, a high level of agreement (>67%) was reached on all items considered. Full consensus was achieved on a comprehensive list of administrative codes to be considered for inclusion and exclusion criteria in defining exposures and outcomes in EHR-based NAFLD research. We also provide suggestions on how to approach commonly encountered methodological issues and identify areas for future research. CONCLUSIONS: This expert panel consensus statement can help harmonize and improve generalizability of EHR-based NAFLD research.

Comparing reliability of ICD-10-based COVID-19 comorbidity data to manual chart review, a retrospective cross-sectional study.

International Statistical Classification of Disease and Related Health Problems, 10th Revision codes (ICD-10) are used to characterize cohort comorbidities. Recent literature does not demonstrate standardized extraction methods. OBJECTIVE: Compare COVID-19 cohort manual-chart-review and ICD-10-based comorbidity data; characterize the accuracy of different methods of extracting ICD-10-code-based comorbidity, including the temporal accuracy with respect to critical time points such as day of admission. DESIGN: Retrospective cross-sectional study. MEASUREMENTS: ICD-10-based-data performance characteristics relative to manual-chart-review. RESULTS: Discharge billing diagnoses had a sensitivity of 0.82 (95% confidence interval [CI]: 0.79-0.85; comorbidity range: 0.35-0.96). The past medical history table had a sensitivity of 0.72 (95% CI: 0.69-0.76; range: 0.44-0.87). The active problem list had a sensitivity of 0.67 (95% CI: 0.63-0.71; range: 0.47-0.71). On day of admission, the active problem list had a sensitivity of 0.58 (95% CI: 0.54-0.63; range: 0.30-0.68)and past medical history table had a sensitivity of 0.48 (95% CI: 0.43-0.53; range: 0.30-0.56). CONCLUSIONS AND RELEVANCE: ICD-10-based comorbidity data performance varies depending on comorbidity, data source, and time of retrieval; there are notable opportunities for improvement. Future researchers should clearly outline comorbidity data source and validate against manual-chart-review.

Stability of diagnostic coding of psychiatric outpatient visits across the transition from the second to the third version of the Danish National Patient Registry.

OBJECTIVE: In Denmark, data on hospital contacts are reported to the Danish National Patient Registry (DNPR). The ICD-10 main diagnoses from the DNPR are often used as proxies for mental disorders in psychiatric research. With the transition from the second version of the DNPR (DNPR2) to the third (DNPR3) in February-March 2019, the way main diagnoses are coded in relation to outpatient treatment changed substantially. Specifically, in the DNPR2, each outpatient treatment course was labelled with only one main diagnosis. In the DNPR3, however, each visit during an outpatient treatment course is labelled with a main diagnosis. We assessed whether this change led to a break in the diagnostic time-series represented by the DNPR, which would pose a threat to the research relying on this source. METHODS: All main diagnoses from outpatients attending the Psychiatric Services of the Central Denmark Region from 2013 to 2021 (n = 100,501 unique patients) were included in the analyses. The stability of the DNPR diagnostic time-series at the ICD-10 subchapter level was examined by comparing means across the transition from the DNPR2 to the DNPR3. RESULTS: While the proportion of psychiatric outpatients with diagnoses from some ICD-10 subchapters changed statistically significantly from the DNPR2 to the DNPR3, the changes were small in absolute terms (e.g., +0.6% for F2-psychotic disorders and +0.6% for F3-mood disorders). CONCLUSION: The change from the DNPR2 to the DNPR3 is unlikely to pose a substantial threat to the validity of most psychiatric research at the diagnostic subchapter level.