Discontinuation of statins in routine care settings: a cohort study.

BACKGROUND: Systematic data on discontinuation of statins in routine practice of medicine are limited. OBJECTIVE: To investigate the reasons for statin discontinuation and the role of statin-related events (clinical events or symptoms believed to have been caused by statins) in routine care settings. DESIGN: A retrospective cohort study. SETTING: Practices affiliated with Brigham and Women's Hospital and Massachusetts General Hospital in Boston. PATIENTS: Adults who received a statin prescription between 1 January 2000 and 31 December 2008. MEASUREMENTS: Information on reasons for statin discontinuations was obtained from a combination of structured electronic medical record entries and analysis of electronic provider notes by validated software. RESULTS: Statins were discontinued at least temporarily for 57 292 of 107 835 patients. Statin-related events were documented for 18 778 (17.4%) patients. Of these, 11 124 had statins discontinued at least temporarily; 6579 were rechallenged with a statin over the subsequent 12 months. Most patients who were rechallenged (92.2%) were still taking a statin 12 months after the statin-related event. Among the 2721 patients who were rechallenged with the same statin to which they had a statin-related event, 1295 were receiving the same statin 12 months later, and 996 of them were receiving the same or a higher dose. LIMITATIONS: Statin discontinuations and statin-related events were assessed in practices affiliated with 2 academic medical centers. Utilization of secondary data could have led to missing or misinterpreted data. Natural-language-processing tools used to compensate for the low (30%) proportion of reasons for statin discontinuation documented in structured electronic medical record fields are not perfectly accurate. CONCLUSION: Statin-related events are commonly reported and often lead to statin discontinuation. However, most patients who are rechallenged can tolerate statins long-term. This suggests that many of the statin-related events may have other causes, are tolerable, or may be specific to individual statins rather than the entire drug class. PRIMARY FUNDING SOURCE: National Library of Medicine, Diabetes Action Research and Education Foundation, and Chinese National Key Program of Clinical Science.

Statin use and calcific uremic arteriolopathy: a matched case-control study.

BACKGROUND: Calcific uremic arteriolopathy (CUA), also known as calciphylaxis, is characterized by vascular calcification, thrombosis and intense inflammation. Prior research has shown that statins have anticalcification, antithrombotic and antiinflammatory properties; however, the association between statin use and CUA has not been investigated. METHODS: This matched case-control study included 62 adult maintenance hemodialysis (HD) patients with biopsy-confirmed CUA diagnosed between the years 2002 and 2011 (cases). All cases were hospitalized at the time of diagnosis. Controls (n = 124) were hospitalized maintenance HD patients without CUA (matched to cases by gender and timing of hospitalization). Univariate and multivariable logistic regression models were applied to compute odds ratio (OR) and 95% confidence intervals (CI) for CUA in statin users, and also to examine previously described associations. RESULTS: The mean age of cases was 58 years. Most were females (68%), and of white race (64%). Statin use was more common in controls than in cases (39 vs. 19%, p < 0.01). Statin use was associated with lower odds of CUA in unadjusted (OR 0.38, 95% CI 0.18-0.79) and adjusted (OR 0.20, 95% CI 0.05-0.88) analyses. Hypercalcemia (OR 2.25, 95% CI 1.14-4.43), hypoalbuminemia (OR 5.73, 95% CI 2.79-11.77), calcitriol use (OR 5.69, 95% CI 1.02-31.77) and warfarin use (OR 4.30, 95% CI 1.57-11.74) were positively associated with CUA in adjusted analyses whereas paricalcitol and doxercalciferol were not (OR 1.33, 95% CI 0.54-3.27). CONCLUSION: Statin use may be negatively associated with odds of CUA. Further large prospective studies with attention to potential confounders are needed to confirm these findings.

Canary: An NLP Platform for Clinicians and Researchers.

Information Extraction methods can help discover critical knowledge buried in the vast repositories of unstructured clinical data. However, these methods are underutilized in clinical research, potentially due to the absence of free software geared towards clinicians with little technical expertise. The skills required for developing/using such software constitute a major barrier for medical researchers wishing to employ these methods. To address this, we have developed Canary, a free and open-source solution designed for users without natural language processing (NLP) or software engineering experience. It was designed to be fast and work out of the box via a user-friendly graphical interface.

Extracting Healthcare Quality Information from Unstructured Data.

Healthcare quality research is a fundamental task that involves assessing treatment patterns and measuring the associated patient outcomes to identify potential areas for improving healthcare. While both qualitative and quantitative approaches are used, a major obstacle for the quantitative approach is that many useful healthcare quality indicators are buried within provider narrative notes, requiring expensive and laborious manual chart review to identify and measure them. Information extraction is a key Natural Language Processing (NLP) task for discovering and mining critical knowledge buried in unstructured clinical data. Nevertheless, widespread adoption of NLP has yet to materialize; the technical skills required for the development or use of such software present a major barrier for medical researchers wishing to employ these methods. In this paper we introduce Canary, a free and open source solution designed for users without NLP and technical expertise and apply it to four tasks, aiming to measure the frequency of: (1) insulin decline; (2) statin medication decline; (3) adverse reactions to statins; and (3) bariatric surgery counselling. Our results demonstrate that this approach facilitates mining of unstructured data with high accuracy, enabling the extraction of actionable healthcare quality insights from free-text data sources.

Reasons for discontinuation of lipid-lowering medications in patients with chronic kidney disease.

BACKGROUND/AIMS: Many patients with chronic kidney disease (CKD) do not receive lipid-lowering therapy despite their high cardiovascular risk. The reasons for this are unknown. METHODS: We have conducted a retrospective cohort study of discontinuation of lipid-lowering drugs in patients with CKD stage 3 and higher treated in practices affiliated with two academic medical centers between 2000 and 2010. Information on medication discontinuation and its reasons was obtained from electronic medical records, including natural language processing of electronic notes using previously validated software. RESULTS: Out of 14,034 patients in the study cohort, 10,072 (71.8%) stopped their lipid-lowering drugs at least once, and 2,444 (17.4%) stopped them for at least 1 month. Patients who had a comorbidity associated with higher cardiovascular risk were less likely to stop lipid-lowering drugs. Insurance request was the most common explicitly documented reason for discontinuation, and adverse reactions were the most common reason for long-term discontinuation. In a multivariable analysis, patients were more likely to stop a lipid-lowering drug because of an insurance request if they had government insurance and they were also more likely to stop a lipid-lowering drug because of adverse reactions if they had a history of multiple adverse reactions to other medications. There was no significant relationship between CKD stage and the reason for discontinuation of lipid-lowering drugs. CONCLUSIONS: Patients with CKD frequently stop lipid-lowering drugs. Insurance requests and adverse reactions are common reasons for the discontinuation. Further research is needed to ensure appropriate lipid-lowering therapy for these individuals at high cardiovascular risk.

Structured vs. unstructured: factors affecting adverse drug reaction documentation in an EMR repository.

Adverse reactions to medications to which the patient was known to be intolerant are common. Electronic decision support can prevent them but only if history of adverse reactions to medications is recorded in structured format. We have conducted a retrospective study of 31,531 patients with adverse reactions to statins documented in the notes, as identified with natural language processing. The software identified statin adverse reactions with sensitivity of 86.5% and precision of 91.9%. Only 9020 of these patients had an adverse reaction to a statin recorded in structured format. In multivariable analysis the strongest predictor of structured documentation was utilization of EMR functionality that integrated the medication list with the structured medication adverse reaction repository (odds ratio 48.6, p < 0.0001). Integration of information flow between EMR modules can help improve documentation and potentially prevent adverse drug events.