Francine L. Maloney

An automated technique for identifying associations between medications, laboratory results and problems

BACKGROUND The patient problem list is an important component of clinical medicine. The problem list enables decision support and quality measurement, and evidence suggests that patients with accurate and complete problem lists may have better outcomes. However, the problem list is often incomplete. OBJECTIVE To determine whether association rule mining, a data mining technique, has utility for identifying associations between medications, laboratory results and problems. Such associations may be useful for identifying probable gaps in the problem list. DESIGN Association rule mining was performed on structured electronic health record data for a sample of 100,000 patients receiving care at the Brigham and Womens Hospital, Boston, MA. The dataset included 272,749 coded problems, 442,658 medications and 11,801,068 laboratory results. MEASUREMENTS Candidate medication-problem and laboratory-problem associations were generated using support, confidence, chi square, interest, and conviction statistics. High-scoring candidate pairs were compared to a gold standard: the Lexi-Comp drug reference database for medications and Mosbys Diagnostic and Laboratory Test Reference for laboratory results. RESULTS We were able to successfully identify a large number of clinically accurate associations. A high proportion of high-scoring associations were adjudged clinically accurate when evaluated against the gold standard (89.2% for medications with the best-performing statistic, chi square, and 55.6% for laboratory results using interest). CONCLUSION Association rule mining appears to be a useful tool for identifying clinically accurate associations between medications, laboratory results and problems and has several important advantages over alternative knowledge-based approaches.

A method and knowledge base for automated inference of patient problems from structured data in an electronic medical record

BACKGROUND Accurate knowledge of a patients medical problems is critical for clinical decision making, quality measurement, research, billing and clinical decision support. Common structured sources of problem information include the patient problem list and billing data; however, these sources are often inaccurate or incomplete. OBJECTIVE To develop and validate methods of automatically inferring patient problems from clinical and billing data, and to provide a knowledge base for inferring problems. STUDY DESIGN AND METHODS We identified 17 target conditions and designed and validated a set of rules for identifying patient problems based on medications, laboratory results, billing codes, and vital signs. A panel of physicians provided input on a preliminary set of rules. Based on this input, we tested candidate rules on a sample of 100,000 patient records to assess their performance compared to gold standard manual chart review. The physician panel selected a final rule for each condition, which was validated on an independent sample of 100,000 records to assess its accuracy. RESULTS Seventeen rules were developed for inferring patient problems. Analysis using a validation set of 100,000 randomly selected patients showed high sensitivity (range: 62.8-100.0%) and positive predictive value (range: 79.8-99.6%) for most rules. Overall, the inference rules performed better than using either the problem list or billing data alone. CONCLUSION We developed and validated a set of rules for inferring patient problems. These rules have a variety of applications, including clinical decision support, care improvement, augmentation of the problem list, and identification of patients for research cohorts.

Clinician attitudes toward and use of electronic problem lists: a thematic analysis

BackgroundThe clinical problem list is an important tool for clinical decision making, quality measurement and clinical decision support; however, problem lists are often incomplete and provider attitudes towards the problem list are poorly understood.MethodsAn ethnographic study of healthcare providers conducted from April 2009 to January 2010 was carried out among academic and community outpatient medical practices in the Greater Boston area across a wide range of medical and surgical specialties. Attitudes towards the problem list were then analyzed using grounded theory methods.ResultsAttitudes were variable, and dimensions of variations fit into nine themes: workflow, ownership and responsibility, relevance, uses, content, presentation, accuracy, alternatives, support/education and one cross-cutting theme of culture.ConclusionsSignificant variation was observed in clinician attitudes towards and use of the electronic patient problem list. Clearer guidance and best practices for problem list utilization are needed.

Improving completeness of electronic problem lists through clinical decision support: a randomized, controlled trial

Background Accurate clinical problem lists are critical for patient care, clinical decision support, population reporting, quality improvement, and research. However, problem lists are often incomplete or out of date. Objective To determine whether a clinical alerting system, which uses inference rules to notify providers of undocumented problems, improves problem list documentation. Study Design and Methods Inference rules for 17 conditions were constructed and an electronic health record-based intervention was evaluated to improve problem documentation. A cluster randomized trial was conducted of 11 participating clinics affiliated with a large academic medical center, totaling 28 primary care clinical areas, with 14 receiving the intervention and 14 as controls. The intervention was a clinical alert directed to the provider that suggested adding a problem to the electronic problem list based on inference rules. The primary outcome measure was acceptance of the alert. The number of study problems added in each arm as a pre-specified secondary outcome was also assessed. Data were collected during 6-month pre-intervention (11/2009–5/2010) and intervention (5/2010–11/2010) periods. Results 17 043 alerts were presented, of which 41.1% were accepted. In the intervention arm, providers documented significantly more study problems (adjusted OR=3.4, p<0.001), with an absolute difference of 6277 additional problems. In the intervention group, 70.4% of all study problems were added via the problem list alerts. Significant increases in problem notation were observed for 13 of 17 conditions. Conclusion Problem inference alerts significantly increase notation of important patient problems in primary care, which in turn has the potential to facilitate quality improvement. Trial Registration ClinicalTrials.gov: NCT01105923.

USB-based Personal Health Records: an analysis of features and functionality.

PURPOSE To determine the features of commercially available USB-based Personal Health Records (PHR) devices, and compare the commercial state of the art to recommendations made by certification committees. METHODS Thirteen USB-based PHRs were identified and analyzed based on data elements used and features provided. Marketing techniques used by the companies were also explored. RESULTS Eight of the thirteen PHRs contained all seven clinical data elements (problems, procedures, medications, providers, allergies, labs, immunizations), three were missing a single element and the remaining two lacked two elements. In the features analysis no single PHR contained all eight features (export data, import data, images, summary print out, emergency entry, teaching material available for problem, username and password supported, Mac-compatible), but two of the devices had seven of the eight features. Finally, scare tactics were used in marketing all but two of the PHR devices. CONCLUSION While PHRs are very important in the health care field, at the present time, USB-based PHRs currently on the market appear to have deficiencies. Tethered or web-based PHRs may be a better option for consumers at present.

Use of an Electronic Problem List by Primary Care Providers and Specialists

ABSTRACTBACKGROUNDAccurate patient problem lists are valuable tools for improving the quality of care, enabling clinical decision support, and facilitating research and quality measurement. However, problem lists are frequently inaccurate and out-of-date and use varies widely across providers.OBJECTIVEOur goal was to assess provider use of an electronic problem list and identify differences in usage between medical specialties.DESIGNChart review of a random sample of 100,000 patients who had received care in the past two years at a Boston-based academic medical center.PARTICIPANTSCounts were collected of all notes and problems added for each patient from 1/1/2002 to 4/30/2010. For each entry, the recording provider and the clinic in which the entry was recorded was collected. We used the Healthcare Provider Taxonomy Code Set to categorize each clinic by specialty.MAIN MEASURESWe analyzed the problem list use across specialties, controlling for note volume as a proxy for visits.KEY RESULTSA total of 2,264,051 notes and 158,105 problems were recorded in the electronic medical record for this population during the study period. Primary care providers added 82.3% of all problems, despite writing only 40.4% of all notes. Of all patients, 49.1% had an assigned primary care provider (PCP) affiliated with the hospital; patients with a PCP had an average of 4.7 documented problems compared to 1.5 problems for patients without a PCP.CONCLUSIONSPrimary care providers were responsible for the majority of problem documentation; surgical and medical specialists and subspecialists recorded a disproportionately small number of problems on the problem list.

Increasing Patient Engagement: Patients’ Responses to Viewing Problem Lists Online

OBJECTIVE To characterize the opinions, emotions, and actions taken by patients who viewed their electronic problem list via an online personal health record (PHR). MATERIALS AND METHODS An online survey of patients who viewed their problem lists, as maintained by their healthcare provider, in a web-based PHR linked to an electronic health record for the first time. RESULTS A total 3,649 patients completed the survey, yielding a response rate of 42.1%. Patient attitudes towards the problem list function were positive overall, with 90.4% rating it at least somewhat useful and 86.7% reporting they would probably or definitely use it again. Nearly half (45.6%) of patients identified at least one major or minor problem missing from their list. After viewing the list, 56.1% of patients reported taking at least one action in response, with 32.4% of patients reporting that they researched a condition on the Internet, 18.3% reported that they contacted their healthcare provider and 16.7% reported changing or planning to change a health behavior (patients could report multiple actions). 64.7% of patients reported feeling at least somewhat happy while viewing their problem list, though others reported feeling sad (30.4%), worried (35.7%) or scared (23.8%) (patients could report multiple emotions). A smaller number of patients reported feeling angry (16.6%) or ashamed (14.3%). Patients who experienced an emotional response were more likely to take action. CONCLUSION Overall, patients found the ability to view their problem lists very useful and took action in response to the information. However, some had negative emotions. More research is needed into optimal strategies for supporting patients receiving this information.

Reducing risk with clinical decision support: a study of closed malpractice claims.

OBJECTIVE Identify clinical opportunities to intervene to prevent a malpractice event and determine the proportion of malpractice claims potentially preventable by clinical decision support (CDS). MATERIALS AND METHODS Cross-sectional review of closed malpractice claims over seven years from one malpractice insurance company and seven hospitals in the Boston area. For each event, clinical opportunities to intervene to avert the malpractice event and the presence or absence of CDS that might have a role in preventing the event, were assigned by a panel of expert raters. Compensation paid out to resolve a claim (indemnity), was associated with each CDS type. RESULTS Of the 477 closed malpractice cases, 359 (75.3%) were categorized as substantiated and 195 (54%) had at least one opportunity to intervene. Common opportunities to intervene related to performance of procedure, diagnosis, and fall prevention. We identified at least one CDS type for 63% of substantiated claims. The 41 CDS types identified included clinically significant test result alerting, diagnostic decision support and electronic tracking of instruments. Cases with at least one associated intervention accounted for

A Picture is Worth 1,000 Words

40.3 million (58.9%) of indemnity. DISCUSSION CDS systems and other forms of health information technology (HIT) are expected to improve quality of care, but their potential to mitigate risk had not previously been quantified. Our results suggest that, in addition to their known benefits for quality and safety, CDS systems within HIT have a potential role in decreasing malpractice payments. CONCLUSION More than half of malpractice events and over

A Systematic Intervention To Improve Serious Illness Communication In Primary Care

40 million of indemnity were potentially preventable with CDS.