To summarize the Canadian health information technology (HIT) policy experience and impart lessons learned to the US as it determines its policy in this area.
Qualitative analysis of interviews with identified key stakeholders followed by an electronic survey.
We conducted semi-structured interviews with 29 key Canadian HIT policy and opinion leaders and used a grounded theory approach to analyze the results. The informant sample was chosen to provide views from different stakeholder groups including national representatives and regional representatives from three Canadian provinces.
Canadian informants believed that much of the current US direction is positive, especially regarding incentives and meaningful use, but that there are key opportunities for the US to emphasize direct engagement with providers, define a clear business case for them, sponsor large scale evaluations to assess HIT impact in a broad array of settings, determine standards but also enable access to resources needed for mid-course corrections of standards when issues are identified, and, finally, leverage implementation of digital imaging systems.
Not all stakeholder groups were included, such as providers or patients. In addition, as in all qualitative research, a selection bias could be present due to the relatively small sample size.
Based on Canadian experience with HIT policy, stakeholders identified as lessons for the US the need to increase direct engagement with providers and the importance of defining the business case for HIT, which can be achieved through large scale evaluations, and of recognizing and leveraging successes as they emerge.
These systems showed that machine learning approaches could be augmented with rule-based systems to determine concepts, assertions, and relations. Depending on the task, the rule-based systems can either provide input for machine learning or post-process the output of machine learning. Ensembles of classifiers, information from unlabeled data, and external knowledge sources can help when the training data are inadequate.
]]>Prescription drugs can be associated with adverse effects (AEs) that are unrecognized despite evidence in the medical literature, as shown by rofecoxib's late recall in 2004. We assessed whether applying information mining to PubMed could reveal major drug–AE associations if articles testing whether drugs cause AEs are over-represented in the literature.
MEDLINE citations published between 1949 and September 2009 were retrieved if they mentioned one of 38 drugs and one of 55 AEs. A statistical document classifier (using MeSH index terms) was constructed to remove irrelevant articles unlikely to test whether a drug caused an AE. The remaining relevant articles were analyzed using a disproportionality analysis that identified drug–AE associations (signals of disproportionate reporting) using step-up procedures developed to control the familywise type I error rate.
Sensitivity and positive predictive value (PPV) for empirical drug–AE associations as judged against drug–AE associations subject to FDA warnings.
In testing, the statistical document classifier identified relevant articles with 81% sensitivity and 87% PPV. Using data filtered by the statistical document classifier, base-case models showed 64.9% sensitivity and 42.4% PPV for detecting FDA warnings. Base-case models discovered 54% of all detected FDA warnings using literature published before warnings. For example, the rofecoxib–heart disease association was evident using literature published before 2002. Analyses incorporating literature mentioning AEs common to the drug class of interest yielded 71.4% sensitivity and 40.7% PPV.
Results from large-scale literature retrieval and analysis (literature mining) compared favorably with and could complement current drug safety methods.