No abstract

In Brief Providing diabetes patients all of the care recommended by current guidelines is a clinical challenge. Geisinger Health System has designed a provider-led, team-based system of care to more consistently and reliably meet this challenge. This system of care uses an all-or-none bundle of diabetes measures and electronic health record tools to improve both process measures and intermediate diabetes outcomes.

over 15,000 children ages 4-8 had at least one height and one weight recorded in the EHR at the same visit with BMI> 85th % for age, sex. 11,276 physician letters were mailed: 9,954 opt in letters over 13-months (Cohorts 1-3) and1,322 opt out letters in 2-months (Cohort 4). Of the 170 families screened, 58/107 (54%) identified opt in letters as recruitment source; 49/63 (78%) families in Cohort 4 reported opt out letters introduced them to the study. Fifty-six families gave written informed consent and/or assent and were randomized to one of two groups -Group Family Behavior Modification (FBM) or Enhanced Information Mailing (EIM) group. Three FBM and two EIM families dropped before treatment started. Conclusions: Physician letters to parents of eligible children identified in the EHR were the most successful recruitment strategies for the primary care obesity prevention study. Opt-out physician letters shortened recruitment time and doubled the number of participants recruited for the final cohort. Background: Rapidly growing numbers of Type 2 Diabetics (T2D) in the US continues to escalate need for evidence-based primary care interventions to reduce complications and costs. Limited time, information, and lack of revenue for chronic disease management create gaps between national guidelines and primary care for diabetes. Methods: Of the over 20,000 diabetics were identified in Geisinger EPIC ® EHR, 3166 T2D with diagnosis on problem list or ICD-9 code had HgA1c>8.0%. Physician letters or emails were sent to invite them to call to Opt OUT if they did not want more information. After a 10 day period, staff called 1932 eligible T2D to invite them to a shared medical primary care visit to participate in a randomized controlled trial comparing a tailored 5-month web-based lifestyle intervention (dLifeG.com) to usual care. 166 Type 2 diabetics gave written informed consent and were randomized 1:2 to control group or intervention group. In <20 minutes with simple computer instructions, T2Ds in intervention group created password protected personalized website to set goals, view weekly lessons, and take interactive quizzes to improve diabetes knowledge and selfmanagement. Weekly emails with lesson topics and links to the dLifeG.com were sent to intervention group participants. At end of study, control group will also have access to website intervention. Results: At midpoint, ~3/4th of the 100 T2 diabetics in intervention group were engaged [defined by the number of site page viewed (0-3090 pages), emails opened (0-100%), and quiz pages consumed (0-578)]. Mid and end of study changes in diabetes knowledge, HgA1c, Blood Pressure, and weight will be discussed. Conclusions: Web-based lifestyle interventions can be employed in primary care to engage a majority of diabetics with HgA1c>8.0 in a self-management lifestyle modification intervention. Importantly, the study also helped identify diabetics who may need additional resources and assistance with chronic disease management.

HMORN-Selected Abstracts and those from people with comorbidities (cancer, chronic lung disease, or congestive heart failure). We estimated the sensitivity, specificity, and positive predictive value (PPV) of ONYX compared to manual review using multivariable logistic regression, accounting for clustering of reports within people. We examined how ONYX's performance varied by age and comorbidity. Results: ONYX classified 26% of reports (1,276/5,000; 38% [841/2,200] of true pneumonias and 16% [435/2,800] of non-pneumonias) as "requiring manual review" based on pre-defined criteria. Reports from older people were more likely to require manual review than those from younger people. Among reports that could be classified, ONYX had a sensitivity of 91% (1,242/1,359), specificity of 92% (2,170/2,365), and PPV of 81% (1,242/1,437, modeled based on pneumonia prevalence in the source database). Sensitivity and specificity were similar regardless of comorbidity. Conclusions: NLP offers potential for identifying pneumonia outcomes from EMR data. Next steps include 1) further training to decrease the proportion of reports requiring manual review and 2) evaluating the accuracy of ONYX in other health systems.

No abstract