OBJECTIVE:Antidepressant medication use (ADM) has been shown to predict diabetes. This article assessed the role of inflammatory markers in this relationship within the Diabetes Prevention Program (DPP).METHODS:DPP participants randomized to metformin (MET), life-style intervention (ILS), or placebo (PLB) were assessed for depression (Beck Depression Inventory [BDI]) annually, ADM use semiannually, serum inflammatory markers (C-reactive protein [CRP], interleukin 6 [IL-6]) at baseline and year 1, and diagnosis of type 2 diabetes mellitus (T2DM) semiannually (for 3.2 years).RESULTS:At baseline (N = 3187), M (SD) body mass index was 34 (6) kg/m and the median (interquartile range) BDI score was 3 (1-7). One hundred eighty-one (5.7%) reported ADM use and 328 (10%) had BDI scores of 11 or higher. CRP and IL-6 levels did not differ by treatment group. Baseline ADM, but not BDI score, was associated with higher levels of baseline CRP adjusted for demographic, anthropometric variables, and other medications (20% higher, p = .01). Year 1 CRP decreased for non-ADM users in the MET (-13.2%) and ILS (-34%) groups and ADM users in the ILS group (-29%). No associations were found with IL-6. CRP and continuous use of ADM predicted incident T2DM in the PLB group. In the ILS group, continuous and intermittent ADM, but not CRP, predicted T2DM. In the MET group, CRP predicted incident T2DM. CRP did not mediate the risk of T2DM with ADM use in any group.CONCLUSIONS:ADM was significantly associated with elevated CRP and incident T2DM. In the PLB group, ADM and CRP independently predicted onset of T2DM; however, CRP did not significantly mediate the effect of ADM.

Objective To assess in the Diabetes Prevention Program and Diabetes Prevention Program Outcomes Study whether diagnosis of diabetes predicted elevated depressive symptoms (DS) or use of antidepressant medicine (ADM) following diagnosis; whether diabetes status or duration had significant effect on DS or ADM use; and to determine the associations between A1C, fasting plasma glucose (FPG), normalization of FPG, and DS or ADM use after diagnosis. Methods Diabetes Prevention Program participants in three treatment arms (intensive life style, metformin, placebo) were assessed for diabetes, glucose control, ADM use, and DS, measured using the Beck Depression Inventory (BDI). Among 3234 participants, 1285 developed diabetes. Depression levels were measured before and after diabetes diagnosis. Results Neither DS nor use of ADM increased after diagnosis; higher FPG was associated with greater ADM use in the intensive life style arm; a 10-mg/dl rise in FPG is associated with greater odds of ADM use. Higher FPG and A1C were associated with higher BDI scores in all three arms; A 10-mg/dl rise in FPG had a 0.07 increase in BDI. A 1% higher A1c was associated with a 0.21-point increase in BDI. Normalization of FPG was associated with lower BDI. When FPG had normalized, there was a decrease of 0.30 points in the BDI score compared when FPG had not normalized. Conclusions Contrary to clinical attributions, diabetes diagnosis did not show an immediate impact on BDI scores or ADM use. Higher glucose levels after diagnosis were associated with a small but significantly higher BDI score and more ADM use. Trial Registration: DPPOS: NCT00038727; DPP: NCT00004992

Background: Substance abuse in the United States is a serious public health concern impacting morbidity and mortality. However, systematic screening and intervention has not been widely adopted into routine practice by health care organizations and routine screening and intervention is not currently in place for primary care at Kaiser Permanente Colorado. Therefore, a formative evaluation was conducted to explore and enhance implementation of the Substance Abuse and Mental Health Services Administration (SAMHSA) screening, brief intervention, and referral to treatment (SBIRT) approach in the organization. Methods: Key clinical stakeholders, including internal and family medicine physicians, primary care nurses, mental health therapists, chemical dependency clinicians, and clinic-based psychologists provided feedback. Two focus groups were also conducted with patient stakeholders: one in English and one in Spanish. Results: All clinical stakeholders promoted clinic-based psychologists to conduct brief intervention and determine referral to treatment as the optimal implementation program. Inclusion of the patient perspective also highlighted the importance of considering this perspective in implementation. Both patient groups were generally supportive of SBIRT, especially the educational value of screening questions defining healthy drinking limits; however, English-speaking patients noted privacy concerns and Spanish-speaking patients noted frequently being asked about drug or alcohol use. Organizationally, systems exist to facilitate drug and alcohol use screening, intervention, and referral to treatment. However, physician time, alignment with other priorities, and lack of consistent communication were noted potential barriers to SBIRT implementation. Conclusions: Clinicians expressed concerns about competing priorities and the need for organizational leadership involvement for successful SBIRT implementation. A unique suggestion for successful implementation is to utilize existing primary care clinic-based psychologists to conduct brief intervention and facilitate referral to treatment. Patient stakeholders supported universal screening, but cultural differences in opinions and current experience were noted, indicating the importance of including this perspective when evaluating implementation potential.

Objectives: Control of hypertension has improved nationally with focus on identifying and treating elevated blood pressures (BP) to guideline recommended levels. However, once BP control is achieved, the frequency in which BP falls out of control and the factors associated with BP recidivism is unknown. The objective of this study is to examine rates and predictors of blood pressure recidivism in adults with controlled hypertension. Study Design: A retrospective cohort study in two large, integrated health care systems in Minnesota and Colorado. Methods: Patients with a prior diagnosis of hypertension based on a combination of ICD-9 codes, receipt of anti-hypertensive medications and/or elevated blood pressure readings were eligible to be included. We defined controlled hypertension as normotensive blood pressure (BP) readings (less than 140/90 mmHg or less than 130/80 mmHg if coexistent diabetes or chronic kidney disease present) at 2 consecutive primary care visits. Following these visits among patients with controlled BP, we followed patients for BP recidivism defined by the mean of the last 2 blood pressure readings greater than 140/90 mm Hg or 130/80 mm Hg for those with diabetes or chronic kidney disease during a mean follow-up period of 13.2 months . Results: A total of 22,275 patients with controlled hypertension were included in this study. The proportion of patients with hypertension recidivism was 16.4%. A linear increase in blood pressure recidivism was noted with time between the index visit and last observed blood pressure reading. Major predictors of recidivism included female gender (OR 1.14, p = 0.02), history of diabetes (OR 2.32, p < 0.001) and black race (OR 1.37, p = 0.02). Age 50-64 displayed a protective effect against recidivism (OR 0.80, p = 0.02). Conclusions: Hypertensive recidivism occurs in a significant portion of patients with controlled hypertension. Patient factors associated with recidivism include age, race and prior history of diabetes among other characteristics. Strategies to minimize hypertension recidivism have significant potential to improve overall levels of blood pressure control and hypertension related quality measures from a health care systems perspective.

Fifty-two American Board of Family Medicine and 19 American Board of Internal Medicine certified physicians completed projects to increase the percentage of hypertensive patients on their patient panels who had controlled blood pressure. Mean panel blood pressure control improved from 79.49% to 84.64%. The choice of improvement option was not associated with the level of improvement or with the participant's perception of the workload related to completing the project.

JoAnn Sperl-Hillen1, Patrick O’Connor1, David Magid2, David Price2, Stephen Asche1, William Rush1, Heidi Ekstrom1, David Brand2, Paul Johnson3 and Karen Margolis1 1HealthPartners 2Kaiser Permanente Colorado 3University of Minnesota

Few studies have directly investigated the association of clinicians' implicit (unconscious) bias with health care disparities in clinical settings.To determine if clinicians' implicit ethnic or racial bias is associated with processes and outcomes of treatment for hypertension among black and Latino patients, relative to white patients.Primary care clinicians completed Implicit Association Tests of ethnic and racial bias. Electronic medical records were queried for a stratified, random sample of the clinicians' black, Latino and white patients to assess treatment intensification, adherence and control of hypertension. Multilevel random coefficient models assessed the associations between clinicians' implicit biases and ethnic or racial differences in hypertension care and outcomes.Standard measures of treatment intensification and medication adherence were calculated from pharmacy refills. Hypertension control was assessed by the percentage of time that patients met blood pressure goals recorded during primary care visits.One hundred and thirty-eight primary care clinicians and 4,794 patients with hypertension participated. Black patients received equivalent treatment intensification, but had lower medication adherence and worse hypertension control than white patients; Latino patients received equivalent treatment intensification and had similar hypertension control, but lower medication adherence than white patients. Differences in treatment intensification, medication adherence and hypertension control were unrelated to clinician implicit bias for black patients (P = 0.85, P = 0.06 and P = 0.31, respectively) and for Latino patients (P = 0.55, P = 0.40 and P = 0.79, respectively). An increase in clinician bias from average to strong was associated with a relative change of less than 5 % in all outcomes for black and Latino patients.Implicit bias did not affect clinicians' provision of care to their minority patients, nor did it affect the patients' outcomes. The identification of health care contexts in which bias does not impact outcomes can assist both patients and clinicians in their efforts to build trust and partnership.

PURPOSE:To assess the impact of personalised physician learning (PPL) interventions using simulated learning cases on control of hypertension and dyslipidaemia in primary care settings. METHODS:A total of 132 primary care physicians, 4568 eligible patients with uncontrolled hypertension, and 15 392 eligible patients with uncontrolled dyslipidaemia were cluster-randomised to one of three conditions: (a) no intervention, (b) PPL-electronic medical record (EMR) intervention in which 12 PPL cases were assigned to each physician based on observed patterns of care in the EMR in the previous year, or (c) PPL-ASSESS intervention in which 12 PPL cases were assigned to each physician based on their performance on four standardised assessment cases. General and generalised linear mixed models were used to account for clustering and to model differences in patient outcomes in the study arms. RESULTS:Among patients with uncontrolled hypertension at baseline, 49.1%, 46.6% and 47.3% (p=0.43) achieved blood pressure (BP) targets at follow-up. Among patients with uncontrolled dyslipidaemia at baseline, 37.5%, 37.3% and 38.1% (p=0.72) achieved low density lipoprotein cholesterol targets at follow-up in PPL-EMR, PPL-ASSESS and the control group, respectively. Although systolic (BP) (p<0.001) and lipid (p<0.001) values significantly improved during the study, the group-by-time interaction term showed no differential change in systolic BP values (p=0.51) or lipid values (p=0.61) among the three study arms. No difference in intervention effect was noted when comparing the PPL-EMR with the PPL-ASSESS intervention (p=0.47). CONCLUSIONS:The two PPL interventions tested in this study did not lead to improved control of hypertension or dyslipidaemia in primary care clinics during a mean 14-month follow-up period. This null result may have been due in part to substantial overall improvement in BP and lipid control at the study sites during the study. TRIAL REGISTRATION NUMBER:NCT00903071.

OBJECTIVE To determine whether depression symptoms or antidepressant medication use predicts weight regain in overweight individuals with impaired glucose tolerance (IGT) who are successful with initial weight loss. RESEARCH DESIGN AND METHODS A total of 1,442 participants who successfully lost at least 3% of their baseline body weight after 12 months of participation in the randomized controlled Diabetes Prevention Program (DPP) continued in their assigned treatment group (metformin, intensive lifestyle, or placebo) and were followed into the Diabetes Prevention Program Outcome Study (DPPOS). Weight regain was defined as a return to baseline DPP body weight. Participant weight and antidepressant medication use were assessed every 6 months. Depression symptoms (Beck Depression Inventory [BDI] score ≥11) were assessed every 12 months. RESULTS Only 2.7% of the overall cohort had moderate to severe depression symptoms at baseline; most of the participants with BDI score ≥11 had only mild symptoms during the period of observation. In unadjusted analyses, both depression symptoms (hazard ratio 1.31 [95% CI 1.03–1.67], P = 0.03) and antidepressant medication use at either the previous visit (1.72 [1.37–2.15], P < 0.0001) or cumulatively as percent of visits (1.005 [1.002–1.008], P = 0.0003) were predictors of subsequent weight regain. After adjustment for multiple covariates, antidepressant use remained a significant predictor of weight regain (P < 0.0001 for the previous study visit; P = 0.0005 for the cumulative measure), while depression symptoms did not. CONCLUSIONS In individuals with IGT who do not have severe depression and who initially lose weight, antidepressant use may increase the risk of weight regain.

PURPOSE:We investigated whether clinicians' explicit and implicit ethnic/racial bias is related to black and Latino patients' perceptions of their care in established clinical relationships.METHODS:We administered a telephone survey to 2,908 patients, stratified by ethnicity/race, and randomly selected from the patient panels of 134 clinicians who had previously completed tests of explicit and implicit ethnic/racial bias. Patients completed the Primary Care Assessment Survey, which addressed their clinicians' interpersonal treatment, communication, trust, and contextual knowledge. We created a composite measure of patient-centered care from the 4 subscales.RESULTS:Levels of explicit bias were low among clinicians and unrelated to patients' perceptions. Levels of implicit bias varied among clinicians, and those with greater implicit bias were rated lower in patient-centered care by their black patients as compared with a reference group of white patients (P = .04). Latino patients gave the clinicians lower ratings than did other groups (P <.0001), and this did not depend on the clinicians' implicit bias (P = .98).CONCLUSIONS:This is among the first studies to investigate clinicians' implicit bias and communication processes in ongoing clinical relationships. Our findings suggest that clinicians' implicit bias may jeopardize their clinical relationships with black patients, which could have negative effects on other care processes. As such, this finding supports the Institute of Medicine's suggestion that clinician bias may contribute to health disparities. Latinos' overall greater concerns about their clinicians appear to be based on aspects of care other than clinician bias.

OBJECTIVESWe assessed implicit and explicit bias against both Latinos and African Americans among experienced primary care providers (PCPs) and community members (CMs) in the same geographic area.METHODSTwo hundred ten PCPs and 190 CMs from 3 health care organizations in the Denver, Colorado, metropolitan area completed Implicit Association Tests and self-report measures of implicit and explicit bias, respectively.RESULTSWith a 60% participation rate, the PCPs demonstrated substantial implicit bias against both Latinos and African Americans, but this was no different from CMs. Explicit bias was largely absent in both groups. Adjustment for background characteristics showed the PCPs had slightly weaker ethnic/racial bias than CMs.CONCLUSIONSThis research provided the first evidence of implicit bias against Latinos in health care, as well as confirming previous findings of implicit bias against African Americans. Lack of substantive differences in bias between the experienced PCPs and CMs suggested a wider societal problem. At the same time, the wide range of implicit bias suggested that bias in health care is neither uniform nor inevitable, and important lessons might be learned from providers who do not exhibit bias.

Alanna Kulchak Rahm1, David Price1, Arne Beck1, Carmen Martin1, Jennifer Boggs1, Thomas Backer2 and James Dearing1 1Kaiser Permanente Colorado 2Human Interaction Research Institute

Irene Blair, PhD1,2, David Price, MD3, Ed Havranek, MD4, John Steiner, MD, MPH2, Rebecca Hanratty, MD4, Diane Fairclough, DrPh5, Tillman Farley, MD6 and Holen Katz, MA1 1University of Colorado Boulder 2Kaiser Permanente Colorado 3Kaiser Permanente Colorado Institute for Health Research 4Denver Health 5University of Colorado Health Sciences Center 6Salud Family Health Center

Introduction: Continuing medical education (CME) is meant to drive and support improvements in practice. To achieve this goal, CME activities must move beyond simply purveying knowledge, instead helping attendees to contextualize information and to develop strategies for implementing new learning. CME attendees face different barriers to implementing learning, depending on both personal and practice specific contexts. We sought to develop a framework, applicable across multiple CME activities, for categorizing barriers that learners anticipated encountering after CME activities.Methods: Building on previous work, qualitative research methods were used to develop an enhanced framework classifying attendee-perceived barriers to implementing CME learnings in practice. Three thousand one hundred thirty (3130) narrative responses on attendee-perceived barriers to implementing learnings were collected from 75 Kaiser Permanente Colorado live CME activities for family medicine, internal medicine, pediatric, and OB/GYN clinicians in 2008 and 2009.Results: Our CME Learning Transfer Barrier Framework contains 27 discrete barriers in 12 barrier categories (including "none"). The barrier framework was applicable across two years of live CME activities for different clinician target audiences.Discussion: Assessing, characterizing, and summarizing barriers to implementing learning during CME activities can provide valuable information to inform subsequent CME interventions, and provide feedback to organizational leaders to inform performance improvement efforts. The framework may be applicable to other CME formats and to CME activities for audiences in different practice settings.

OBJECTIVE To assess the association between antidepressant medicine use and risk of developing diabetes during the Diabetes Prevention Program (DPP) and Diabetes Prevention Program Outcomes Study (DPPOS). RESEARCH DESIGN AND METHODS DPP/DPPOS participants were assessed for diabetes every 6 months and for antidepressant use every 3 months in DPP and every 6 months in DPPOS for a median 10.0-year follow-up. RESULTS Controlled for factors associated with diabetes risk, continuous antidepressant use compared with no use was associated with diabetes risk in the placebo (adjusted hazard ratio 2.34 [95% CI 1.32–4.15]) and lifestyle (2.48 [1.45–4.22]) arms, but not in the metformin arm (0.55 [0.25–1.19]). CONCLUSIONS Continuous antidepressant use was significantly associated with diabetes risk in the placebo and lifestyle arms. Measured confounders and mediators did not account for this association, which could represent a drug effect or reflect differences not assessed in this study between antidepressant users and nonusers.

OBJECTIVE To assess the effect of a targeted age-specific computerized alert to reduce D-dimer testing in elderly patients. STUDY DESIGN A single-crossover cluster randomized trial of computerized alerts during physician order entry involving 8 ambulatory care clinics in a group-model integrated care delivery system. METHODS The rate of completed D-dimer tests per 1000 patient visits, ratio of completed venous ultrasonography to completed D-dimer tests, and rate of completed venous ultrasonography per 1000 patient visits. RESULTS The rate of completed D-dimer tests per 1000 visits among patients 65 years and older in intervention clinics decreased from 5.02 to 1.52 (95% confidence interval [CI], -4.20 to -2.80; P <.001), which persisted throughout the study period. The rate of completed D-dimer tests per 1000 visits among patients 65 years and older in control clinics decreased from 3.14 to 2.11 (95% CI, -1.66 to -0.04; P <.001 for interaction). After activation of the alert in control clinics, the rate of completed D-dimer tests per 1000 visits among patients 65 years and older decreased from 2.11 to 0.81 (95% Cl, -1.79 to -0.80; P <.001). After activation of the alert in each clinic group, the ratios of completed venous ultrasonography to completed D-dimer tests increased from 1.17 to 4.05 (95% CI, 2.52-3.22) and from 2.25 to 7.29 (95% CI, 3.74-6.35) in intervention clinics and control clinics, respectively (P <.001 for both). CONCLUSION An electronic age-specific alert targeted to a specific condition reduced D-dimer testing in this elderly population of outpatients and demonstrated a persistent effect.