Donald E. Nease

The State of the Evidence for Integrated Behavioral Health in Primary Care

Integrated behavioral health care is a complex, multifaceted healthcare delivery approach that is geared towards addressing mental and behavioral health concerns in primary care. There are a number of different models for integrated behavioral health care, with components that can be conceptualized as structures of care, processes of care, or principles of care. Common models include the IMPACT model (care management for depression), the three-component model (care management, enhanced mental health support, and a prepared practice), and the primary mental health care model of colocated integrated behavioral health care (on-site mental health specialists who collaborate with primary care providers), among others. Meta-analysis has shown that integrated behavioral health care improves health outcomes, although the extant evidence primarily pertains to depression. It is not well known which components of integrated behavioral health care are either necessary or sufficient for improving outcomes. There are many evidence gaps in integrated behavioral health care, including implementation and dissemination and the effects of integrated behavioral health care on disease contexts other than depression, behavioral medicine (e.g., lifestyle change in primary care), diverse populations, and cost and sustainability outcomes. Multiple methodologies should be deployed to address these gaps, including quasi-experimental, mixed methods (quantitative and qualitative), and observational designs.

Clinical Reminders Designed and Implemented Using Cognitive and Organizational Science Principles Decrease Reminder Fatigue

Background: Response rates to point-of-care clinical reminders typically decrease over time. We hypothesized that this “reminder fatigue” could be prevented by (1) applying sound human factors engineering and cognitive science principles in designing the reminder system, and (2) implementing the reminders with rigorous attention to organizational science principles. Methods: This was a retrospective cohort enumeration from January 1, 2006, through July 31, 2012, in a set of 5 academically affiliated family medicine practices. We modeled the odds ratio of clinician action in response to a reminder according to the number of reminders issued during the encounter, the number of problems on the patients problem list, patient age, and time (number of months since launch) using logistic regression with clustering by encounter. Results: There were issued 988,149 reminders at 453,537 encounters during the sampling frame. Action was taken in response to 60.1% of reminders, and discussion or consideration was documented in another 26.8%. The odds ratios for action in response to reminders over time, by number of prompts during the encounter, and by number of problems were 1.01, 1.18, and 1.02, respectively. Key design features included issuing reminders only when a service was due, allowing clinicians to attend to reminders when doing so fit their workflow (vs forcing attention at a specific time), keeping reminders very short and simple (action item only, no explicative material), and a team meeting and buy-in process before each new reminder was implemented. Conclusions: Reminder fatigue over time, with increasing numbers of reminders and with increasing complexity of patients, is not inevitable. A reminder system designed and implemented in accordance with the principles of cognitive science and human factors engineering can prevent reminder fatigue.

Perceived Barriers and Facilitators of Using a Web-Based Interactive Decision Aid for Colorectal Cancer Screening in Community Practice Settings: Findings From Focus Groups With Primary Care Clinicians and Medical Office Staff

Background Information is lacking about the capacity of those working in community practice settings to utilize health information technology for colorectal cancer screening. Objective To address this gap we asked those working in community practice settings to share their perspectives about how the implementation of a Web-based patient-led decision aid might affect patient-clinician conversations about colorectal cancer screening and the day-to-day clinical workflow. Methods Five focus groups in five community practice settings were conducted with 8 physicians, 1 physician assistant, and 18 clinic staff. Focus groups were organized using a semistructured discussion guide designed to identify factors that mediate and impede the use of a Web-based decision aid intended to clarify patient preferences for colorectal cancer screening and to trigger shared decision making during the clinical encounter. Results All physicians, the physician assistant, and 8 of the 18 clinic staff were active participants in the focus groups. Clinician and staff participants from each setting reported a belief that the Web-based patient-led decision aid could be an informative and educational tool; in all but one setting participants reported a readiness to recommend the tool to patients. The exception related to clinicians from one clinic who described a preference for patients having fewer screening choices, noting that a colonoscopy was the preferred screening modality for patients in their clinic. Perceived barriers to utilizing the Web-based decision aid included patients’ lack of Internet access or low computer literacy, and potential impediments to the clinics’ daily workflow. Expanding patients’ use of an online decision aid that is both easy to access and understand and that is utilized by patients outside of the office visit was described as a potentially efficient means for soliciting patients’ screening preferences. Participants described that a system to link the online decision aid to a computerized reminder system could promote a better understanding of patients’ screening preferences, though some expressed concern that such a system could be difficult to keep up and running. Conclusions Community practice clinicians and staff perceived the Web-based decision aid technology as promising but raised questions as to how the technology and resultant information would be integrated into their daily practice workflow. Additional research investigating how to best implement online decision aids should be conducted prior to the widespread adoption of such technology so as to maximize the benefits of the technology while minimizing workflow disruptions.

Evaluation of the Role of Training in the Implementation of a Depression Screening and Treatment Protocol in 2 Academic Outpatient Internal Medicine Clinics Utilizing the Electronic Medical Record

Systematic approaches to depression identification and management are effective though not consistently implemented. The research team implemented a depression protocol, preceded by training, in 2 faculty-resident practices. Medical assistants used the Patient Health Questionnaire (PHQ)-2 for initial screening; providers performed the PHQ-9. These were documented in the electronic medical record. Logistic regression was performed to assess the association of provider type, clinic site, and training attendance with documentation of PHQ-9 after positive PHQ-2s, and with repeat PHQ-9s after positive PHQ-9s. In logistic regression analysis, training attendance was positively associated with documentation of PHQ-9 after a positive PHQ-2 (odds ratio [OR] = 2.4 [confidence interval (CI) = 1.3-4.3]) and repeated documentation of a PHQ-9 after a positive PHQ-9 (OR = 2.5 [CI = 1.1-5.3]). This study describes the successful implementation of a stepped-care approach to depression care. The positive association of training with compliance with protocol procedures indicates the importance of training in the implementation of practice change.

Engaging Patients in Primary and Specialty Care

Abstract Engaging patients as partners in their care is essential for improving health, enhancing the delivery of care, and reducing costs. The need for patient engagement applies across the cancer control continuum, across clinician type (primary and specialty care), and across delivery settings (hospital, ambulatory practices, and the community). Patient engagement is broadly defined as involving patients in their care to actively process health information, deciding how that information fits into their lives, and acting on decisions. While effectively engaging patients in care can be challenging, health information technology (HIT) can support and facilitate the process. HIT that is more patient centered can collect patient-reported information; aggregate existing clinical information; translate medical content into lay language; provide individualized, evidence-based recommendations; and facilitate patient action. This process empowers patients and clinicians with information, engages patients in medical decisions, promotes adoption of healthy behaviors to prevent cancer, encourages screening to identify cancer at an earlier stage when it is more amenable to treatment, facilitates transitions in care, supports treatment, and enhances survivorship planning. To ensure that HIT has the intended impact on patient engagement attention is needed to the technology’s design, adoption, and use by patients and clinicians.

A Community Engagement Method to Design Patient Engagement Materials for Cardiovascular Health

PURPOSE EvidenceNOW Southwest is a cluster-randomized trial evaluating the differential impact on cardiovascular disease (CVD) care of engaging patients and communities in practice transformation in addition to standard practice facilitation support. The trial included development of locally tailored CVD patient engagement materials through Boot Camp Translation (BCT), a community engagement process that occurred before practice recruitment but after cluster randomization. METHODS We introduce a cluster randomization method performed before recruitment of small to medium-size primary care practices in Colorado and New Mexico, which allowed for balanced study arms while minimizing contamination. Engagement materials for the enhanced study arm were developed by means of BCT, which included community members, practice members, and public health professionals from (1) metropolitan Denver, (2) rural northeast Colorado, (3) Albuquerque, and (4) rural southeast New Mexico. Outcome measures were messages and materials from BCTs and population characteristics of study arms after using geographic-based covariate constrained randomization. RESULTS The 4 BCTs’ messages and materials developed by the BCT groups uniquely reflected each community and ranged from family or spiritual values to early prevention or adding relevance to CVD risk. The geographic-based covariate of a cluster randomization method constrained randomization-assigned regions to study arms, allowing BCTs to precede practice recruitment, reduce contamination, and balance populations. CONCLUSIONS Cluster-randomized trials with community-based interventions present study design and implementation challenges. The BCTs elicited unique contextual messages and materials, suggesting that interventions designed to help primary care practices decrease CVD risk may not be one size fits all.

Stakeholder engagement in diabetes self-management: patient preference for peer support and other insights

Background Self-management support (SMS) for patients with diabetes can improve adherence to treatment, mitigate disease-related distress, and improve health outcomes. Translating this evidence into real-world practice is needed, as it is not clear which SMS models are acceptable to patients, and feasible and sustainable for primary care practices. Objective To use the Boot Camp Translation (BCT) method to engage patient, practice, community resource and research stakeholders in translation of evidence about SMS and diabetes distress into mutually acceptable care models and to inform patient-centred outcomes research (PCOR). Participants Twenty-seven diabetes care stakeholders, including patients and providers from a local network of federally qualified health centres participated. Methods Stakeholders met in-person and by conference call over the course of 8 months. Subject matter experts provided education on the diabetes SMS evidence. Facilitators engaged the group in discussions about barriers to self-management and opportunities for improving delivery of SMS. Key Results BCT participants identified lack of social support, personal resources, trust, knowledge and confidence as barriers to diabetes self-management. Intervention opportunities emphasized peer support, use of multidisciplinary care teams and centralized systems for sharing information about community and practice resources. BCT informed new services and a PCOR study proposal. Conclusions Patients and family engaged in diabetes care research value peer support, group visits, and multidisciplinary care teams as key features of SMS models. SMS should be tailored to an individual patients health literacy. BCT can be used to engage multiple stakeholders in translation of evidence into practice and to inform PCOR.

Strategies and impacts of patient and family engagement in collaborative mental healthcare: protocol for a systematic and realist review

Introduction Collaborative mental healthcare (CMHC) has garnered worldwide interest as an effective, team-based approach to managing common mental disorders in primary care. However, questions remain about how CMHC works and why it works in some circumstances but not others. In this study, we will review the evidence on one understudied but potentially critical component of CMHC, namely the engagement of patients and families in care. Our aims are to describe the strategies used to engage people with depression or anxiety disorders and their families in CMHC and understand how these strategies work, for whom and in what circumstances. Methods and analysis We are conducting a review with systematic and realist review components. Review part 1 seeks to identify and describe the patient and family engagement strategies featured in CMHC interventions based on systematic searches and descriptive analysis of these interventions. We will use a 2012 Cochrane review of CMHC as a starting point and perform new searches in multiple databases and trial registers to retrieve more recent CMHC intervention studies. In review part 2, we will build and refine programme theories for each of these engagement strategies. Initial theory building will proceed iteratively through content expert consultations, electronic searches for theoretical literature and review team brainstorming sessions. Cluster searches will then retrieve additional data on contexts, mechanisms and outcomes associated with engagement strategies, and pairs of review authors will analyse and synthesise the evidence and adjust initial programme theories. Ethics and dissemination Our review follows a participatory approach with multiple knowledge users and persons with lived experience of mental illness. These partners will help us develop and tailor project outputs, including publications, policy briefs, training materials and guidance on how to make CMHC more patient-centred and family-centred. PROSPERO registration number CRD42015025522.

Split-Session Focus Group Interviews in the Naturalistic Setting of Family Medicine Offices

PURPOSE When recruiting health care professionals to focus group interviews, investigators encounter challenges such as busy clinic schedules, recruitment, and a desire to get candid responses from diverse participants. We sought to overcome these challenges using an innovative, office-based, split-session focus group procedure in a project that elicited feedback from family medicine practices regarding a new preventive services model. This procedure entails allocating a portion of time to the entire group and the remaining time to individual subgroups. We discuss the methodologic procedure and the implications of using this approach for data collection. METHODS We conducted split-session focus groups with physicians and staff in 4 primary care practices. The procedure entailed 3 sessions, each lasting 30 minutes: the moderator interviewed physicians and staff together, physicians alone, and staff alone. As part of the focus group interview, we elicited and analyzed participant comments about the split-session format and collected observational field notes. RESULTS The split-session focus group interviews leveraged the naturalistic setting of the office for context-relevant discussion. We tested alternate formats that began in the morning and at lunchtime, to parallel each practice’s workflow. The split-session approach facilitated discussion of topics primarily relevant to staff among staff, topics primarily relevant to physicians among physicians, and topics common to all among all. Qualitative feedback on this approach was uniformly positive. CONCLUSION A split-session focus group interview provides an efficient, effective way to elicit candid qualitative information from all members of a primary care practice in the naturalistic setting where they work.

Improving the assessment of depression remission with the Remission Evaluation and Mood Inventory Tool.

Objective The Remission Evaluation and Mood Inventory Tool (REMIT) is a practical 5-item self-report measure of key positive mood states associated with recovering from depression, as distinct from depressive symptoms per se. The study goal was to identify a clinically useful threshold for interpreting REMIT responses in the context of mild to moderate depressive symptoms. Methods This was a secondary analysis of a cross-sectional dataset initially used to develop and validate the REMIT. Primary care patients being treated for depressive symptoms of either mild or moderate severity (n = 247 and 240, respectively) rated their perceived degree of depression remission prior to completing the Patient Health Questionnaire-8 (PHQ-8) and the REMIT. We summed the totals of the latter two measures to form the PHQ + REMIT index. Results Receiver Operating Characteristics analysis indicated that the PHQ + REMIT threshold ≥13 was associated with good sensitivity (92%) and acceptable specificity (43%) to the absence of patient-perceived remission. In contrast, the PHQ had only 21% specificity at this sensitivity level. Area under the curve was 0.815 (95% C.I.: 0.765–0.865), which was significantly greater than that of the PHQ-8 alone (area under the curve = 0.745, 95% C.I.: 0.691–0.805, p(diff) = 0.0002). Threshold performance was unaffected by adjustment for demographic characteristics and variation in remission percentage. Compared with standard symptom-based classification, using the REMIT reclassified 27% of mildly symptomatic patients as remitted. Conclusions Using the REMIT with patients who have mild to moderate depressive symptoms improves the assessment of patient-perceived remission, which is indicated by a summed PHQ + REMIT index of less than 13. Longitudinal research is needed to test whether this broadened patient-centered approach to assessing remission improves clinical decision making and long-term outcomes.