Determinants of the de-implementation of low-value care: a multi-method study.

BACKGROUND: There is an urgent need to understand the determinants (i.e., barriers and facilitators) of de-implementation. The purpose of this study was to develop a comprehensive list of determinants of the de-implementation of low-value care from the published literature and to compare this list to determinants identified by a group of stakeholders with lived experience with de-implementation. METHODS: This was a two-phase multi-method study. First, a systematic review examined published barriers and facilitators to de-implementation. Articles were identified through searches within electronic databases, reference lists and the grey literature. Citations were screened independently and in duplicate and included if they were: 1) written in English; and 2) described a barrier or facilitator to de-implementation of any clinical practice in adults (age ≥ 18 years). 'Raw text' determinants cited within included articles were extracted and synthesized into a list of representative determinants using conventional content analysis. Second, semi-structured interviews were conducted with decision-makers (unit managers and medical directors) and healthcare professionals working in adult critical care medicine to explore the overlap between the determinants found in the systematic review to those experienced in critical care medicine. Thematic content analysis was used to identify key themes emerging from the interviews. RESULTS: In the systematic review, reviewers included 172 articles from 35,368 unique citations. From 437 raw text barriers and 280 raw text facilitators, content analysis produced 29 distinct barriers and 24 distinct facilitators to de-implementation. Distinct barriers commonly cited within raw text included 'lack of credible evidence to support de-implementation' (n = 90, 21%), 'entrenched norms and clinicians' resistance to change (n = 43, 21%), and 'patient demands and preferences' (n = 28, 6%). Distinct facilitators commonly cited within raw text included 'stakeholder collaboration and communication' (n = 43, 15%), and 'availability of credible evidence' (n = 33, 12%). From stakeholder interviews, 23 of 29 distinct barriers and 20 of 24 distinct facilitators from the systematic review were cited as key themes relevant to de-implementation in critical care. CONCLUSIONS: The availability and quality of evidence that identifies a clinical practice as low-value, as well as healthcare professional willingness to change, and stakeholder collaboration are common and important determinants of de-implementation and may serve as targets for future de-implementation initiatives. TRIAL REGISTRATION: The systematic review was registered in PROSPERO CRD42016050234 .

Translating evidence to patient care through caregivers: a systematic review of caregiver-mediated interventions.

BACKGROUND: Caregivers may promote the uptake of science into patient care and the practice of evidence-informed medicine. The purpose of this study was to determine whether caregiver-mediated (non-clinical caregiver-delivered) interventions are effective in improving patient, caregiver, provider, or health system outcomes. METHODS: We searched the MEDLINE, Embase, PsycINFO, Cumulative Index of Nursing and Allied Health, and Scopus databases from inception to February 27, 2017. Interventions (with a comparison group) reporting on a quality improvement intervention mediated by a caregiver and directed to a patient, in all ages and patient-care settings, were selected for inclusion. A three-category framework was developed to characterize caregiver-mediated interventions: inform (e.g., provide knowledge), activate (e.g., prompt action), and collaborate (e.g., lead to interaction between caregivers and other groups [e.g., care providers]). RESULTS: Fifty-six studies met the inclusion criteria, and 64% were randomized controlled trials (RCTs). The most commonly assessed outcomes were patient- (n = 40) and caregiver-oriented (n = 33); few health system- (n = 10) and provider-oriented (n = 2) outcomes were reported. Patient outcomes (e.g., satisfaction) were most improved by caregiver-mediated interventions that provided condition and treatment education (e.g., symptom management information) and practical condition-management support (e.g., practicing medication protocol). Caregiver outcomes (e.g., stress-related/psychiatric outcomes) were most improved by interventions that activated caregiver roles (e.g., monitoring blood glucose) and provided information related to that action (e.g., why and how to monitor). The risk of bias was generally high, and the overall quality of the evidence was low-moderate, based on Grading of Recommendations Assessment Development and Evaluation ratings. CONCLUSIONS: There is a large body of research, including many RCTs, to support the use of caregiver-mediated interventions that inform and activate caregivers to improve patient and caregiver outcomes. Select caregiver-mediated interventions improve patient (inform-activate) and caregiver (inform-activate-collaborate) outcomes and should be considered by all researchers implementing patient- and family-oriented research. SYSTEMATIC REVIEW: PROSPERO, CRD42016052509 .

Validation of Caregiver-Centered Delirium Detection Tools: A Systematic Review.

OBJECTIVES: To summarize the validity of caregiver-centered delirium detection tools in hospitalized adults and assess associated patient and caregiver outcomes. DESIGN: Systematic review. SETTING: We searched MEDLINE, EMBASE, PsycINFO, CINAHL, and Scopus from inception to May 15, 2017. PARTICIPANTS: Hospitalized adults. INTERVENTION: Caregiver-centered delirium detection tools. MEASUREMENTS: We drafted a protocol from the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Two reviewers independently completed abstract and full-text review, data extraction, and quality assessment. We summarized findings using descriptive statistics including mean, median, standard deviation, range, frequencies (percentages), and Cohen's kappa. Studies that reported on the validity of caregiver-centered delirium detection tools or associated patient and caregiver outcomes and were cohort or cross-sectional in design were included. RESULTS: We reviewed 6,056 titles and abstracts, included 6 articles, and identified 6 caregiver-centered tools. All tools were designed to be administered in several minutes or less and had 11 items or fewer. Three tools were caregiver administered (completed independently by caregivers): Family Confusion Assessment Method (FAM-CAM), Informant Assessment of Geriatric Delirium (I-AGeD), and Sour Seven. Three tools were caregiver informed (administered by a healthcare professional using caregiver input): Single Question in Delirium (SQiD), Single Screening Question Delirium (SSQ-Delirium), and Stressful Caregiving Response to Experiences of Dying. Caregiver-administered tools had better psychometric properties (FAM-CAM sensitivity 75%, 95% confidence interval (CI)=35-95%, specificity 91%, 95% CI=74-97%; Sour Seven positive predictive value 89.5%, negative predictive value 90%) than caregiver-informed tools (SQiD: sensitivity 80%, 95% CI=28.4-99.5%; specificity 71%, 95% CI=41.9-91.6%; SSQ-Delirium sensitivity 79.6%, specificity 56.1%). CONCLUSION: Delirium detection is essential for appropriate delirium management. Caregiver-centered delirium detection tools show promise in improving delirium detection and associated patient and caregiver outcomes. Comparative studies using larger sample sizes and multiple centers are required to determine validity and reliability characteristics.