Implementation science communications最新文献

Background: Machine learning (ML)-a field of study dedicated to the principled extraction of knowledge from complex data-can benefit implementation science, quality improvement (QI), and primary care research. Given the general complexity of implementation research and the need to develop strategies for understanding relationships among practice characteristics and practice facilitation strategies, we chose the Implementation Research Logic Model (IRLM) as an underlying structure for the data and to identify relationships that might be associated with outcomes. This study illustrates this novel method involving ML and an IRLM in the context of a practice facilitation-supported QI program in primary care.

Methods: We applied advanced statistical methods within a machine learning framework to data from the Healthy Hearts in the Heartland (H3) study, including practice facilitation data and practice and staff participation survey, to assess the relationship between practice attributes and practice facilitator strategies and their impact on successful implementation of QI interventions. We used PCA for feature selection, incorporated practice facilitators' knowledge for contextual factor validation, and employed Structural Equation Modeling (SEM) to analyze relationships among contextual factors, latent variables, practice facilitation strategies, and outcomes.

Results: We selected 20 contextual factors and identified practice facilitation strategies and mapped them to the IRLM. Cronbach's alphas of contextual factors in the five domains (Intervention characteristics, outer setting, inner setting, characteristics of individuals, and implementation process) are 0.71, 0.82, 0.72, 0.89, 0.86, respectively. We used structural equation modeling to analyze the relationships among contextual factors, latent variables, practice facilitation strategies (Doing Tasks, Project Management, Consulting, Teaching, and Coaching), and outcomes (number of implemented QI interventions and Change Process Capability Questionnaire (CPCQ) score). All five facilitation strategies had statistically significant associations with the implementation of QI interventions (all P < 0.05).

Conclusions: The combination of ML and the theory behind the IRLM can be used to identify relationships between inner and outer context determinants and implementation strategies and study outcomes in pragmatic research study datasets. All the proposed strategies in H3 were statistically associated with completed QI interventions; and the strategies had more impact on the implementation of interventions than CPCQ change. By understanding the relationship between outcomes, practice determinants and coaching strategies, practice facilitators can better help primary care practices adapt and implement interventions and build capacity to adapt to change.

Background: Unhealthy alcohol use is a public health problem with significant health, social and economic impacts. Alcohol screening and brief intervention (ASBI) in adult primary care is an evidence-based approach enabling early identification and intervention of unhealthy alcohol use. However, large-scale implementation and sustainment of ASBI in routine clinical practice remains a challenge, and little is known about its population-level impact. Using electronic health record (EHR) data in a large integrated healthcare system in Northern California that implemented systematic ASBI in adult primary care in mid-2013, this observational study examined: 1) trajectories of ASBI performance over 5 years post systematic implementation, and 2) their associations with both later ASBI performance and alcohol use outcomes.

Methods: Using the health plan's EHR data, we calculated annual screening rates of adults with a primary care visit, and brief intervention (BI) rates among those with a positive screen (i.e., reporting alcohol consumption exceeding the age and sex specific daily and weekly low-risk National Institute on Alcohol Abuse and Alcoholism guidelines), for 57 medical facilities from years 2014 to 2021. We conducted latent class growth analysis using annual screening and BI rates to characterize trajectories of ASBI performance from years 2014-2018. Multivariable mixed-effects models were fit to examine the associations of ASBI performance trajectories with later ASBI performance and facility-level alcohol use outcomes.

Results: Three distinct screening performance trajectory groups (low-, middle- and high-performance) and four distinct BI performance trajectory groups (low-, improving-, middle- and high-performance) were identified. Facilities in the low-BI-performance group had panels of patients living in more deprived neighborhoods compared to the other 3 BI performance groups. After accounting for repeated measures and adjusting for time and patient panel characteristics, we found that screening and BI performance trajectories during 2014-2018 were significantly associated with screening and BI rates 2019-2021, respectively. We also observed a steeper decline in percentages reporting "exceeding daily drinking limits" and "having 5 + binge drinking days" over time among patients of facilities in the improving- and high-BI-performance groups.

Conclusions: Early success in ASBI performance is associated with long-term sustainability and may be associated with long-term population-level drinking outcomes.

Background: Implementation science has become increasingly important for improving uptake of healthcare innovations, which typically involves a broad range of stakeholders. The Acceptability of Intervention Measure (AIM), Intervention Appropriateness Measure (IAM), and Feasibility of Intervention Measure (FIM) are generic and adaptable outcome measures to assess the implementation of innovations across various settings and populations. However, their use in Denmark requires translation into the Danish language and a cross-cultural adaptation into the Danish healthcare context.

Methods: The study aimed to translate and cross-culturally adapt the AIM, IAM, and FIM for use in Danish healthcare settings. The translation process followed Beaton's guidelines, encompassing six stages: translation, synthesis, backward translation, expert committee review, pretesting, and appraisal of the adaptation process. Both quantitative (questionnaires) and qualitative (interviews) methods were applied during pretesting to evaluate the Danish versions.

Results: All stages of the translation and adaption process were completed. Linguistic challenges were identified, such as ensuring distinction between items, but they were resolved during the expert review. Pretesting with 33 Danish healthcare professionals showed that items were generally clear and relevant, but some overlap between AIM, IAM, and FIM items caused confusion. For example, IAM item 4 ("… seems like a good match") was difficult to interpret, leading to missing responses, and FIM item 3 was revised to improve clarity.

Conclusions: The translation and cross‑cultural adaptation, including pretesting, of the AIM, IAM, and FIM resulted in Danish versions that maintained conceptual alignment with the originals. While additional evaluation across interventions, contexts, and practices will strengthen the evidence base, the current versions already provide a practical tool for assessing implementation outcomes in Danish healthcare contexts.

Background: Electronic health record (EHR) systems have been used to support the implementation of evidence-based care. Growing evidence suggests that EHR systems can also support de-implementation of low-value care. However, a review of this literature has not been conducted. This scoping review will: 1) summarize how EHR-based interventions themselves have been used in primary care settings to de-implement low-value care, 2) summarize the effectiveness of these EHR interventions, 3) describe de-implementation strategies and outcome measures that have been used, and 4) describe facilitators and barriers that influence EHR-based de-implementation interventions.

Methods: We conducted a search using MEDLINE, CINAHL, Embase, and Web of Science on January 19, 2024 for peer-reviewed papers on EHRs and de-implementation in primary care. We inductively developed themes of how the EHR was used to support de-implementation. We mapped de-implementation strategies to a previously published taxonomy on implementation strategies, de-implementation outcomes to a previously published taxonomy on these outcomes, and facilitators and barriers to the Consolidated Framework for Implementation Research. We stratified study findings by EHR intervention type.

Results: We included 50 studies. EHRs supported de-implementation using four intervention types: 1) EHR alerts, 2) order sets and preference lists, 3) documentation templates, and 4) communication tools among the care team. The proportion of studies that showed favorable effectiveness in reducing low-value care ranged from 16.7% (communication tools) to 50.0% (documentation templates). Common strategies to support EHR-based de-implementation interventions included auditing and providing feedback, conducting educational meetings, and distributing educational materials. Twenty-two studies reported some assessment of de-implementation outcomes. Most EHR intervention types had numerous multi-level facilitators and barriers identified.

Conclusions: This scoping review identified multiple EHR-based interventions that health systems use to support de-implementation and their effectiveness. Although promising, the evidence base is limited by the general lack of frameworks used for intervention development and de-implementation, unclear theoretical rationale to support the use of selected de-implementation strategies, and the unclear validity of de-implementation outcomes used. Additional research is needed to develop and validate frameworks and outcomes for de-implementation to strengthen the evidence base.

Trial registration: None.

Background: Relationships are foundational to successful implementation of innovations in healthcare. In genomic medicine, multidisciplinary teams with good communication are most effective to provide safe genomic care; however, working together could be challenging due to the distinct work culture, worldviews, and clinical approaches held by different professional groups. In this paper, we explored the various strategies used to build relationships and foster collaboration as part of a Change program that supported the use of genomic testing and counselling in specialty areas.

Methods: Qualitative interviews were conducted with 36 participants across 3 professional categories (genetic counsellors, medical specialists, and nurses/allied health workers) to ask about their experiences working together in innovative models of genomic care across 7 clinical specialties. Data analysis was conducted through a two-staged inductive and deductive coding process: firstly to identify the categories based on the attributes of the Relational Theory and then coded against the Theoretical Model for Trusting Relationships and Implementation (the 'Model').

Results: Eight out of nine strategies to build/strengthen relationships described in the 'Model' were identified in the interview data. They included three technical strategies and five relational strategies. The inter-connections were present between relational and technical strategies, as well as within the relational category, whereby some served to reinforce one another. Two additional strategies emerged from the interview data but were not included within the 'Model,' including: negotiating boundary work and accepting differences used at inter-professional level. Specifically, genetic counsellors either reconstructed the professional boundary by taking over tasks beyond their role or adopted a boundary-preserving strategy to balance the social order within the team.

Conclusions: Our study highlights how relationship-building strategies can be leveraged in genomic multidisciplinary teams and can inform decisions about creating conditions that promote positive relationships and relational competence, ultimately leading to successful implementation of innovations into organisations/systems.

Background: The Normalization Process Theory (NPT) is increasingly used for evaluating and understanding implementation processes of complex care interventions. One key tool for applying the NPT in research and practice is the NoMAD questionnaire, which offers a structured approach to examination of the four constructs that according to the NPT are central in implementation and normalisation processes. We aimed to evaluate the psychometric properties of the Swedish version S-NoMAD.

Methods: Secondary analysis was performed on pooled S-NoMAD survey data from six implementation studies in different health and social care contexts. The NPT factor structure was tested by confirmatory factor analysis (CFA). Internal construct reliability was tested using Cronbach's alpha. Validity was confirmed by assessing the fit of the CFA using the fit measures Comparative Fit Index, Tucker-Lewis Index, root mean square error of approximation and standardised root mean square residual. Pearson correlations amongst the latent construct and general questions about the intervention were calculated.

Results: The estimation results of the CFA indicate that the four-factor model implied by the NPT fits the data reasonably well. The factor loadings are of good sizes and the fit indices do not imply a mis-specified model. A good internal construct validity, indicated by a good model fit to the NPT four-construct model and acceptable to good internal reliability, was shown. External validity was also demonstrated.

Conclusions: The CFA results indicate that the S-NoMAD has good psychometric properties for capturing perceptions of people involved in various Swedish implementation studies conducted in both health and social care contexts, demonstrating its general applicability. They show that the S-NoMAD, unlike the majority of instruments for evaluation of implementation processes, is not context- and intervention-specific. The findings highlight the utility of the S-NoMAD and show that it meets some important criteria for pragmatic measures. Further studies are warranted on different interventions implemented in diverse contexts regarding the meaning of the magnitude of the NoMAD scores in order to clarify its potential value as a tool for assessment of implementation strategies and on psychometric properties beyond construct validity and internal construct reliability, for example on test-retest reliability and longitudinal studies focusing on responsiveness.

Background: According to phasic models of implementation, a Preparation phase designed to enhance the implementation climate should be completed prior to the Implementation phase. Yet preparatory activities and outcomes are rarely reported or assessed in implementation research. Project MIMIC (Maximizing Implementation of Motivational Incentives in Clinics) was a hybrid type 3 effectiveness-implementation trial that compared two multi-component, phasic strategies to implement contingency management (CM) in opioid treatment programs. The current secondary analysis assessed the comparative effectiveness of the two strategies on 5-month Preparation phase outcomes: attainment of knowledge and fidelity benchmarks, implementation climate at the end of the Preparation phase, and time required for providers to complete the final preparatory/pre-implementation activity of enrolling and scheduling their first CM patient.

Methods: Twenty-eight opioid treatment programs and 186 staff were cluster-randomized to receive the Addition Technology Transfer Center (ATTC) control strategy (didactic workshop + performance feedback + consultation) or the theory-driven Enhanced-ATTC (E-ATTC) experimental strategy. During the Preparation phase, the E-ATTC strategy consisted of the ATTC strategy plus monthly Implementation Sustainment Facilitation sessions rooted in principles of team-based motivational interviewing to cultivate a strong implementation climate and accelerate successful completion of the Preparation phase.

Results: Across the 28 OTPs and 186 staff, attainment of knowledge and fidelity benchmarks favored the E-ATTC but did not differ significantly by condition. Implementation climate ratings after the Preparation phase were high in both conditions with no conditional differences. Providers randomized to E-ATTC completed their final preparatory activity at significantly higher rates than those randomized to ATTC. Cox regression revealed that receipt of the E-ATTC strategy was also associated with significantly faster completion of the final Preparation activity.

Conclusions: Consistent with hypotheses, the theory-driven implementation strategy was associated with higher levels of and faster time to completion of preparatory activities, a key indicator of readiness for implementation. Counter to expectations, this was not driven by differences in implementation climate. High ratings of implementation climate at baseline limited our ability to detect change over time, highlighting a need for alternate strategies to measure putative mechanisms of change. This analysis adds to the scant literature reporting Preparation phase strategies and outcomes, which are strong predictors of successful implementation.

Trial registration: This study is registered in Clinicaltrials.gov (NCT03931174).