Article type
Abstract
Background
A gap exists between research evidence and implementation of that evidence in real-world settings at the population level. Computable evidence in the form of Fast Evidence Interoperability Resources (FEvIR) helps matching individual patient data with evidence containing effect estimates for outcomes of interventions to find people who would benefit most.
A regional quality improvement and benchmarking network has proved promising in implementing and measuring change on a practice level. Population health tools and clinical decision support (CDS) tools help prioritize interventions for people from the whole population.
Objectives
Outreaching and improving the care of people at the highest risk for adverse cardiovascular outcomes via a regional quality improvement network and population health tools used by physicians and nurses.
Methods
Electronic health record data from a population of 25,000 were extracted for real-time clinical CDS and into a population health dashboard. The data were processed using risk prediction and decision support tools, allowing identification of and outreach to individuals in need of interventions. Workshops of physicians and nurses under the umbrella of a regional quality network were organized to set 1-2 quality improvement goals at a time for each practice and to monitor progress. Patients were involved by providing information on benefits and harms of interventions and by discussing and adjusting the relative importance of outcomes when choosing interventions.
Results
During the first implementation round, average low-density lipoprotein cholesterol in the target population was reduced by more than 0.5 mmol/L, and average systolic blood pressure by more than 5 mm Hg. The numbers of people avoiding major coronary event, stroke, or heart failure were estimated on the basis of these intermediary effects. Comparative effectiveness of the interventions was estimated by calculating net effects by using a rating of relative importance of outcomes.
External evaluation by a national quality register showed that progress in the region was more rapid than the national average.
Conclusions
The evidence ecosystem using computable evidence was implemented locally by health care professionals, with equitable outreach of people who would benefit most, iteratively defining and monitoring reachable goals. The methods can be applied to any populations and interventions.
A gap exists between research evidence and implementation of that evidence in real-world settings at the population level. Computable evidence in the form of Fast Evidence Interoperability Resources (FEvIR) helps matching individual patient data with evidence containing effect estimates for outcomes of interventions to find people who would benefit most.
A regional quality improvement and benchmarking network has proved promising in implementing and measuring change on a practice level. Population health tools and clinical decision support (CDS) tools help prioritize interventions for people from the whole population.
Objectives
Outreaching and improving the care of people at the highest risk for adverse cardiovascular outcomes via a regional quality improvement network and population health tools used by physicians and nurses.
Methods
Electronic health record data from a population of 25,000 were extracted for real-time clinical CDS and into a population health dashboard. The data were processed using risk prediction and decision support tools, allowing identification of and outreach to individuals in need of interventions. Workshops of physicians and nurses under the umbrella of a regional quality network were organized to set 1-2 quality improvement goals at a time for each practice and to monitor progress. Patients were involved by providing information on benefits and harms of interventions and by discussing and adjusting the relative importance of outcomes when choosing interventions.
Results
During the first implementation round, average low-density lipoprotein cholesterol in the target population was reduced by more than 0.5 mmol/L, and average systolic blood pressure by more than 5 mm Hg. The numbers of people avoiding major coronary event, stroke, or heart failure were estimated on the basis of these intermediary effects. Comparative effectiveness of the interventions was estimated by calculating net effects by using a rating of relative importance of outcomes.
External evaluation by a national quality register showed that progress in the region was more rapid than the national average.
Conclusions
The evidence ecosystem using computable evidence was implemented locally by health care professionals, with equitable outreach of people who would benefit most, iteratively defining and monitoring reachable goals. The methods can be applied to any populations and interventions.