Evidence-informed decision-making is a common concept in the health industry, based on integrating the most appropriate research evidence to make a decision. The process should also consider factors like its effect on equity, public opinion, the feasibility of the implementation, effectiveness, and stakeholders’ acceptability. This systematic decision-making approach applies to numerous processes, like identifying a high-priority issue and selecting the available interventions. EIDM can also be used to monitor implementation and evaluate the impact made. The concept focuses on making better decisions and effectively using scarce resources while avoiding harm, thus generally enhancing health. The evidence-informed approach to making decisions impacts sectors like health systems, clinical practice, and public health.
EIDM originated from the evidence-based medicine movement and was perceived as a scientific and systematic way of practicing medicine. Over the years, evidence-based medicine has been described as using the most appropriate current evidence to make informed decisions about caring for every patient while integrating clinical expertise. However, data-driven public health expands this systematic approach to cover the focus of the larger population while integrating the community’s preferences. The evidence-informed approach terms can, therefore, vary in the aspect of the specified area of practice. Emphasizing evidence-informed in place of the evidence-based aspect of making a decision shows that the process there are other factors beyond the proof of safety and effectiveness. Therefore, these elements must be considered in the decision-making process in an organized manner. In addition, the factors can be backed up with current research evidence.
Evidence-Informed Decision-Making Process
As a systematic and transparent approach to the decision-making process, EIDM involves several steps. The first action applied in evidence-informed decision-making is problem identification. It involves recognizing a high-priority issue and understanding the context and causes of a specific decision. For instance, the identified issue could be a risk factor of drug abuse within a community or a high prevalence of a specific medical condition within a particular region. In this step, one should focus on the most affected group regarding age, gender, ethnicity, or geographical area. Some of the research evidence that might explain the scope of the problem include routine health information, surveys, and surveillance and monitoring data. If possible, attempts should be made to find health disparities through disaggregated data. A good understanding of the determinants of the problem and the decision-making context is also essential in this step.
Once the problem is identified, the next step should be selecting an intervention by designing solutions. Depending on the procedure and outcome, a specific health concern may consider a single choice or solution or several possibilities. For instance, while evidence-based practices may comprise three or more interventions, health technology assessment might only include a single medication or technology. However, the procedure is similar as they entail identifying questions and the eligibility criteria using population, intervention, and comparator (PICO). The third step applied in evidence-informed decision-making is designing implementation considerations. Once the decision to apply an intervention is taken, practical criteria are used to choose interventions and build the deployment strategy. The plan is further informed by identifying the barriers and facilitators to the process and assessment for research evidence of the solutions. Lastly, the considered executions are then monitored, and their impacts evaluated. During this EIDM process stage, data is collected systematically to determine if the intervention is executed based on the stakeholders’ expectations. In addition, a systematic assessment is made to identify the intervention policy’s impact, efficiency, and relevance.
Searching and Retrieving Appropriate Research-Based Evidence
The Evidence Funnel
Searching and retrieving the most relevant research-based evidence is often guided by the evidence funnel, which involves three significant levels. The first level is evidence inquiry, which is carried out in primary research and involves collecting individual data. At this level, the unit of analysis is the individuals. In addition, the level includes case reports, randomized control trials, surveys, case series, quasi-experimental studies, and cross-sectional studies. In addition, primary research also includes qualitative research like focus group discussions or interviews. This level consequently feeds into the second level of the evidence funnel, evidence synthesis, which exists in secondary research. This type of research uses the already collected data in primary research. Therefore, primary studies are this level’s significant unit of analysis, with systematic reviews used to address numerous research questions. Lastly, at the tertiary research level, evidence products are used to summarize the results in the secondary research. These products are elements of connecting research to action and include health technology assessments, guidelines, patient decision aids, and evidence briefs for policy.
Types of Research Evidence
In the EIDM process, different questions require different types of evidence. However, prior to selection, the quality of the evidence has to be determined. The study design is one of the traditional frameworks for understanding research evidence, with various designs placed in a hierarchy based on declining internal validity. The first level of evidence includes systematic reviews and randomized controlled trials. This level is often followed by the second level, which involves quasi-experimental studies and their reviews. The third level includes a non-experimental review of randomized control trials and systematic qualitative reviews with or without meta-synthesis. This research evidence is followed by level four which involves consensus panel reports, editorial, and expert opinion based on scientific evidence. The last and lowest level involves literature reviews, case reports, and other recognized experts’ opinions based on experimental evidence. This suggested hierarchy of quantitative evidence reveals the study’s biases, with the possibility of bias being higher moving down the hierarchy.
Key Features and Methods Associated with RCTs, Survey Designs, and Qualitative Research
Qualitative research methods focus on obtaining data that answers the research questions on how and why phenomenon. As opposed to quantitative research, the design gathers information in a non-numerical, written format. This ensures that the information provided is sufficient and that analysis is done based on participants’ answers, as opposed to making assumptions. Qualitative research designs include historical studies, phenomenology, grounded theory, and action research. These designs are covered through focus groups, case studies, oral history, and observation.
RCTs
Randomized control studies are designs of studying which involve comparing two or more interventions by randomly assigning participants to a control or experimental group based on the variables of the research. RCTs are based on various essential features and principles. First, RCTs apply randomization, whereby participants are randomly assigned to one of the two groups in the trials, thus reducing biases. Randomized control trials use either single or double blinding to minimize participants’ ability to know that they are receiving an intervention. In addition, randomized control trials’ significance of the results is statistically determined based on a predetermined algorithm.
Survey Designs
Survey design techniques gather research data by presenting questions to a predefined group of participants. The method can collect qualitative and quantitative data. This is enabled through the use of open-ended and closed-ended questions. Survey designs can either be in paper form or online form. They contain structured questions requiring respondents to provide their answers based on personal experience. The research design’s nature allows a researcher to collect a huge amount of data within a short time.
Sources of Bias in RCTs, Survey Designs and Qualitative Research.
Research bias occurs when a researcher introduces a systematic inaccuracy in the sample data, thus distorting the research outcome. It might also occur when a researcher’s taste and opinion impact the study. The inaccuracies divert the research from its genuine conclusions to proving the researcher’s expectations. There are various potential causes of bias in research. For instance, in the RCTs, when participants are aware that they are part of the active group, their positive expectations may affect the results of the trials. Such prejudice is known as performance bias. Another significant type of bias in most research is information bias. This type of prejudice occurs in the recall, data collection, and recording and involves missing data.
In the RCTs, participants can drop out of the study before it ends. However, if the researcher fails to include the participants who opted out, it may lead to attrition bias. Another common type of prejudice is selection bias, which happens when there is a difference between the selected participant groups, in the aspect of the demographic data, especially in RCTs. A research study may become biased if the experimental variables influence the control variables. This type of prejudice is known as confounding bias. In addition, research may contain reporting bias if there exists a systematic difference between the reported research information and the unreported data. However, the potential study biases can be identified prior to conducting the study through the hierarchy of evidence, thus enhancing prevention measures.