Quality Improvement and Research: Can They Work Synergistically?

Quality improvement (QI) and research are both essential and necessary for advancing patient care. Unfortunately, they are often misinterpreted for one another, which can lead to ethical, practical, and regulatory challenges [1]. Despite these two activities serving distinct purposes, they can work synergistically and form part of a translational continuum of evidence-informed practice and innovation [2]. It is not uncommon that QI reveals questions that are worthy of formally researching. This article will explore what QI and research are, and how they can be misinterpreted. This will be followed by a brief discussion on how they can co-exist, using a case study focusing on the management of compromised first permanent molars (cFPM), to demonstrate this. Quality improvement is an overarching term used to describe a systematic continuous approach to improving healthcare outcomes. Once a problem is identified, QI processes work towards findings solutions that aim to improve service provision, and ultimately patient outcomes [2]. Rather than a single method, QI methods include audit, service evaluation and quality improvement project (QIP). Research, in contrast, is hypothesis-driven, methodologically rigorous, and designed to generate findings that are generalisable beyond the local context [3]. Research requires ethical approval (rather than local oversight in QI) and is governed by strict standards to ensure their reliability and validity. The three QI approaches and research will now be discussed in more detail. Audits are well recognised in healthcare as a method that measures current practice against a defined gold standard, or accepted benchmark. In essence, they compare what is happening against what should be happening to assess whether a standard of care is being met. They are often simple to carry out but run the risk that if the standard is outdated, vague, or not evidence-based, the audit loses relevance. Key criticisms of audits are they often have a very narrow focus and are cyclical processes, usually requiring multiple re-audits to demonstrate and evaluate change. This requirement can often lead to superficial compliance rather than any meaningful improvement [2, 4]. Similarly, they are rarely useful beyond the local setting, limiting their remit. Service evaluations are used to show how a service is performing, helping understand how effective, acceptable or efficient current services are. Like an audit, they focus on describing what is happening in current practice; however, differ in that they do not compare outcomes against a gold standard. The lack of a re-evaluation is a major drawback, in that service evaluations can identify a problem but might not actually solve it. Implementing change is challenging, as a new service evaluation would be required to fully assess any improvement. In addition, service evaluations are very context and location specific, meaning they lack generalisability [2, 5]. A QIP attempts to make measurable improvements to patient care within a specific context and setting. QIPs are often action-orientated and use iterative approaches, for example Plan-Do-Study-Act, to rapidly test and refine changes in clinical practice [2, 6]. Adopting iterative methods permits continuous evaluation of change and allows an adaptive response to unforeseen obstacles and unintended effects upon implementation [4, 6]. Therefore, QIPs are able to negate the issue of a ‘single-bullet’ intervention, or change to practice, that is usually the outcome of an audit or service evaluation [6]. However, QIPs are not the panacea of all QI methods. Clinicians can be resistant to doing a QIP as they are more time-consuming and sustaining improvements in an iterative way can easily be lost in busy clinical departments [2]. Similarly, they lack generalisability given how context specific they are [6]. However, if completed well, and maintained, they have the potential to make significant changes to a department over a sustained period of time [2]. Research is the attempt to derive generalisable, new knowledge and test a hypothesis to expand on what is already known. The ability to create new understanding and providing answers to questions that have not yet been explored is the major difference between it and QI approaches. Research studies follow rigorous and robust methodologies that collect and analyse data to permit conclusions to be drawn. They require strict peer-review and governance processes, with formal ethical approval needed in almost all cases [3]. They have the potential to have good external validity, with the findings being applicable to other populations or settings; however, if not careful in the design, this can be a limitation. Undertaking research can be challenging. They are expensive and time-consuming. Studies often take months or even years to complete, causing significant delays to implementing findings into daily practice [7]. Despite the differences, both QI and research share a commitment to evidence-informed practice and improved patient outcomes. An overview of these approaches, using cFPM as an example, is shown in Table 1. Increasingly, healthcare recognises the value of integrating QI and research in a translational continuum. Research and QI approaches are frequently conflated. This could have significant ramifications given the ethical and legal requirement differences between the two. As described above, research tests a new idea and aims to generate generalisable evidence [3, 5]. Any project that meets these criteria is research. Labelling research as ‘QI’ incorrectly is relatively common [2, 5]. It is likely to be naivety; however, it can be deliberate to avoid the ethical and regulatory scrutiny required in research studies [5]. Any project that tests a novel intervention (e.g., success of restorative material) or collects new knowledge from patients (e.g., opinions on discolouration following placement of silver diamine fluoride) is research, and should be treated as such. Incorrectly labelling a research project as a QI initative runs the risk of ethical breaches, especially if patient data is used without appropriate consent. Similarly, findings may be invalid as the methods employed are likely not to be as robust and rigorous as they should be. In contrast, research can be labelled incorrectly, when in fact it is QI. In these cases, it is often a mistake; however, it could be that clinical teams wish to promote the project as research, instead of QI, as it seems more prestigious or publishable [5]. Misinterpreting QI as research is of less concern than the converse. However, in doing so, the iterative nature of QI approaches might be lost or unnecessary delays are result from the often protracted ethical review and regulatory approval processes [3]. There are some simple ways to avoid misinterpretation. The easiest method is to use established decisions tools. The UK's Health Research Authority provides an online decision aid to determine if a project is research, audit, or service evaluation [8]. Alternatively, teams can seek early advice from local research and development departments to help determine the correct classification. In reality, QI and research regularly co-exist, often complementing one another in a translational loop (see Figure 1) [2, 4, 5]. QI initiatives help identify research-worthy questions and/or provide insights, grounded in clinical relevance, that help shape the design of the research study. In converse, research findings which lead to evidence-based changes of practice [7] can then be adapted and tested locally through QI approaches, ensuring uptake and sustainability [2, 7]. In practice, translating evidence is challenging in dentistry with variable and complex health conditions and diversity of practice settings being cited as the main obstacles to implementation [7]. Despite this, when done well, the co-existence can promote faster translation of evidence into practice, supporting the continuous improvement and innovation across healthcare systems [1, 2, 7]. The question of how best to manage cFPM—whether to restore or extract—was identified as a key clinical challenge, leading to the initiation of the DECIDE project. A combination of QI approaches were undertaken to (a) establish whether the acknowledged direct impacts associated with cFPM in children were being discussed as part of the patient assessment process [9], and (b) what treatment options were being discussed and offered. The results of these QI initiatives highlighted the commonly reported impacts (pain, eating, sleep loss and daily activities) were mostly being recorded [9]; however, there was variation and lack of consensus in the management options being used in the local context. While useful, the QI approaches highlighted unresolved clinical questions. Following the translational loop, shown in Figure 1, a follow-on research study was designed and undertaken to ascertain how cFPM were being managed across the United Kingdom by general dentists and specialists in paediatric dentistry. The findings identified substantial differences between, and within, these professional groups [10]. The findings from this research were unable, in isolation, to inform a change in clinical practice. As a result, a wider programme of research (funded as part of an NIHR doctoral research fellowship) was undertaken to (a) establish adolescents' and adults' views and experiences around managing cFPM using semi-structured interviews, (b) elicit the public's preferences for managing cFPM, including determining societal willingness to pay (WTP), using a discrete choice experiment and (c) determine the most efficient way of managing cFPM over the lifetime of a patient using mathematical modelling [11]. The findings of these studies are reported elsewhere [12-14]; however, it was clear that making the decision to restore or extract is not binary. Instead, the need to make shared decisions jointly between the patient, parent and healthcare professional for cFPM became abundantly clear [13-15]. Despite disseminating this message through publications [13-15] and regional and international presentations, it was felt at a local departmental meeting that shared decision making was not routinely being undertaken for cFPM. As part of a QIP, following PDSA principles, ‘planned’ baseline data showed that shared-decision making was not common practice. For the ‘Do’ stage, age-appropriate information is currently being developed by colleagues at Newcastle Dental Hospital for both children and parents to help facilitate this process. Data will be collected to ‘study’ whether the intervention was effective, before ‘acting’ on what to do next. This QIP will continue whilst a further research study looking to explore how these interventions can be embedded into a decision-aid to support decision-making of cFPM across multiple settings will then follow. The isolated case study demonstrates the intricate and intimate relationship QI and research can have. Despite QI and research often being misinterpreted for one another, they can and should co-exist as part of translational continuum of evidence-informed practice. Dr Greig Taylor is funded by an NIHR Clinical Lectureship. The views expressed are those of the author(s) and not necessarily those of the NIHR or the Department of Health and Social Care. I would like to thank my colleagues, Rachel Osborne and Orion O'Brien, for allowing me to share the ongoing work of developing age-appropriate leaflets for cFPM due to molar-incisor hypomineralisation. The author declares no conflicts of interest.