Why Quality Improvement Must Be Reformed

Introduction

Healthcare quality has been a hot topic for several decades now. With constant ongoing research raising the bar for what qualifies as “good care”, ensuring that a hospital or clinic has healthcare processes that can adequately assess, improve and implement changes to existing processes becomes exceedingly difficult.

The Basis for Quality Improvement

Quality improvement literature – the basis for today’s best practices - often builds on established principles created for the manufacturing industry such as Lean Six Sigma _[1]. It is thanks to quality improvement processes that multi-disciplinary care teams evolved to measure and control institutional standards, in addition to ensuring consistency of application for every patient _[2]. Interventions like DVT prophylaxis, avoiding catheter-based infections and adequate utilization of non-pharmacological interventions like incentive spirometers and making sure a patient ambulates daily can all be measured and monitored. Systems that were created based on a single provider’s ability to be responsible for all aspects of care are consistently outperformed by multi-disciplinary cohesive teams and their ability to provide evidence-based best practice care. Creating robust multi-disciplinary teams that can be held accountable for various aspects of care are one of the biggest contributions of quality improvement systems to modern medicine_{[3, 4]}.

Systems that were created based on a single provider’s ability to be responsible for all aspects of care are consistently outperformed by multi-disciplinary cohesive teams

Limitations of Quality Improvement

In spite of decades of quality improvement research, there are many limitations to how much quality improvement teams can improve patient care in a given hospital _[5].Although within the context of a quality improvement study measurable changes in patient outcomes can be demonstrated, this rarely translates into real-life changes in patient care that is sustainable. The cited reasons for this limitation mention: a lack of quality improvement culture, inadequate power for the quality improvement (QI) team to enforce a change, inadequate training or skills of the quality improvement team itself, the “magic bullet” approach to QI interventions and the “project based” nature of quality improvement studies that focus on one small intervention at a time without an ability to change hospital-wide processes _{[2, 5]} .

Quality improvement itself needs to be improved.

This article aims to outline another dimension that often limits quality improvement studies, which is the failure to study an aspect of patient care within a context of the entirety of patient care.  The external validity of most quality improvement studies is thus flawed because the conditions within the study are necessarily different from the less controlled and less scrutinized environment of usual patient care.

The external validity of most quality improvement studies is thus flawed because the conditions within the study are necessarily different from the less controlled and less scrutinized environment of usual patient care.

For a typical QI study, the team often isolates a process that needs to be modified to generate a measurably better outcome (Figure 1). Looking at this process in isolation encourages the QI team to identify a “process owner” which is the provider responsible for this task, and a “process outcome” which is a measurable intervention that the patient needs to receive. This framework limits the field of view of the QI team in such a way that interventions often don’t take into account other processes the healthcare provider is responsible for. For example, providers can experience “alert-fatigue” with alerts from an electronic medical record placing a limitation on how much this intervention can be used to aid work flow_[6].

Figure 1: Lean Process map designed for quality improvement. The output of this process may or may not be an outcome directly relevant to the patient. This while perfectly suited to evaluating processes in sito (under experimental conditions), might fail to capture how this process impacts other issues of potentially high value to the patient. Eg. Ordering and giving DVT prophylaxis on admission can result in patient discomfort due to pressure stockings or injections that they feel are unnecessary. Failure of a provider to discuss the risk-benefit of this intervention might negatively impact a patient. Trying to have this discussion with a patient who is still anxious about their diagnosis or treatment plan may be perceived as poor quality care that is not prioritizing a patient’s values. This process when implemented well can yield to metrics that look good but may not necessarily be associated with good care for the patient or their satisfaction. In fact, patients can lose trust in healthcare teams whose priorities are not aligned with their own during any given phase of care.

A holistic approach to evaluating and improving quality of health care delivery remains the elusive holy grail of sustainable and effective quality improvement. Some important work has been done in this regard, with recent literature describing the need for better contextualizing the determinants of quality rather than focusing on numerous quality improvement interventions individually_[2]. One such determinant is the efficiency of healthcare, which is quite difficult to measure. Systematic reviews of efficiency measures in health care found that there is poor validity and standardization across health care institutions. Furthermore, they are not always linked to quality of care delivered and sometimes merely measure cost rather than actual efficiency _{[2, 7]}. Adding to the complexity, the diversity of healthcare systems globally makes it more challenging to simply import methodologies from another system. What constitutes “good enough” care is highly impacted by available resources, expertise and the needs of the local population.

The diversity of healthcare systems globally makes it more challenging to simply import methodologies from another system.

Contextualizing Quality Improvement Projects

To aid quality improvement teams better contextualize the processes, we propose referring back every process to a patient centered diagram. Evaluating patient-related process in this way can help contextualize the impact and meaningfulness of a process being addressed. This can help providers and process owners make better decisions about what is a top priority in real time and avoid out of place alerts or stops that many quality improvement studies use to improve a process evaluated in situ (Figure 2).

Figure 2:  The patient as the process main stakeholder. The spokes of the sun are the different components of healthcare delivery that the patient can evaluate and is directly affected by. Various improvement processes can be mapped out to one of the spokes of the wheel and incorporated with clarity into a bigger process that impacts the patient directly and possible batching or “stops” can be placed to make sure that things occur in an order that makes sense to the patient and is considered valuable by them.

A similar diagram is required for any provider that is involved in a process being improved. Rather than look at a provider as a process owner only, we acknowledge that a provider is instrumentally involved in a large number of processes for different patients and possibly even on a systems level. Mapping the workflow of highly functioning providers like doctors can lead to much more meaningful evaluations of how systems can be optimized to serve complex providers that wear as many hats as physicians do (Figure 3).

Figure 3: Provider centered diagram. A typical list of processes a hospitalist is involved in. Other specialties have similar complexity of responsibilities. For example, surgical specialties have operating theatre duties and similar care coordination responsibilities. Patient’s complexity may vary on average, but the overall cognitive load required for all physicians is comparable. The central chaotic shape was chosen to reflect that each of these processes can occur to varying degrees from day to day but the physician would be considered the process owner for all of them as they occur in parallel.

The phases of patient care described in Figure 2 apply to every patient. They can all occur together for patients who are discharged from an emergency department or spread throughout a hospital stay. The phases of clinical care can vary geographically or across care teams as we outlined in our article “Challenges in Designing Excellence Oriented Hospital Systems – Part 2”. Mapping any quality improvement effort to the phases of patient care can add necessary context and offer opportunities to bring to the center the patient’s values and needs before confirming the efficacy of a quality improvement effort.

Quality improvement needs to be both patient and provider centered

It would be remiss to examine hospital processes without honestly taking the needs of providers into account. Concerns about physician suicides increased wellness interventions in hospital systems such as mindfulness training, gym memberships and free counselling services, in most cases achieving the same success as free pizza which has turned into a meme about ineffective admin support that fails to address core issues _[8].

In an environment such as a hospital filled with high performing professionals, why is it so difficult to support these professionals in the jobs they spent over a decade qualifying to do? I propose that the problem is more related to impossible systems of practice that overburden physicians with high cognitive load tasks with processes that focus more on efficiency than supporting clinicians. To illustrate this, the work flows a clinician can be involved in must be examined. This should begin to give perspective on the complex decision making and high cognitive load clinicians spend eight hours a day doing. A deeper understanding of these work flows can aid QI teams in addressing structural, digital and staff support issues to help clinicians perform optimally and sustainably.

Concerns about physician suicides increased wellness interventions in hospital systems such as mindfulness training, gym memberships and free counselling services, in most cases achieving the same success as free pizza which has turned into a meme about ineffective admin support that fails to address core issues

Figure 3 demonstrates why neat QI interventions can break down when they target behavioral change for a physician. The competing needs on a physician’s attention immediately show the challenges to making changes to behavior that do not keep in mind the overall work-day of a physician. The past decade has witnessed increasing complexity of medical decision making, increased complexity of billing, increased complexity of documentation requirements either for billing or medico-legal reasons and increased care coordination duties as multidisciplinary teams became larger in addition to ever changing standards of medical practice. Patients have also become more informed and expect (also deserve) shared decision making for all aspects of care. We can immediately see how delivering high quality care is impossible without appropriate support for the physicians making all these decisions in addition to patient counseling and care coordination.

Conclusion

Reform is needed in quality improvement departments such that a broader context is taken into account when attempting to implement a change. Although much more challenging, failure to adequately take into account the patient’s requirements as well as the clinician’s needs will continue to lead to ineffective interventions that fail to stand the test of time beyond the QI study itself. Future articles will dive more deeply into the nuances of transitions of care and how this adds a layer of complexity to patient care even beyond the needs of the patient and the provider.

• Amaratunga T and Dobranowski J 2016 Systematic Review of the Application of Lean and Six Sigma Quality Improvement Methodologies in Radiology J. Am. Coll. Radiol. 13 1088-1095.e7
• Walshe K and Freeman T 2002 Effectiveness of quality improvement: learning from evaluations BMJ Qual. Saf. 11 85–7
• Taberna M, Gil Moncayo F, Jané-Salas E, Antonio M, Arribas L, Vilajosana E, Peralvez Torres E and Mesía R 2020 The Multidisciplinary Team (MDT) Approach and Quality of Care Front. Oncol. 10 85
• Epstein N E 2014 Multidisciplinary in-hospital teams improve patient outcomes: A review Surg. Neurol. Int. 5 S295–303
• Dixon-Woods M and Martin G P 2016 Does quality improvement improve quality? Future Healthc. J. 3 191–4
• Hussey P S, de Vries H, Romley J, Wang M C, Chen S S, Shekelle P G and McGlynn E A 2009 A systematic review of health care efficiency measures Health Serv. Res. 44 784–805
• Sinskey J L, Margolis R D and Vinson A E 2022 The Wicked Problem of Physician Well-Being Anesthesiol. Clin. 40 213–23
• Castro Lopes S, Guerra-Arias M, Buchan J, Pozo-Martin F and Nove A 2017 A rapid review of the rate of attrition from the health workforce Hum. Resour. Health 15 21
• Willard-Grace R, Knox M, Huang B, Hammer H, Kivlahan C and Grumbach K 2019 Burnout and Health Care Workforce Turnover Ann. Fam. Med. 17 36–41
• Nang Mie Mie Htun  null, Reyer J A, Yamamoto E, Yoshida Y and Hamajima N 2016 Trends in attrition among medical teaching staff at universities in Myanmar 2009-2013 Nagoya J. Med. Sci. 78 27–40
• Haddad L M, Annamaraju P and Toney-Butler T J 2022 Nursing Shortage StatPearls (Treasure Island (FL): StatPearls Publishing)
• Footracer K G 2015 Alert fatigue in electronic health records JAAPA Off. J. Am. Acad. Physician Assist. 28 41–2