Article Text
PDF
Abstract
High-quality discharge summaries are essential for promoting patient safety during transitions between care settings. When the diagnosis list in the discharge summary is not accurate, the subsequent care provider will not have the latest medical history list and the care and safety of the patient will be compromised. Discrepancies in the secondary diagnosis capture rates have been identified in close to 30% of patients admitted to Sengkang Community Hospital (SKCH) during internal audits. Our project aimed to improve the rates of secondary diagnoses coding in the discharge summaries of patients who were admitted to SKCH using skills of change management in our interventions. Plan-Do-Study-Act cycles used in combination with change management skills led to the success of our quality improvement project. Remarkably, we managed to achieve close to 100% of the secondary diagnoses capture rate after a 5-month period.
- Patient safety
- Diagnosis
- Hospital medicine
Data availability statement
All data relevant to the study are included in the article or uploaded as online supplemental information.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Inaccurate diagnoses in discharge summaries compromise patient care and lead to poor transition between healthcare settings. Many quality improvement projects fail to sustain and spread due to challenges such as a lack of leadership support, limited stakeholder engagement and a poor commitment to continuous improvement.
WHAT THIS STUDY ADDS
By using a mixed-model change management approach in our quality improvement project, we managed to greatly improve our rates of completed secondary diagnoses, sustain the result and successfully spread the initiatives.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
The project highlights the importance of integrating models for change management with concepts commonly applied for improvement and implementation to achieve success, sustain changes and drive positive outcomes.
Introduction
Diagnoses have important implications for patient care, research and policy. High-quality discharge summaries are essential for promoting patient safety during transitions between care settings. The discharge summary is an important communication tool for promoting the quality, safety and continuity of care.1 When the diagnosis list in the discharge summary is not accurate, the subsequent care provider will not have the latest medical history list and the care and safety of the patient will be compromised. The patient will be at risk for negative health outcomes from the inaccurate clinical picture as clinical decision-making is tailored to a correct understanding of the patient’s health problem,2 leading to inappropriate management, poor quality of care and a higher risk of readmission.3,4 Moreover, when diagnoses are omitted, it affects health policy decisions that rely on diagnostic information, thus affecting research, surveillance, clinical audit, quality improvement (QI) hospital payment policies and resource allocation.5,6,7
The Institute of Medicine reports Best Care Lower Cost: The Path to Continuously Learning Health Care in America concluded that diagnostic and treatment options are expanding and changing at an accelerating rate, placing new stresses on clinicians and patients, as well as potentially impacting the effectiveness and efficiency of care delivery.8 With a rising number of patients with multimorbidities, healthcare professionals must deal with the increased administrative work process of keying in the secondary diagnoses to ensure that discharge summaries are comprehensive as a form of communication to the subsequent care provider. Internal quality audits have demonstrated that up to 30% of our discharge summaries do not have an accurate and comprehensive list of secondary diagnoses. As a community hospital, our patients are elderly with multimorbidities. Without an accurate list of diagnoses, the subsequent care provider, patient and their caregivers will not know the full extent of their medical problems with potential lapses in the medical treatment plans.
Successful QI work in healthcare is both a science and an art. Although QI holds promise for improving the quality of care and patient safety, hospitals that adopt QI often struggle with its implementation.9 An effective QI project not only requires an understanding of the methodology and science of improvement but also a mastery of the concepts of change management.10 Hospital leaders must ensure operational efficiencies through effective quality initiatives using skills from change management to improve patient safety and enhance staff satisfaction.
Our project aimed to improve the rates of secondary diagnoses coding in the discharge summaries of patients who were admitted to Sengkang Community Hospital (SKCH) using skills of change management in our interventions. Our mission statement was to improve the coding rates of secondary diagnoses in the discharge summaries of patients who were admitted to SKCH from 70% to 100% within 5 months.
Methods
This QI project was conducted in the clinical area of SKCH, based in Singapore. The clinical area of SKCH has 400 beds, comprising 12 wards, providing inpatient rehabilitation care, subacute care and palliative care to patients requiring transition of care back into the community. The mean age of all patient admissions is 75.8 years by the end of 2023, with an average length of stay being 26.4 days. The majority of admissions were mainly for rehabilitation. Our hospital operates in a multidisciplinary team setting composing of healthcare professionals from various specialties such as medical, nursing, allied health and administrative personnel. All domains collaborate to provide comprehensive care to patients.
In our practice, we do not have the support of clinical coders. Instead, to generate the diagnoses list, resident physicians abstract relevant information from patient’s medical records and decide which diagnoses are relevant for the episode of care. Medical diagnoses are manually identified and keyed into the electronic medical record by our physicians as part of the discharge process. These diagnoses are then linked to appropriate International Classification of Diseases -10 codes by the computer system. Hence, the QI team comprised only of physicians as this work process is solely executed by them.
The hospital’s electronic medical system, Eclipsys Sunrise Clinical Manager, was reviewed to extract baseline data on diagnoses coding in discharge summaries between October 2022 and March 2023. The QI team consists of seven physicians, consisting of two consultants and five resident physicians, working at the postacute and continuing care department at SKCH. A cause-and-effect analysis was performed to identify the root causes for the incomplete secondary diagnosis coding rates. Voting was then conducted on the root causes identified and these were then plotted on a Pareto chart. After two rounds of voting, according to the Pareto principle, three factors were identified to be contributing to 80% of the effect (figure 1). These factors were then selected for targeted interventions. Data were collected after each round of the Plan-Do-Study-Act (PDSA) cycle, which was used to assess improvement in coding rates and the impact of the interventions.
Pareto chart analysis. PDPA; personal data protection act; SCM; sunrise clinical manager; AOR, “at own risk”
Patient and public involvement
No patients were involved in this project.
Change management principles used in driving the solutions
Healthcare workers are a unique group of individuals. Unlike other industries, healthcare staff look after patients who are dependent on them. Medicine is a long tradition and people working in healthcare have fixed opinions on specific care practices, cultural beliefs and standards to uphold. This dynamic is very different from practices in other industries that are trained to be agile, embrace changes and continuously promote innovative measures. Common reasons for healthcare staff resisting changes are safety concerns with new practices, change affecting the psychological safety of staff, traditional patient care norms, fear of criticism and punitive actions when implementing changes and staff perceiving leaders and change model as patronising and having distrust in leadership.11 Therefore, change management is a critical component to the successful implementation of QI projects in healthcare, as it helps to successfully navigate transitions, adapt to new processes and achieve sustainable improvements. Studies have shown that organisational changes in healthcare are more likely to succeed when healthcare professionals have the opportunity to influence the change, feel prepared for the change and recognise the value of the change, including perceiving the benefit of the change for patients.12
Digital innovations, technological advances, changing disease patterns, developing medical technologies and ageing populations require healthcare organisations and professionals to change almost constantly.13 14 As the saying goes, ‘Change is the only constant in life’. Change management models are designed to act as compasses that help organisations and individuals navigate difficult transitions and guide them towards ensuring that changes are accepted and put into practice. For our QI project, we used a mixed-model approach, adopting basic principles from common change management models such as Lewin’s change model,15 Kotter’s eight-step theory,16 Deming’s cycle17 and Kübler-Ross Change Curve18 to leverage on best practices, tactics and strategies to facilitate a successful implementation of the change process. The various improvement tools and change methods that are used in this project are further detailed in the following paragraphs. They share many characteristics in common and were deployed to gain maximum benefit.
Results
A cause-and-effect analysis identified the root causes for the inadequate secondary diagnoses coding rates. The factors were broadly defined as physician factors and work process factors. Physician factors include uncertainty regarding the definition of secondary diagnoses and lack of time. Work process factors identified include a cumbersome coding work process on the electronic medical system.
On identification of the root causes, our team initiated the PDSA cycle and the first intervention was to educate all clinicians on the importance of accurate coding through formal departmental teaching sessions. There was an improvement of 10% from 78% to 88% postintervention 1 (figure 2). We introduced the second intervention, which was using a visually appealing reminder and nominated ward champions to advocate and promote the change. This led to a further 5% increase in the completion rate, bringing it to 92%. Finally, we collaborated with the IT department to create a simplified coding workflow to improve efficiency for the cumbersome coding process and for the sustainability of change (figure 2). We continue to collect data every 3 monthly after the completion of our QI project and it is well noted that we managed to sustain a rate close to 100% even until.
Run chart showing baseline and effect of interventions. SKCH, Sengkang Community Hospital.
Change management tools (a mix-model approach)
Creating a shared purpose and communicating the change vision (Lewin’s three-stage model of change and Kotter’s change model)
According to Lewin and Kotter, change must only happen when there’s a strong motivation to undergo it. The first step in our project is to use a sequential and systematic approach to change management to create a sense of urgency and form a strategic vision. We first created a shared goal and a sense of urgency by giving departmental briefings of what the current situation is, which is a gap between what is listed in our patients’ discharge summaries and what is the true reflection of our patients’ comprehensive medical conditions. Having a shared purpose is crucial because improvement efforts work best if there is an explicit connection between the change and people’s values. We aimed to mobilise the junior staff in the department and to gather support and commitment as they are the ones completing the discharge summaries. We also ensured that the senior doctors who vet the discharge summaries are aligned with the aim of this QI project to reach a shared understanding and unite the doctors at various levels towards a common goal.
PDSA improvement tool and Kubler-Ross change curve model
The Deming cycle, or PDSA, is an iterative process for solving problems and improving processes for continuous improvement. After two rounds of voting, according to the Pareto principle, three factors were identified as contributing to more than 80% of the effect (figure 1). All members of the QI team participated in both rounds of voting. The factors were classified into physician factors and work process factors. Junior doctors in the department were unclear about the need for a complete list of secondary diagnoses and some did not understand what secondary diagnoses entail. In a busy hospital setting with multiple admissions and discharges per day, the physicians did not think that creating a comprehensive and accurate list of secondary diagnoses was important. To address this, we used the Kubler-Ross change curve model to plan and implement our change strategy, as well as to communicate and engage our physicians. We customised our communication styles according to the emotions of the individuals to promote buy-in for the project. We initiated the PDSA cycle and the first intervention was to ensure that all clinicians understand the importance of accurate coding. Formal training, including case-based discussion during departmental teaching sessions, was held. Feedback sessions and surveys were conducted 1-month postinterventions to assess the understanding of the importance of complete secondary diagnoses to all staff in the department. For the second intervention, we used a visually appealing reminder which was placed in a prominent area of the staff work area and nominated ward champions to advocate and promote the change. This provides a constant and easy reminder for physicians to complete the secondary diagnoses list on discharge. From the feedback and surveys, the takeaway point was that while physicians are aware of the need, this is overlooked in a time-constrained environment which is what we have identified as a root cause as well.
Remove barriers to change (Lewin’s three-stage model of change and Kotter’s change model)
The work process factor was identified as a cumbersome process to search for and input the complete list of secondary diagnoses on the hospital’s electronic medical system. The characteristics of patients that our community hospital serves are those with multimorbidities and dependent on others for their activities of daily living. Hence, the complete list of secondary diagnoses is often lengthy. To address this, we collaborated with the IT department to create a simplified coding workflow to improve efficiency. By working with external stakeholders like the IT department, we were able to make our work processes more efficient. The results of the iterative cycles allowed the team to optimise and refine the coding work process on our electronic medical system. System-wide implementation of this refined work process greatly lowered the inertia to produce a comprehensive and accurate secondary diagnosis list and the physicians gladly adopted the new processes, promoting the success of this QI project. By tapping on advancement in technology, we can innovate and implement the best solution and reinvent standard work for a quality sustainable process.
Celebrating small wins (Lewin’s three-stage model of change and Kotter’s change model)
Celebrating small wins helps to fortify success, build momentum, increase team morale and enable the team to not only make progress continuously but want to make progress continuously. We did that by recognising and acknowledging individuals’ efforts during our team meeting. Celebrating someone else’s success acknowledges their accomplishments and their contribution to the greater goals. This allows for gratification and as a motivation for success. In our regular departmental meeting, our team took 5 min to celebrate the progress of our project, complimented the clinical team for their effort and recognised their contributions in producing an accurate and comprehensive list of secondary diagnoses for our patients on discharge.
Discussion
Change management frameworks are designed to make the changes easier to implement and, more importantly, to solidify the change as the new norm. We used the result of the 5-month PDSA to learn how users interacted with a reminder system and a refined work process. The iterative cycles produced ideas, comments and suggestions and allowed us to optimise the visually optimised reminder and electronic medical system work process before organisational spread.
Figure 2 shows the beginning of the PDSA cycle with subsequent interventions to demonstrate the success of our QI project. Remarkably, we managed to achieve close to 100% of the hospital discharge summary having a complete list of secondary diagnoses, with the most significant improvement occurring after removing the barriers to success by refining and optimising current work processes. Leveraging the right change management tools can help healthcare organisations attain support for major changes in the organisation.
At the end of every month, we gathered feedback and collected compliments from community providers on how the comprehensive list of secondary diagnoses enhanced patient care and promoted communications between care settings. The positive feedbacks and compliments act as ‘wins’ by serving as tangible evidence that our change process is yielding beneficial results. Over time, it builds momentum, boost morale and foster a culture of continuous improvement. Our team actively tracks and celebrates the small ‘wins’ on the way to our goal. We are pleased to report that the rates of our patients’ discharge summaries having a complete list of secondary diagnoses increased from 68% to 98% after the final intervention (figure 2). We continued to track the result of our QI project till May 24 and it has shown to be stable and consistent, with more than 98% of the discharge summary having complete secondary diagnoses.
Limitations
Even though our improvement efforts were vastly successful, our project does have its limitations. Change models are often designed and used to guide large-scale organisational changes while our project only involves the physician in the clinical department which consists of only 35 people. Smaller scale project means that the initiatives are easier to implement and there is often less resistance on the ground. Hence, the full potential and benefit of the change model may not have been demonstrated.
In addition, this QI project was driven by consultants and resident physicians who are already considered senior in rank. Hence the driving of the change model was well received and adopted by the rest of the department. However, in settings where these changes are led by junior doctors or other cadres of staff, this outcome may vary due to a lack of leadership support.
Finally, leadership and staff members did not use the model proactively but required multiple reminder sessions and prompts by the team to practice them and the findings of our project may not be generalisable in other institution that uses clinical coders for diagnosis entry.
Sustainability
Our results showed a sustained improvement in the completion of secondary diagnoses capture rate despite the usual turnover of junior physicians and the planned removal of visual reminders after the QI project was closed. The rate of completed secondary diagnoses from August 23 to May 24 showed a sustained result with the secondary diagnoses capture rate ranging from 97% to 99%. This result suggests that the change management concepts used to drive this project were well used and successful. The effective engagement of clinical staff with was done by creating a shared purpose and communicating the change vision. Even though there was the usual turnover of junior physicians in the department, the scheduled training sessions during new staff orientation on the importance and benefits of capturing the secondary diagnoses, act synergistically with the ease of coding for secondary diagnoses with the enhancement of the electronic medical system. This allows all physicians in the department to consistently perform this task, ensuring the success and sustainability of this project.
Conclusion
Change management models are simple and systematic tools that can be used across a variety of healthcare systems to improve quality within an organisation. By focusing on small tests to create and measure change, improvement is seen faster, compared with the use of a typical research measurement. Using them as guiding principles in a healthcare setting, our team successfully incorporated change management techniques to promote success in our QI project. The project highlights the success when we integrate models for change management with models commonly applied for improvement and implementation to support positive changes in healthcare.
Data availability statement
All data relevant to the study are included in the article or uploaded as online supplemental information.
Ethics statements
Patient consent for publication
Ethics approval
No institutional review board was needed for quality improvement projects according to local guidelines.
Acknowledgments
The authors would like to acknowledge the Department of Post-acute Care and Continuing Service for their participation in this quality improvement project. We would also like to thank Dr Luke Low and Dr Xu Bangyu for their encouragement and guidance in this project.
Footnotes
Contributors Both authors contributed equally to the planning, conduct and reporting of the work described in the article. AMW is the guarantor for this article.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.