Evaluation of a staff-led clinical redesign program to improve patient flow through outpatient clinics : a mixed methods study

An increased demand for health services in Australia is driven by an ageing population, increased consumer expectations, expensive technologies and a growing burden of chronic diseases. The complexity of the health system presents challenges for patients with chronic health conditions as they may be under the care of multiple health professionals across primary and secondary settings. As primary care practitioners serve as the 'gateway' to the secondary care, a referral is required to enter this wider health system. In publicly funded hospitals, secondary care is usually provided by outpatient clinics and is triaged based on clinical urgency. This results in the patient being placed on a wait list. Waiting times for outpatient care in Australia are not subject to the same level of scrutiny as elective surgery waiting times, time spent in emergency departments or inpatient length of stay. The only nationally collected metrics are the number of 'service events' (appointments), the types of services provided, demographic information of the users and how the services are funded. This study is important as it highlights the problems of analysing data from an area of health which places minimal value on collecting and maintaining accurate statistics and places a focus on an under-researched area of health. The aim of this research was to evaluate a staff-led clinical redesign program where an external body (the University of Tasmania) worked in collaboration with the health system. This project was part of a federally funded state-wide program to improve the effectiveness, efficiency and long-term sustainability of Tasmania's health system. Outpatient clinics were one of five key areas targeted for redesign in Tasmania's public hospitals. From internal hospital data, outpatient clinics that had a long wait time to first appointment, a high Did Not Attend (DNA) rate, a low discharge rate and a high number of hospital and patient-initiated cancellations were invited to participate. Lean methodology has shown success in redesigning healthcare processes which involve a linear sequence. Patient flow is the successive movement of people through a sequence of processes along a pathway of care. In this study, patient flow encompassed all the steps between referral into the outpatient clinic, obtaining an appointment and transfer back to community care and is referred to as a 'value stream.' This mixed methods study had an embedded research design where the secondary data set (in this case, the qualitative data) was embedded in the primary data set (quantitative) and used to answer the following primary research question: Does the application of clinical redesign improve patient flow through Plastic Surgery and Ophthalmology Outpatient clinics? Patient flow was defined by the following measures: ‚Äö Percentage of patients who waited longer than the clinically recommended time before attending their first outpatient appointment by triage category ‚Äö Median wait time to the first appointment by triage category ‚Äö Did Not Attend (DNA) rate ‚Äö Discharge rate ‚Äö Hospital and patient appointment cancellation rates ‚Äö The number of overdue follow-up appointments Method This was a mixed methods observational study in which a staff-led clinical redesign program was evaluated. The intervention began with staff from all working areas of two outpatient clinics being instructed on the Lean practices of standard work processes and the reduction of waste during a 2-day workshop. The aim of the workshop was for staff to map the current value stream to identify areas of waste and inefficiency and agree on a 15-month redesign strategy (with the help of a redesign consultant and a project officer). The purpose of this research was to assess if the staff-led redesign program improved patient flow parameters by comparing the pre-study and intervention metrics. A unique feature of the study was that the hospital only provided raw appointment and waitlist data for analyses. This was a deciding factor on the choice of an embedded research design. The qualitative data (patient and staff surveys, field notes and redesign meeting notes) facilitated the construction of a visual model of patient flow, which fully informed the choice of patient flow metrics for the quantitative analysis. The qualitative data was also used to confirm the internal validity of the quantitative results. As a measure of patient flow, the primary objective was to assess the change in the proportion of patients who waited longer than the clinically recommended time limit for their respective triage categories (category 1 <30 days, category 2 <90 days and category 3 <365 days) for a first appointment. Another feature of an embedded design is that the secondary data set can be used to answer a separate research question. In this research the qualitative data from both clinics were combined into one data set and thematically analysed to answer the secondary research question: What are the factors influencing the implementation and success of the redesign initiatives? Results Clinic demand and patient access were the main foci of the redesign activities for the Plastic Surgery Working Group. A new model of staff flow during clinic sessions was introduced, along with a nurse-led clinic dedicated to complicated dressing changes. A 'Physio first' model of wrist care was trialled but failed to see any patients. New staff guidelines were written to enforce current practices with an emphasis on safely discharging patients back to community care. The percentage of category 1 patients who waited more than 30 days for their first appointment decreased during the study, from 43.5% to 28.6% (p< 0.00001). This mainly reflected a decrease in the number of long-waiters, as the median wait time only changed slightly (from 8 to 6 days). Although the percentage of category 2 patients waiting longer than 90 days remained high (97.4% vs 96.4%), the median wait time decreased significantly (560 to 405 days, p<0.0001). The median wait time for category 3 patients did not show a significant change (1112 to 1038 days, p=0.3). The other measures of patient flow were DNA rate, discharge rate and appointment cancellations. The overall clinic DNA rate did not change significantly (13.7% vs 12.7%, p=0.06). When the data was examined for targeted interventions, there was a decrease in the DNA rate of return appointments in the registrar hand clinic (15.6% vs 13.1%, p=0.02), with the biggest improvement in the registrar hand clinic on Tuesday afternoons (17.0% vs 11.8%. p=0.00005). There was a modest improvement in the discharge rate for return appointments (24.7% vs 26.6 %, p=0.001), which may have been higher if the data set was complete. The hospital and patient cancellation rates could not be calculated with any accuracy because of the multiple methods of processing cancelled appointments by the hospital. Clinic demand and patient access were also the main foci of the Ophthalmology Outpatient clinic redesign activities. Diabetic retinopathy screening referrals were no longer accepted, a practice consistent with larger Australian eye hospitals. A systematic discharge program was implemented for all current diabetic retinopathy patients with mild or no disease, back to community optometrists/ophthalmologists. Due to an incomplete data set the change in the discharge rate during the study could not be accurately calculated (although this initiative alone was known to discharge at least 285 patients). The number of overdue follow-up appointments as an outcome measure was not part of the initial study plan, as it was not a recognised element of patient flow through the clinic system. It was the Ophthalmology Working Group who recognised patient flow as the delicate balance between allocating appointments to patients from the wait list, post-operative appointments and patients who were due follow-up appointments. During the Intervention period, as the number of patients on the outpatient wait list decreased, the number of patients overdue for their follow-up appointment increased. The cause was multifactorial, as there were an additional 102 surgeries during the Intervention period and each of these patients required at least one follow-up appointment, as well as the concentrated effort to allocate appointments to category 2 patients from the outpatient wait list. Data integrity provided challenges throughout the analysis, especially when interpreting the Ophthalmology Outpatient clinic results. The business rules for how appointments were made and cancelled were not established prior to the project, which resulted in an incomplete data set and the method of triaging of the referrals to the clinic changed three time during the Pre-study period. Despite this, the largest change in the waiting time to first appointment which could be attributed to the redesign program was for category 2 patients (median wait time decreased from 183 to 123 days, p<0.0001). Three overarching themes were identified as factors influencing the implementation and success of the redesign initiatives in both clinics: Context, People and Process. Context described how the local characteristics of the project impacted the study (e.g. physical space, funding arrangements and available data). The People theme included human capital and how local engagement and understanding influenced the outcome of the program. The final theme ‚ÄövÑv¨ Process ‚ÄövÑv¨ was the largest and most diverse of all the themes, as nearly all the initiatives were affected by a process issue. To emphasise the importance of timing and sequence of steps in the redesign project, a process map was developed as an extension of the thematic analysis findings. This was an expanded version of the conventional 'plan-do-study-act' cycle, often used in health care redesign programs. The cycle highlighted the importance of considering and completing each step before progre...