Development and Analysis of a Treatment Timeliness Dashboard for Patients With Head and Neck Cancer at a Tertiary Academic Multidisciplinary Clinic

J Clin Pathways. 2024;10(6):31-36. doi:10.25270/jcp.2024.10.02

In 2021, the Commission on Cancer approved the first quality metric for head and neck cancer: time to initiation of postoperative radiation therapy (PORT) less than or equal to 6 weeks for surgically treated head and neck squamous cell carcinoma.¹ However, anecdotal observation belies the complexity of having patients start PORT within the standardized 6-week timeline. Unexpected barriers to access and care contribute to these complexities. Understanding timeliness to treatment rep­resents a critical opportunity to maximize patient experiences and quality-related outcomes, including disease survival.²

Effective care coordination can help address the complexities of timeliness and improve outcomes for patients with cancer.³ Nurse navigation is a well-established concept in current oncology practice, but the specific role of a nurse navigator often varies between practices.⁴ In large oncology practices, a single navigator may not be enough to support each patient’s individual care coordination during their can­cer journey. Frontline outpatient oncology nurses are often better staffed and have greater visibility into the personalized needs of patient care.

As an academic institution, our Head and Neck Program recognized the com­plex multidisciplinary challenges of cancer care delivery in head and neck cancer. To address these challenges, we developed a Lean methodology-based disease-line model comprising a co-located multi-disciplinary care (MDC) clinic. Unlike the traditional sequen­tial care model, wherein newly-diagnosed patients attend multiple consultations with different oncology providers, the MDC clinic provides simultaneous access to several oncology subspecialties in a single visit.⁵ Outpatient nursing care coordi­nation was also proposed as the core model for pre-visit coordi­nation, day-of flow, and post-visit care implementation, along with tracking quality metrics to collect and analyze treatment timelines. Timeliness to treatment delivery was identified as a core value to patients, caregivers, and clinicians. Timeliness to treatment was considered an important metric to measure patient experience, quality care, and the overall health of the MDC program. The team worked for more than 5 years to develop an internal treatment timeliness dashboard for all patients with head and neck cancer seen at our facility using a primary nursing care coordination model.

This study was approved for human subject research by the Insti­tutional Review Board at our institution (STUDY00002258).

Lean Process

The Head and Neck Program and Strategic Planning office used value stream analysis (VSA), a lean-based tool designed to evaluate any standard process by eliminating wasteful work processes.^6,7 Experts across multiple disciplines developed a current and ideal state flow map for the patient experience. Gaps were identified to better streamline access to care. During the process, time points were recognized as key performance indicators for the program and patient care. Leadership identi­fied four timeliness metric measurements (in days):

1. Time from first contact or referral date to first MDC visit
2. Time from first MDC visit to date of treatment plan given
3. Time from first MDC visit to first treatment start
4. Time from first treatment start to second (adjuvant) treatment start

Secondarily, we tracked the number of patients who were offered an MDC visit within 7 business days as well as the num­ber of patients offered a clinical trial.

Nursing Framework

The core model was centered around outpatient nursing care coordination. Standard work was developed to educate nurs­ing staff, including visit preparation, day-of flow, and post-visit care implementation. Nurses would document the timeliness metrics and use the timeliness data to drive awareness of any treatment delays or gaps. Care coordination for new patients focused on achieving care timeline goals, for example, by initi­ating primary concurrent chemoradiation (CRT) within a cer­tain number of days of the initial MDC visit.

Data Collection Transition

As part of their new care coordination role, the frontline nurs­ing team entered data daily into a Microsoft Excel spreadsheet. To assist with data collection, a basic data dictionary was out­lined to define metrics and additional data fields. However, monthly reviews revealed significant concerns about the accu­racy and completeness of the data. Staffing shortages and nurs­ing education gaps contributed to these issues. The MDC clinic was a daily event, and the pace at which patients were added led to a large, cumbersome spreadsheet to the point that it was no longer a feasible system for tracking. Based on nursing team feedback, the team decided to transition data entry to a web-based database, REDCap (Research Electronic Data Capture), in January 2021.^8,9 The transition was guided by the Lean A3 process of problem solving, which included defining the prob­lem, root cause analysis, target state, and implementation.¹⁰

Continued State

Throughout 2021, data were prospectively collected and in­putted into the REDCap database. It again became evident that collected data were neither accurate nor complete, which made reliable process improvement projects difficult to develop and implement. This lack of data fidelity was thought to be sec­ondary to an overlap with the primary nursing tasks, staffing shortage, and daily data entry for the frontline nurses. To clean the database and provide quality control, our Head and Neck Program decided to hire a data analyst; the role was intended to optimize the workflow of the outpatient nursing team to ac­curately use the database to identify for any delays in care if pa­tients were not meeting expected treatment target timeframes.

An MPH-trained data analyst joined the team in early 2022. The analyst helped rectify inaccuracies and incomplete records and updated any data that were either missing or incorrect in REDCap based on physician notes in the electronic medical record. In addition, the analyst revised, for the first time since its creation, the data dictionary that outlined the structure of the REDCap database and descriptions of each data field. With the inclusion of a data analyst, frontline nurses were able to continue to use REDCap for initial coordination of a patient’s oncology treatment journey and focus on its use as a tool for monitoring treatment timeliness and delays of care.

A total of 2676 patient records from May 2019 through April 2023 were cleaned fully for analysis. Preliminary analysis was performed to identify changes with the implementation of the MDC. The descriptive analysis included the following:

• Identifying patient population distribution by anatomic site of the primary tumor
• Calculating median and mean days of four timeliness measures
• Calculating percentage of patients offered a new MDC appointment within 7 business days
• Calculating percentage of patients offered a clinical trial at our institution
• Calculating percentage of patients who did not receive
  • adjuvant treatment (after surgery),
  • CRT (as definitive nonsurgical treatment), or
  • surgery (as definitive treatment)
• Identifying qualitative reasons for patients not seeking treatment for the above three treatment modalities

REDCap’s interface allows users to generate reports that calcu­late the mean and median of the four timeliness metrics. These reports were generated for patients seen in the MDC for each year: 2019, 2020, 2021, 2022, and 2023. An additional report and set of calculations were generated to determine the overall timeliness and other quality metrics across all 5 years.

The senior data analyst presents timeliness data monthly to our Head and Neck Program Steering Committee. The Steering Committee comprises disease-line physicians from each disci­pline (surgery, medical oncology, radiation oncology), depart­ment administrators, nursing specialty directors, and operation leaders. The monthly presentations track timeliness of patient care per calendar month and timeliness trends over several months to identify any deviations, for example, new patient volumes increasing new appointment wait times. The Steering Committee identifies demand and operational variabilities and enacts action plans to reduce and eliminate deviations.¹¹

A total of 2676 patients were seen in our Head and Neck MDC clinic between May 2019 and April 2023.12 These encounters include patients who were seen in the MDC but did not neces­sarily receive treatment at our institution. Patients may have been seen again in the clinic if they experienced a recurrence from a previously treated cancer or if they experienced a sec­ond head and neck cancer. Considering these factors, a total of 2549 unique patients were seen at the Head and Neck MDC clinic over the 5-year period.

Head and neck cancers include several anatomic subsites. Oral cavity and oropharynx primary sites represented 56% of the patient population, followed by larynx and skin cancers of the head and neck region (Figure 1).

To assess improvement in timeliness, we measured the four metrics for each year and conducted an overall analysis for all 5 years combined (Table 1). Over the 5-year period, we observed a decrease in the number of median days between designated dates across three of the four metrics. There was no change in timeliness for the metric “Time from the first MDC visit to treatment plan date.” This indicates that treatment plan deliv­ery occurred on the date of the first MDC visit.

We observed an improvement in the percentage of patients who were offered their first MDC visit within 7 business days. In 2019, 23.6% of patients were offered an MDC new patient visit within 7 business days compared with 52.6% in 2023. The percentage of patients who were offered a clinical trial re­mained stable at around 20% throughout all 5 years, with an overall average of 20.1% of patients offered a clinical trial at our institution. This percentage does not include the number of pa­tients subsequently enrolled in a clinical trial offered to them.

Out of 867 patients who were recommended a form of ad­juvant treatment, 61.4% received postoperative treatment at the Head and Neck MDC clinic, and 38.6% did not receive adju­vant treatment (Table 2). Of this subset, 59.1% did not receive adjuvant treatment at the MDC clinic due to distance from the clinic (Figure 2); these patients opted to receive treatment at a facility closer to their home. About 31% of patients refused the recommended adjuvant treatment plan, and 6% of this cohort opted to receive treatment with a “non-institution doctor”; pa­tients opted to receive adjuvant treatment at another specific in­stitution for reasons other than distance. A small proportion of patients had disease progression (1.8%), died before treatment (0.6%), or were lost to follow-up (1.5%). Loss to follow-up sig­nifies that the patient did not return to the MDC clinic, and there is no information available in the electronic medical re­cord indicating whether or not the patient received treatment elsewhere.

For 521 patients, primary CRT was recommended as their first treatment modality. About 80.4% of patients underwent primary CRT at the institution, of whom 19.6% did not re­ceive treatment at the facility; the majority of patients opted to receive CRT closer to home (40.2%) or with a “non-institution doctor” (35.3%). About 6.9% of patients refused recommended CRT, 2.9% of patients died before the start of CRT regimen, and another 2.9% of patients had disease progression. After the initial MDC visit, 11.7% of patients in this cohort were lost to follow-up.

Of 1495 patients who were recommended to undergo sur­gery, 82.4% received primary surgery at the institution, where­as 13.6% did not receive surgery at the institution. Most of the patients who did not receive surgery refused surgery recom­mendations from the MDC team (40.0%) or opted to receive surgery with a “non-institution doctor” (26.1%). After the initial MDC visit, 19.7% were lost to follow-up. Additionally, 7.4% of patients opted to receive surgery closer to home, 7.4% of patients died before their recommended definitive surgery, and 3.4% had disease progression before surgery.

Our Head and Neck clinic’s experience with developing an internal timeliness dashboard using primary nursing care co­ordination highlights the importance of quality improvement projects to improve health care delivery systems within current resources. Although this was a prolonged and iterative process, preliminary results have shown that the implementation of the co-located MDC clinic with nursing care coordination has im­proved timeliness of care.

There was steady improvement in timeliness across two of the four metrics over the 5-year period. The first metric, “Time from first contact to first MDC visit,” remained steady at 10 median days overall. “Time from first MDC visit to treatment plan delivery date” also remained steady at 0 median days. This is a major improvement when comparing this metric between the Head and Neck clinic’s former sequential care model and the MDC model. Previously published data predating spring 2019 revealed that “Time from first consult visit to treatment plan delivery date” was 18 median days.⁵ Implementation of the MDC model eliminated the lag in care of the sequential model between a patient’s first visit and establishment of a treatment plan. Instead of attending several appointments and waiting for a treatment plan to be devised, the treatment plan could now be formulated concurrently at the time of the first MDC visit and assigned a primary nurse coordinator who is aware of the plan of care.

For the metric “Time from treatment plan delivery to first treatment start date,” timeliness improved for this metric over the 5 years with an overall timeliness of 27 median days. The sequential care model metric for spring 2019 was 23 median days.5 This may not seem like an improvement, but the timeli­ness of 27 median days is equivalent to the time from first clinic visit to treatment start date for the MDC model. For the se­quential model, the time from the first clinic visit to treatment start date was 41 median days.⁵ Therefore, with the elimination of lag time of first clinic visit to treatment plan delivery date with the implementation of the MDC model, there has been an establishment of an earlier treatment start date for patients by 14 median days compared with the prior sequential care model.

We also observed a steady improvement in timeliness in the “Time from first treatment start date to adjuvant treatment start date” metric from 50 median days to 43 median days over time. This metric comprises a smaller cohort who received adjuvant treatment after their surgical procedure. The timeline is a lon­ger period because patients needed time to recover from their surgical procedure before being seen again in the MDC as an Established Disease Clinic (EDC) visit. During the EDC visit, patients meet with the same MDC team to determine if additional (adjuvant) treatment is needed. After the EDC visit, patients who undergo treatment, such as adjuvant radiation or adjuvant concurrent chemoradiation, undergo the care coordination process of radiation planning and simulation and potential dental extractions facilitated by frontline nurses. Insurance authorization creates additional time constraints at all stages of care.

This project also showcases the importance of establishing and updating standardized work practices for data collection, cleaning, and analysis. It became evident early on that input­ting and updating data was arduous for the nursing team soon after implementing the MDC clinic. The team did not antici­pate an increase in patient volume when the clinic opened, and how clinic volume continued to grow over the years. Despite having improved their data input skills, the nursing team con­tinued to struggle with updating existing records as patients received treatments. Therefore, many records were left incom­plete. The inclusion of a public health data analyst in 2022 helped alleviate this problem by providing support on the back end of data collection. Nurses would create new records and enter initial data into the REDCap database, while the data analyst updated and completed records to later perform analy­sis for the team.

There are several limitations to this project. First, there was no formal development of any nursing pre- and post-surveys for frontline feedback except for subjective conversation during nurse staff meetings. Second, only 242 encounters were record­ed into the Excel spreadsheet for 2019. Records were not con­sistently added from May 2019 through September 2019. These missing records affect the accuracy of results, particularly for the 2019 and overall analyses. The inconsistency of inputting data ultimately led the team to change the nursing care coordination model to improve data input. Third, the REDCap database con­tained limited information. The database only contained basic fields, such as patient name, medical record number, primary site, treatment type, timeliness metrics, and basic clinical trial information. The public health data analyst incorporated ad­ditional information including sociodemographic and clinico­pathologic information in an updated version of the database. This will facilitate future quality improvement and operational projects for the clinic. Fourth, the database also only includes treatment information received at our institution; any treatment received elsewhere was not included. Analysis only pertains to the timeliness of the MDC clinic and does not reflect the timeli­ness of other institutions. Lastly, some of the data were collect­ed during the COVID-19 pandemic, which may have skewed referral and treatment patterns.

The implementation of a treatment timeliness dashboard alongside a co-located MDC improved timeliness of patient access and care compared with a previously used sequential care model at our institution. Establishing and updating stan­dardized work practices is of utmost importance in order to produce reliable and clean data for quality improvement and operational projects while allowing frontline nursing teams to use the data as a tool for timeliness visibility and monitoring for any deviations in care.

Our Head and Neck MDC continues to optimize the time­liness dashboard both in monitoring the health of the program on a monthly basis and for development of future process im­provement quality initiatives. The team looks forward to fur­ther investigating the abilities of the electronic medical record to better integrate with REDCap to measure timeliness while also alerting clinicians and nursing staff if patients do not re­ceive treatment within the delivery timeframes. Identifying common barriers and gaps in care will allow the team to further develop care improvements. The team is eager to move forward from this experience of developing and improving a timeliness dashboard to align with national quality improvement projects where timeliness of care can impact overall treatment outcomes and survival.

Affiliations:

¹Department of Otolaryngology-Head and Neck Surgery, Winship Cancer Institute at Emory Midtown, Emory University School of Medicine, Atlanta, GA; ²Business Transformation Office, Emory Healthcare, Atlanta, GA

Contributions:

Martha J. Ryan MSN, FNP-C and Melissa F. Riedel MPH, ODS-C contributed equally.

Correspondence:

Melissa F. Riedel, MPH, ODS-C

Phone: (703) 434-1347

Email: melissa.forbach@emory.edu

Disclosures:

The authors disclose no financial or other conflicts of interest.