Conclusions
A quality gap exists between national guideline recommendations and delivery of post-operative radiation therapy in patients treated surgically for HNSCC. A majority of patients do not initiate post-operative radiotherapy within guideline-recommended 6 weeks.8 Our single academic institution review mirrored those results. The majority of patients (88.6%) did not start RT within 6 weeks of surgery, and the average time to initiate of PORT was 7.9 weeks.
Risk factors for a delay in initiation of PORT include, but are not limited to, lower seriocomic status, black race, public insurance, or uninsured, increasing severity of comorbidities, increased post-operative LOS, 30 day readmissions, fragmented care8, and receiving adjuvant therapy at a non-academic center.10,15
Our case-series sought to determine potentially modifiable risk factors which are associated with a prolonged initiation of PORT. Many patient related factors that might delay PORT cannot be altered - such as patient age or tumor stage at presentation. However, there exists some patient and treatment related factors which offer an opportunity for quality improvements and thus earlier initiation of PORT. While previous studies have identified several factors associated with a delay in PORT, our study assessed additional potentially modifiable risk factors such as gastrostomy tube placement, rehabilitation facility placement, and post-operative complications which were all associated with, or at least showed a trend towards, a delay in PORT (Figure 4). Additionally, our single-institution study allows for collection of these more granular data points (such as wound complications and infections), which may not be available in the larger database studies.
Post-operative length of stay, 30-day readmission and post-operative complications were associated with delays. Interestingly, patients who received adjuvant chemotherapy had significantly lower intervals between surgery and initiation of adjuvant therapy (10 days fewer, p = 0.087). Additionally, delays were decreased if surgery and adjuvant radiotherapy were performed at the same institution. Need for dental extractions, gastrostomy tube placement, and rehabilitation facility placement were associated with a longer interval to initiation of PORT, but were not statistically significant.
The finding that the addition of adjuvant chemotherapy to radiation was associated with earlier initiation of adjuvant therapy seems counterintuitive. We postulate that the reduced interval for these patients may be due to earlier recognition for needed adjuvant therapy. A patient with more advanced clinical disease at initial presentation may be referred in a timelier fashion than a patient whose need for adjuvant therapy may be unclear at initial presentation. Referrals in the latter type of cases are typically made only after final pathology results are released, and perhaps only after a tumor board discussion; this decision may not occur until up to 2-3 weeks after surgery. However, this finding may be in part explained by the fact that adjuvant chemotherapy is only be offered to a healthier subset of patients. These patients would be more likely to start therapy sooner than their unhealthy counterparts.
Fragmented care has previously been associated with delayed initiation of PORT and poor adherence to treatment regimens,10and poorer outcomes.16,17 We found a statistically significant decrease in PORT delays for those who received both surgery and adjuvant therapy at our academic facility. Academic centers may improve timeliness of care due to ease of intra-institutional referrals, care coordinators, and multidisciplinary tumor boards which include surgeons, radiation, and medical oncologists. Additional benefit may include the patient volume of an academic center,17higher level of comfort with management of HNSCC-related complications, and fewer treatment interruptions or early terminations. Adjuvant therapy at an academic center has also been shown to reduce total “package time” (time from surgery to completion of adjuvant therapy), likely for the reasons stated above. A reduction of package time is associated with increased overall and recurrence free survival.14-15 This difference in fragmented vs. unfragmented care may be patient related as well. Those who choose to continue their adjuvant therapy at our institution, often despite significant commutes, may be more motivated, educated, have better family support, or better access to transportation.
Improving adherence rates for all patients will be challenging, as there are often unavoidable delays that push initiation of PORT beyond 6 weeks. Therefore, it is critical to identify those modifiable factors which may be contributing to PORT delays and directly address those factors through quality improvement processes.
Divi et al12 demonstrated the utility of a structured quality improvement project in decreasing the time from surgery to initiation of adjuvant therapy. Twelve interventions which sought to address the three key drivers of delay (delayed dental extraction, delayed referral to radiation oncology, and poor patient or team engagement) were implemented. They noted a reduction in avoidable delays from 24 to 9%. Additionally, Janz et al13demonstrated the following care processes improved timeliness of PORT: preoperative RT consultation, pathology report available within 7 days of surgery, time from surgery to PORT referral no longer than 10 days, and time for PORT consultation to its start no longer than 21 days.
In this study, we also noted that patients with longer LOS and those who had postoperative complications had significant delays in initiating adjuvant therapy. This presents an opportunity for intervention. The patients who are at risk for longer stays (e.g. those requiring free flap reconstruction, those without caregivers at home) or who are at high risk for complications (e.g., those with multiple comorbidities) should be identified and preoperative referrals to adjuvant care providers can be made. If the need for G-tube placement18 or dental extractions can be anticipated, these can be accomplished pre- or intra-operatively. However, complications and length of stay are not always modifiable factors, as unforeseen events often unfold during patients’ post-operative course, which may not be predicted. For example, a patient without significant comorbidities may still develop a wound complication which would prolong their stay.
We have initiated a structured institutional multidisciplinary process at our institution in hopes of reducing delays. This initiative will generate automated referrals and include a checklist led by a Head and Neck Cancer-specific nurse navigator to identify at-risk patients, facilitate consult appointments, review of pathology reports, and ensure timely presentation at tumor board. The specific details of the initiative are currently in development in conjunction between our surgeons, radiation oncologists, medical oncologists, and nurse navigator.
There were limitations to this study. Our sample size was adequate to determine statistical significance in delay for several risk factors. However, some risk factors (gastrostomy tube placement, need dental extractions, rehabilitation facility placement) showed a trend towards delay, but were not statistically significant, as might have been expected from prior studies.8
There was also a lack of availability of some outside treatment facility records, which resulted in exclusion of those patients. Some records did not include the day of therapy initiation, number of fractions received, number of fractions completed, or total treatment dose.
Our single academic institutional study also makes the results challenging to generalize to those not practicing in a similar setting.
Finally, comparing all non-academic centers as a single aggregate to our academic institution may unfairly generalize their outcomes. It is possible some non-academic centers have better outcomes than others, or even our institution, but this was not assessed in the data.
This study is important because it identifies modifiable risk factors for a delay in initiation of PORT, a known metric for quality care in HNSCC. Future studies will address the success of a structured institutional multidisciplinary approach and the use of a dedicated nurse navigator in reducing delays in adjuvant therapy, and assessing its effect on outcome measures, such as recurrence rate and disease-free survival in HNSCC.
In this single institution case series, a majority of patients with HNSCC did not adhere to recommendation of initiating adjuvant radiotherapy within 6 weeks of surgery. Identification of modifiable risks factors/barriers that delay initiation of PORT is crucial to reduce avoidable delays. A structured quality improvement project, which directly addresses these modifiable risk factors, would be expected to improve guideline recommended adherence to post-operative RT time intervals, and ultimately improve survival outcomes in patients with HNSCC.