DISCUSSION

Section:

ChooseTop of pageAbstractINTRODUCTIONMETHODSRESULTSDISCUSSION <<REFERENCES

We have shown that patients taking oral oncolytics have a high symptom burden, with many of these symptoms frequently reported as moderate-to-severe. Furthermore, we identified discordance between patient-reported symptom severity and oncologist documentation in the medical record. Moderate-to-severe symptoms reported by patients were documented in the medical record approximately 60% of the time. Finally, we found that 37% of patients reported less-than-excellent adherence to their oral oncolytic medication.

One possible explanation for our findings is that symptoms are discussed and managed but not documented in the medical record because of the burdens associated with documentation.^21-24 Lack of communication or miscommunication between oncologists and their patients during appointments may occur if agenda setting does not take place or because many other issues, such as prognosis, goals of care, financial concerns, dealing with uncertainty, and numerous symptoms, are addressed. Another explanation is that patients’ reporting of their symptoms is influenced by the method of collection—self-report via a questionnaire versus open-ended questions (eg “How are you feeling?”) from their oncologist. Finally and importantly, implementing PROs in clinical oncology practice has multiple challenges. Much of the experience to date is in the setting of clinical research. A framework for successful implementation should be comprehensive in addressing the clinical setting (eg where in treatment and/or monitoring it will be used and with whom communication will occur), the PROM tool itself (eg valid, reliable, simple), the feedback and documentation system, and support of implementation.²⁵

Of the symptoms reported by the patients, depression and anxiety were infrequently documented in the medical record. Almost half of patient assessments indicated moderate-to-severe depression or anxiety, with only 20% of depression and anxiety reported by patients documented in the EMR. A recent report recapitulates our results, finding that patients who reported depression and anxiety were least likely to have the symptom acknowledged and be offered assessments and interventions.²⁶ These symptoms are not only independent risk factors for nonadherence, but both have been associated with higher rates of cancer mortality.^27,28 The discrepancy between patient report versus that documented by the provider for depression and anxiety identifies an important area in which additional research would be helpful for determining whether the cause is a result of patient discomfort in discussing depression and anxiety with their provider and/or provider avoidance of discussing depression and anxiety.

The proportion of nonadherence was 37% in this study, which is in line with previous reports of this patient sample.^29-33 It is important to note that patients vary in what they consider excellent adherence. Although total pill consumption is important for assessing adherence, other factors that may be pertinent to a particular oral oncolytic, such as timing of administration with regard to time of day or surrounding meals would be important factors of adherence to assess in future endeavors. Only a small percentage of patients (7%) had adherence to their oral oncolytic therapy documented in their medical record. This may indicate either a lack of assessment, lack of documentation of assessment, or a lack of communication between the patient and provider regarding oral oncolytic adherence. Previous literature has shown that the use of PROMs for patients taking oral oncolytics leads to greater adherence to the regimen.¹⁵

There are several limitations to this study. First, it was conducted at a single practice and results may not be generalizable to other oncology practices. In addition, given the lack of clear communication of the PROM to the provider, it is unclear how the patients completed the document and whether symptoms were truly current versus something the patient previously experienced. A provider-patient discussion provides an opportunity for clarifying the symptoms reported by the patient, taking the symptoms into context, given prior visits and symptom-based therapies, and determining whether symptoms are current or no longer relevant, given the clinical scenario. Because the PROM was not consistently reviewed at the time of the visit, we were not able to determine how documentation would vary after communication with the patient about the document. There lies an inherent challenge to abstracting data from the oncologist’s documentation and variability in provider documentation from one provider to another. Although 100% documentation for pain may indicate hypervigilance to this potential adverse effect, inclusion of pain in a standard ROS designed to uncover dysfunction or adverse effects may account for this. By including the ROS in the data abstract, information included in this section of the oncologist documentation may not represent what was truly discussed and documented in that clinic visit. Future endeavors should aim to assess whether the use of PROMs for patients taking oral oncolytic therapy improves patient provider communication and enhances clinical management in regard to symptoms and, ultimately, adherence to oral oncolytic therapy.

These limitations notwithstanding, PROMs have the opportunity to accurately capture symptom burden, severity, and adherence to cancer therapy for patients taking oral oncolytic therapy. PROMs may also provide oncologists with a mechanism for enhancing patient self-management education and identifying symptoms and adherence that require intervention at an earlier point. In an era of health care reform, new payment and delivery models to improve care and lower costs through episode-based payment models will link outcomes associated with patient symptom burden and adherence to their treatment (hospital admissions, emergency room visits, resources used) to provider reimbursement. Thus, the consistent use of a PRO tool to assess and address patient outcomes may be beneficial, specifically in the realm of oral oncolytic therapy.

In conclusion, inconsistencies were seen between patient self-report of symptoms and adherence on a PROM and clinician reporting via documentation in the EMR. More symptoms and higher symptom severity were reported via patient self-report. Documentation contained within the EMR relating to patient symptom burden and adherence to their oral oncolytic lacks relevant information. This may indicate either communication errors or a lack of assessment and provides a poor representation of the patient’s symptom burden and adherence to their regimen. This analysis is important, because patients who are receiving oral oncolytics now have more responsibility in identifying and self-managing their symptoms. Use of a PROM for patients receiving oral oncolytics has the opportunity to improve symptom management and adherence to their regimen.