Lung cancer is the second most common cancer treated within VA and is the leading cause of cancer related death among Veterans. The care of patients with lung cancer has been demonstrated to be heterogenous. The Office of Quality and Performance assessed the quality of lung cancer care within VA. Findings indicate the national rate for initial pain screening is 70% with large variation among sites (2%-100%). These measures recognize delays in delivery of palliative care in the treatment of lung cancer likely results in unnecessary symptom burden and poorer quality of life. Despite guideline recommendations to integrate palliative care into curative care, the majority of patients do not receive palliative care services until all curative options are exhausted.
The purpose of the study was to: 1) assess the feasibility of recruiting patients and delivering a nurse-led telephone based palliative care intervention for patients with newly diagnosed lung cancer; 2) determine the acceptability of the intervention; 3) estimate the effect of the intervention on quality-of-life, symptom burden and patient satisfaction.
We conducted a randomized controlled trial at the Puget Sound Health Care System with 40 outpatients diagnosed with lung cancer, any stage or type, within two months. Patients assigned to the intervention received 8 phone calls from a nurse over 3 months. The calls entailed symptom assessment and management, education about lung cancer treatments, goals of care conversations and psychosocial needs. Exclusion criteria: patients with cognitive impairment, inability to sign informed consent or currently enrolled in palliative or hospice care. Process measures: time to conduct calls, nurse recommendations to providers and referrals to ancillary services. Outcomes measures (quality of life, symptom burden and patient satisfaction of care questionnaires) were administered at baseline and study completion. Demographic information was retrieved through CPRS. Analysis was based on the intent to treat. We conducted descriptive statistics and linear regression models calculating Cohen's d to estimate the effect of the intervention on quality-of-life, symptom burden and patient satisfaction.
Aim 1. Feasibility. Recruiting and retaining subjects for a nurse-led telephone-based palliative care study is feasible. We recruited 40 patients over 18 months. We screened 261 subjects and invited 125 eligible subjects to participate in the study. Sixty-one subjects declined to participate, 41 subjects completed informed consent, 40 subjects were randomized and 36 subjects completed the study. The main reasons subjects declined to participate were feeling overwhelmed with their cancer treatment or feeling the study would not benefit them. Four subjects were either lost to follow up or withdrew from the study due to enrolling in hospice or feeling too burdened by the cancer treatment.
Patient characteristics. Patients were white males (95%), average age of 66.7 (13.1 SD), diagnosed with early stage lung cancer (stages I-II (70%)), NSCLC (90%). The majority of patients (80%) in the intervention had early stage cancer with surgical resection representing their primary treatment.
Patient Calls. Eight intervention calls were planned over 12 weeks. The nurse completed an average of 6.7 (2.5 SD) calls. Call preparation averaged 7minutes/patient and average call duration was 18 minutes (range 10-48). Time to complete the post-call note in CPRS averaged 10 minutes. Total time for 1 call/patient= 40-50 minutes. The nurse's caseload during any given 3 month intervention period was 5-7 subjects equating to approximately 25% FTE.
Nurse to Provider Communication. The nurse communicated patient information to providers through a research note entered in the EHR. Among the 17 patients in the intervention, the nurse made 20 recommendations to providers on 8 subjects. Recommendations mainly focused on prescribing new medications for symptoms or titrating existing medications. Other recommendations included referrals to social services or specialty clinics. All but one provider followed the recommendations and entered orders into EHR.
Aim 2. Study Acceptability. Exit interviews with patients (n=17) completing the intervention showed an overall patient satisfaction with the program = 9.5M (1.0 SD) (0= not satisfied-10 extremely satisfied). All patients with the exception of one felt the number of telephone sessions was right. The majority of patients (76%) did not feel there were too many surveys to complete. All patients rated the nurse as very helpful, facilitating access to care and communication with physicians. Patients commented the intervention decrease the burden of face-to-face visits.
Aim 3. Quality-of-life (QOL) and symptom burden. We measured patients' quality of life using the Total Outcome Index (TOI) subscale of the validated Functional Assessment of Cancer Therapy - Lung Scale (FACT-L)The mean TOI score range is 55.3-59.9 (14 SD) and the clinically relevant change scores are estimated at 5-7 points in full clinical trials. Patients' (n=40) mean quality of life at baseline was 52.0 (16.9 SD). The mean change in QOL was 4.7 (17.7 SD) in the control group and 2.7 (9.5 SD) in the intervention group, Cohen's d (-0.14). An adjusted Cohen's d accounting for cancer stage showed the difference to be less substantial (-0.04).
Satisfaction with VA care. We measured patient satisfaction with care using the FAMCARE-13. The FAMCARE is a 13 item, 5 point likert-scale validated questionnaire measuring patient satisfaction with cancer care and assessing interactions with health care providers.. Scores range from 13-65 with scores of 52 > being satisfied with care. In full randomized clinical trials the estimated MID is 5 points from baseline to 12 weeks. Patients (n=40) mean satisfaction at baseline was 57.9 (9.1SD) indicating overall satisfaction with care within 2 months of a lung cancer diagnosis. The mean change in FAMCARE was 1.02 (7.2 SD) in the control group and 0.76 (9.6 SD) in the intervention group, Cohen's d (-0.03). An adjusted Cohen's d accounting for cancer stage showed the difference to be equally small (0.02).
This pilot study demonstrates that a nurse-led telephone-based palliative care intervention for outpatients newly diagnosed with lung cancer is feasible to conduct and acceptable to patients. These results support the VA Blueprint of Excellence by providing personalized, proactive, patient-driven health care and improving access and health outcomes.
- Reinke LF. Symptom Assessment and Management of Patients Newly Diagnosed with Lung Cancer. Paper presented at: American Thoracic Society Annual International Conference; 2015 May 19; Denver, CO.
- Reinke LF. How to overcome fear of death by talking about dying. Presented at: American Thoracic Society Annual International Conference; 2014 May 16; San Diego, CA.