The results of simple interventions involving community health workers suggest that improvements in value-based cancer care need not come from health-care professionals. According to data presented at the 2019 ASCO Quality Care Symposium,¹ reliance upon community health workers trained to assess patient symptoms and make referrals led to significant differences in health-care utilization, including decreased visits to the emergency department in the last month of life and increased use of hospice for the intervention group. Authors of the study also reported significant reductions in total costs of care among patients receiving the intervention, as compared with the control.

“This study shows that basic but essential cancer care services can be provided to patients outside of clinic visits by lay health workers,” said Manali I. Patel, MD, Assistant Professor of Medicine at Stanford University School of Medicine, Stanford, California, and the Veterans Affairs Palo Alto Health Care System, Palo Alto, California. “This was an all-stakeholder approach, designed from the beginning with the input of patients, caregivers, providers, and payers. The payer incentivized this value-based–care approach prior to any of the new alternative care models that we’re seeing now. Ultimately, we showed improved value in cancer care delivery at the end of life through this simple intervention,” she added.

“This study shows that basic but essential cancer care services can be provided to patients outside of clinic visits by lay health workers.”

— Manali I. Patel, MD

Tweet this quote

As Dr. Patel reported, annual health-care costs related to cancer are expected to exceed $200 billion in 2020, but this high cost is not necessarily associated with better care delivery. A majority of patients continue to suffer from undertreated symptoms, noted Dr. Patel, which leads to high rates of hospital admissions, longer hospital stays, inpatient deaths, psychological and physical distress, as well as financial burdens to both patients and the health-care system.

Study Details

To improve value-based cancer care, Dr. Patel and colleagues designed an intervention using lay health workers or community health workers, embedded in the practice to assess patient symptoms. The clinical team risk-stratified patients into either high-risk or low-risk categories based on comorbidities, baseline symptom burden, cancer stage, and the likelihood of symptoms over time. High-risk patients received a weekly phone call from a lay health worker, who conducted symptom assessments using the Edmonton Symptom Assessment Scale and screened for depression using the Patient Health Questionnaire-9. These symptoms were then reviewed with an embedded physician assistant or nurse practitioner in the clinic who conducted the clinical intervention. For patients in the low-risk group, this assessment occurred on a monthly basis.

From November 2016 through September 2018, researchers enrolled all newly diagnosed health-plan beneficiaries with solid and hematologic malignancies and compared outcomes with all patients diagnosed with cancer in the year prior to the intervention. For all patients who died within 12 months of diagnosis, the researchers compared the risk of death using Cox models. Health-care use in the last month of life and the total costs of care from diagnosis until death were also analyzed by generalized linear regression models, with all models adjusted for age, disease stage, comorbidities, diagnosis, and length of follow-up.

Significant Differences in Health-Care Utilization and End-of-Life Costs

As Dr. Patel reported, of the 832-patient cohort, 189 patients in the control group died and 180 in the intervention group died. As this was a historical control, she added, there were differences a priori in demographic characteristics. Although the majority of patients in the intervention arm were men, the majority of patients in the control group were women. However, average age and race/ethnicities were similar across both groups.

The authors also reported differences with respect to cancer diagnoses between the arms, mostly driven by differences in breast cancer, but the number of patients with stage IV disease was similar in both groups, as were comorbidities. Although no differences in survival were found between the groups, patients were selected on the basis of having died within 12 months of diagnosis, said Dr. Patel, so there was limited variability in the cohort.

Nevertheless, researchers found significant differences in health-care utilization, with statistically significant greater hospice use among the intervention group and lower emergency department use in the last month of life. Patients in the intervention group also had lower mean inpatient admissions than the control arm. In addition, researchers identified differences in terms of the percentage of patients who died in acute care facilities (37% in the intervention group vs 63% in the control group; P = .02).

Finally, researchers reported significantly lower median total health-care costs from diagnosis until death for patients receiving the intervention vs the control. This finding translated to a nearly $7,000 per-patient difference in total costs. 

DISCLOSURE: This study was sponsored by the National Institutes of Health. Dr. Patel reported no conflicts of interest.

REFERENCE

1. Patel MI, Ramirez D, Agajanian R, et al: Enhancing community capacity to deliver value-based cancer care at the end-of-life. 2019 ASCO Quality Care Symposium. Abstract 4. Presented September 7, 2019.