Anticipatory Advance Care Planning Visits and COVID-19 Treatment Intensity Among Medicare Fee-for-Service Beneficiaries: A Retrospective Observational Study.

BACKGROUND

In March 2020, professional organizations issued guidelines for anticipatory COVID-19 advance care planning (aACP) with high-risk older adults.

OBJECTIVE

To examine responses to these guidelines and associated COVID-19 treatment intensity.

DESIGN

Retrospective regression discontinuity design (RDD) using 2020 Medicare Parts A and B claims to assess aACP receipt, by beneficiary COVID-19 mortality risk. Adjusted logistic regression to assess predictors of aACP and the association between aACP and COVID-related mechanical ventilation.

PATIENTS: Medicare fee-for-service beneficiaries 66 and older enrolled on January 1, 2020.

MAIN MEASURES: ACP visits assessed via current procedural terminology billing codes 99497 and 99498: classified as anticipatory if the place of service was ambulatory and outside an annual wellness visit. COVID-19 mortality risks based on age and/or co-morbidity and practice attribution were assessed pre-lockdown. Hospitalization within 2 weeks of COVID-19 diagnosis and associated mechanical ventilation (MV) were assessed post-lockdown.

KEY

RESULTS

In total, 24,935,234 beneficiaries received 470,046,404 encounters in 2020; 1,578,331 were for ACP; of these, 318,813 (20%) were classified as anticipatory. Lockdown abruptly decreased all encounters. The RDD estimate found aACP decreased by 54% (95% CI= -0.62, -0.46); the relative decline in aACP was smaller for highest COVID-19 mortality risk groups: multimorbidity (-42%; 95% CI= -0.48, -0.35), patients with advanced cancer (-44%; 95% CI= -0.52, -0.36), dementia (-35%; 95% CI= -0.43, -0.27), or ESRD (-38%; 95% CI= -0.55, -0.21). Care from an integrated health system was associated with less aACP. Among 1,314,986 beneficiaries diagnosed with COVID-19, aACP was associated with an increase in adjusted 14-day hospitalization (OR = 1.21; 95% CI=1.17-1.26) but a decrease in invasive mechanical ventilation if hospitalized (OR = 0.85; 95% CI=0.77-0.96).

CONCLUSIONS

Providers followed recommendations to conduct anticipatory COVID-19 ACP with their high-risk patients, which was associated with reduced COVID-19 treatment intensity. This underscores the need for effective identification and dissemination of clinical risk factors in the next pandemic to guide anticipatory decision making.