Use of Advance Care Planning Billing Codes in a Tertiary Care Center Setting

Natural Language Processing to Identify Home Health Care Patients at Risk for Becoming Incapacitated With No Evident Advance Directives or Surrogates

OBJECTIVES

Home health care patients who are at risk for becoming Incapacitated with No Evident Advance Directives or Surrogates (INEADS) may benefit from timely intervention to assist them with advance care planning. This study aimed to develop natural language processing algorithms for identifying home care patients who do not have advance directives, family members, or close social contacts who can serve as surrogate decision-makers in the event that they lose decisional capacity.

DESIGN

Cross-sectional study of electronic health records.

SETTING AND PARTICIPANTS

Patients receiving post-acute care discharge services from a large home health agency in New York City in 2019 (n = 45,390 enrollment episodes).

METHODS

We developed a natural language processing algorithm for identifying information documented in free-text clinical notes (n = 1,429,030 notes) related to 4 categories: evidence of close relationships, evidence of advance directives, evidence suggesting lack of close relationships, and evidence suggesting lack of advance directives. We validated the algorithm against Gold Standard clinician review for 50 patients (n = 314 notes) to calculate precision, recall, and F-score.

RESULTS

Algorithm performance for identifying text related to the 4 categories was excellent (average F-score = 0.91), with the best results for "evidence of close relationships" (F-score = 0.99) and the worst results for "evidence of advance directives" (F-score = 0.86). The algorithm identified 22% of all clinical notes (313,290 of 1,429,030) as having text related to 1 or more categories. More than 98% of enrollment episodes (48,164 of 49,141) included at least 1 clinical note containing text related to 1 or more categories.

CONCLUSIONS AND IMPLICATIONS

This study establishes the feasibility of creating an automated screening algorithm to aid home health care agencies with identifying patients at risk of becoming INEADS. This screening algorithm can be applied as part of a multipronged approach to facilitate clinician support for advance care planning with patients at risk of becoming INEADS.

Preoperative Code Status Discussion Workflows: Targets for Improvement in Multidisciplinary Pathways

CONTEXT

Discussion of perioperative code status is an important element of preoperative care and a component of the American College of Surgeons' Geriatric Surgery Verification (GSV) program. Evidence suggests code status discussions (CSDs) are not routinely performed and are inconsistently documented.

OBJECTIVES

Because preoperative decision making is a complex process spanning multiple providers, this study aims to utilize process mapping to highlight challenges associated with CSDs and inform efforts to improve workflows and implement elements of the GSV program.

METHODS

Using process mapping, we detailed workflows relating to (CSDs) for patients undergoing thoracic surgery and a possible workflow for implementing GSV standards for goals and decision-making.

RESULTS

We generated process maps for outpatient and day-of-surgery workflows relating to CSDs. In addition, we generated a process map for a potential workflow to address limitations and integrate GSV Standards for Goals and Decision Making.

CONCLUSION

Process mapping highlighted challenges associated with the implementation of multidisciplinary care pathways and indicated a need for centralization and consolidation of perioperative code status documentation.

Actual vs Documented Advance Care Planning in a Primary Care Clinic

Introduction

Advance care planning (ACP) is a complex and multifaceted entity that has significant impact on patient care. ACP takes many forms, may be underbilled, and can have significant ramifications on quality care metrics. We performed a retrospective chart review for patients over 70 years in age in our family medicine resident clinic to evaluate the ways in which ACP is charted and the gap between billed and nonbilled ACP.

Methods

The first 50 patients over 70 years in age seen between August 25, 2020 and September 25, 2020 were selected for standardized chart review. Billing for ACP was defined as Current Procedural Terminology codes=-10 codes 99497 or 99498. Primary outcomes were the percentage of patients with ACP and incidence of ACP documents. Secondary outcome was the proportion of documented ACP conversations in office visits which had billing for ACP.

Results

Forty-eight patients over 70 years in age were identified with an average age of 80.9 years old. Forty-one of 48 patients (85.4%) had some form of ACP and 12 (25%) had formal ACP documents. Of 25 patients with documented ACP conversations in office visits, eleven patients (44%) had ACP which had been formally billed.

Conclusion

The majority of our patients had some form of ACP ranging from inpatient discussions of code status to outpatient visits regarding end-of-life care. However, ACP was underbilled in our practice. Physicians are often evaluated based on quality care metrics such as billed ACP which may not accurately reflect the work physicians are doing.

Advance care planning among Medicare beneficiaries with dementia undergoing surgery

IMPORTANCE

Advance care planning (ACP), in which patients or their surrogates discuss goals and preferences for care with physicians, attorneys, friends, and family, is an important approach to help align goals with actual treatment. ACP may be particularly valuable in patients with advanced serious illnesses such as Alzheimer's disease and related dementias (ADRDs) for whom surgery carries significant risks.

OBJECTIVE

To determine the frequency, timing, and factors associated with ACP billing in Medicare beneficiaries with ADRD undergoing nontrauma inpatient surgery.

DESIGN

This national cohort study analyzes Medicare fee-for-service claims data from 2016 to 2017. All patients had a 6-month lookback and follow-up period.

SETTING

National Medicare fee-for-service data.

PARTICIPANTS

All patients with ADRD, defined according to the Chronic Conditions Warehouse, undergoing inpatient surgery from July 1, 2016 to June 30, 2017.

EXPOSURES

Patient demographics, medical history, and procedural outcomes.

MAIN OUTCOME

ACP billing codes from 6 months before to 6 months after admission for inpatient surgery.

RESULTS

This study included 289,428 patients with ADRD undergoing surgery, of whom 21,754 (7.5%) had billed ACP within the 6 months before and after surgical admission. In a multivariable analysis, patients of white race, male sex, and residence in the Southern and Midwestern United States were at the highest risk of not receiving ACP. Of all patients who received ACP, 5960 (27.4%) did so before surgery while 12,658 (52.8%) received ACP after surgery. Timing of ACP after surgery was associated with an Elixhauser comorbidity index of 3 or higher (1.23, p = 0.045) and major postoperative complication or death (odds ratio 1.52, p < 0.0001).

CONCLUSIONS AND RELEVANCE

Overall ACP billing code use is low among Medicare patients with ADRD undergoing surgery. Billed ACP appears to have a reactive pattern, occurring most commonly after surgery and in association with postoperative mortality and complications. Additional study is warranted to understand barriers to use.