Age and Education Are Associated with Advance Care Planning Among Women with Metastatic Breast Cancer

Abstract

Objective:

To assess the frequency of advance care planning (ACP) and associated demographic and clinical factors among women with metastatic breast cancer (MBC).

Methods:

We analyzed secondary data from a longitudinal study of quality of life among women with MBC. Participants (N = 118) were recruited from breast oncology clinics at an NCI-designated comprehensive cancer center. Participants completed surveys assessing demographic and clinical characteristics. Data on ACP completion (yes/no) were abstracted from the electronic medical record. Chi-square or t-tests assessed bivariate relationships between demographic and clinical characteristics and ACP completion. Characteristics that were significantly related (p < 0.01) were entered into a multivariable logistic regression.

Results:

Only 38% (45/118) of participants had completed ACP. In multivariable analyses, older age and greater education were positively associated with ACP completion (ORage = 1.04 [1.01–1.08]; OReducation = 2.32 [1.04–5.20]).

Conclusions:

ACP interventions should target women with MBC who are younger and with lower levels of educational attainment.

Keywords

advance care planning advance directives advanced cancer metastatic breast cancer

Introduction

Advance care planning (ACP) is the process of discussing and recording patient preferences for future medical care.¹ ACP is associated with reduced hospitalizations, increased use of hospice and palliative care, and decreased cost of care at the end of life.^2–5 ACP is a key component of the National Comprehensive Cancer Network (NCCN) guidelines for palliative care⁶ and is a Centers for Medicare and Medicaid services care quality indicator for all individuals aged 65 and older.⁷

However, only 12–34% of patients with advanced cancer have completed ACP.^8–11 When ACP does occur, it is often in the last few months of life and/or in the inpatient setting.^8,12 This contrasts with NCCN guidelines⁶ and patient preferences for ACP to occur early.¹³ Thus, improvements in ACP are urgently needed in order to provide quality cancer care.

Understanding which individuals are more or less likely to have completed ACP may guide tailored intervention strategies. The metastatic breast cancer (MBC) population is ideal for addressing this research question, as MBC is typically considered to be incurable,¹⁴ but novel treatment approaches have resulted in significantly increased survival times following MBC diagnosis.^15–17 This population can thus provide valuable, novel insight into ACP among those for whom ACP is salient, but are not at the end of life. Leveraging secondary data from an observational study of quality of life among women with MBC,^18,19 we aimed to: (1) characterize the frequency of ACP among women with MBC and (2) identify specific subgroups of women with MBC who are less likely to have completed ACP.

Methods

Procedures and participants

We conducted secondary data analyses from a longitudinal study of quality of life among women with MBC.^18,19 All procedures were reviewed and approved by the Institutional Review Board at Georgetown University (IRB #00005081).

Participants were recruited between January 2023 and April 2024 from breast medical oncology clinics in Washington, DC, Baltimore, MD, and Hackensack, NJ. Eligibility criteria and recruitment procedures for the parent study have been previously reported.^18,19 Briefly, eligible participants were assigned female at birth; age ≥18; diagnosed with MBC; able to speak English or Spanish; and had a working telephone capable of receiving SMS messages and accessing the Internet (due to the design of the parent study). Participants completed a brief baseline survey and four weeks of follow-up surveys. Upon completion of the follow-up period, trained research staff manually reviewed the electronic medical record (EMR) and abstracted key variables of interest during the period between study enrollment and completion or withdrawal (M = 58 days, standard deviation [SD] = 36 days).

Measures

Demographic characteristics

Participants self-reported age, race, ethnicity, education level, partner status, and current caregiving responsibilities.

Comorbidities

We used a simplified comorbidity index based on the Charlson Comorbidity Index.^20–22 Participants indicated whether they had been diagnosed with any of 12 health conditions (e.g., heart attack, asthma), and up to two “other” medical conditions. Possible scores were 0 to 14 (higher = more comorbidities).

Health literacy

The BRIEF Health Literacy Screening Tool^23,24 includes 4 items rated on a 5-point Likert scale (1 = “never” to 5 = “always”). Total scores range from 4 to 20 (higher = better); scores ≥16 indicate adequate health literacy.

Clinical characteristics

Location of MBC care (Washington, Baltimore, or NJ) and time since MBC diagnosis were abstracted from the EMR.

ACP completion

EMR-documented completion of ACP (yes/no) included: (1) a Medical Orders for Scope of Treatment form, a Medical Order for Life Sustaining Treatment form, or a Physician Orders for Life Sustaining Treatment form; (2) an advance directive or living will; and/or (3) a medical durable power of attorney.

Data analysis

Analyses were conducted with IBM SPSS (version 29). Cases with missing data were removed listwise. Chi-square or t-tests (as appropriate) assessed bivariate relationships between demographic and clinical characteristics and ACP completion. Characteristics that were significantly related (p < 0.10) at the bivariate level were subsequently entered into a multivariable logistic regression model. Goodness of fit was evaluated by Hosmer and Lemeshow’s test.²⁵

Results

Of 125 consenting patients, 118 (94%) completed the baseline survey and were included in these analyses (see Table 1). About one-third (n = 45, 38%) had completed ACP. In bivariate analyses (see Table 2), ACP completion was significantly associated with age (p = 0.037), such that participants who had completed ACP were older on average (M = 61 years) than participants who had not completed ACP (M = 56 years). ACP completion was also marginally associated with education (p = 0.089), such that participants with a Bachelor’s degree or greater were more likely to have completed ACP (45%) than those with less than a Bachelor’s degree (29%).

Table 1.

Participant Demographic and Clinical Characteristics, Overall, and by ACP Completion

	Total (N = 118)	Completed ACP (n = 45)	No ACP (n = 73)	p ^a
Age (M, SD)	57.6 (12.2)	60.6 (12.7)	55.8 (11.5)	0.037
Race and ethnicity^b (n, %)
American Indian/Alaska Native	1 (0.8)	1 (2.2)	0 (0.0)	0.381
Asian	9 (7.6)	4 (8.9)	5 (6.8)	0.730
Black/African American	42 (35.6)	15 (33.3)	27 (40.0)	0.687
Hispanic/Latina	12 (10.2)	3 (6.7)	9 (12.3)	0.369
Middle Eastern/North African	2 (1.7)	0 (0.0)	2 (2.7)	0.524
Native Hawaiian/Pacific Islander	1 (0.8)	0 (0.0)	1 (1.4)	1.000
White	58 (49.2)	25 (55.6)	33 (45.2)	0.275
Education level				0.089
<Bachelor’s degree	51 (43.2)	15 (33.3)	36 (49.3)
≥Bachelor’s degree	67 (56.8)	30 (66.7)	37 (50.7)
Partner status (n, %)				0.647
Partnered	62 (52.5)	22 (48.9)	40 (54.8)
Not partnered	53 (44.9)	21 (46.7)	32 (43.8)
Missing/Prefer not to answer	3 (2.5)	2 (4.4)	1 (1.4)
Caregiving responsibilities (n, %)				0.431
Yes	40 (33.9)	17 (37.8)	23 (31.5)
No	77 (65.3)	27 (60.0)	50 (68.5)
Missing/Prefer not to answer	1 (0.8)	1 (2.2)	0 (0.0)
Number of comorbidities (M, SD)	2.6 (1.8)	2.56 (1.89)	2.64 (1.71)	0.806
BRIEF health literacy score (n, %)				0.138
Low/Marginal (≤16)	24 (20.3)	6 (13.3)	18 (24.7)
Adequate (≥17)	94 (79.7)	39 (86.7)	55 (75.3)
Years since MBC diagnosis (M, SD)	4.3 (4.4)	4.4 (4.7)	4.2 (4.2)	0.826
Location of care				0.589
Washington, DC	68 (57.6)	28 (62.2)	40 (54.8)
Baltimore, MD	16 (13.6)	6 (13.3)	10 (13.7)
Hackensack, NJ	34 (28.8)	11 (24.4)	23 (31.5)

Bold values indicate p-values less than 0.1.

p Value for chi-square test, Fisher’s exact test, or student’s t-test, as appropriate.

As participants could select more than one response, it does not total 100%.

ACP, advance care planning.

Table 2.

Results of Multivariable Logistic Regression Model Examining Demographic and Clinical Characteristics Associated with ACP Completion (N = 118)

Explanatory variable	OR	95% CI	p
Age (years)	1.04	[1.01, 1.08]	0.019
Education (ref: <Bachelor’s degree)	2.32	[1.04, 5.20]	0.041

Age and education were entered in a multivariable logistic regression model with ACP completion specified as the dependent variable. Both age and education remained significant in the multivariable analyses. Participants were 4% more likely to have completed ACP for each additional year of age (Odds Ratio = 1.04, 95% Confidence Interval: 1.01–1.08). Compared to participants with less than a Bachelor’s degree, participants with a Bachelor’s degree or greater were more than twice as likely to have completed ACP (Odds Ratio = 2.32, 95% Confidence Interval: 1.04–5.20). The final multivariable model demonstrated adequate goodness of fit (χ²=5.55, p = 0.698).

Discussion

This study extends the prior literature by assessing the frequency of EMR-documented ACP and associated demographic and clinical factors among women with MBC. Rates of ACP among women with MBC were similar to those previously reported for patients with advanced cancer^8–11 and suggest room for improvement.

Participants who were younger and with lower levels of education were less likely to have EMR-documented ACP. While the relationship between education level and ACP completion is well-documented in other populations,^26–28 the prior literature on age and ACP is less clear. While some studies show that ACP engagement increases with age,^26,29–32 Prater and colleagues³³ found that odds of having documented ACP decreased with age, when controlling for prognosis. These conflicting results may suggest a more complex, nonlinear relationship between age and ACP completion. In the context of MBC, our results suggest additional outreach may be needed to encourage ACP among younger patients and those with lower levels of education. Tools and educational materials tailored to life stages and in plain language may have particular utility.³⁴

Prior research demonstrates that non-Hispanic White cancer patients are more likely to have completed ACP than Black and Latino cancer patients,^35,36 but we did not observe differences in ACP completion by participant race or ethnicity. While we enrolled a relatively diverse sample in terms of race and ethnicity, they were homogenous in other ways: all were receiving care at an NCI-designated comprehensive cancer center, most had a college education or greater, and nearly all had adequate levels of health literacy. Given the complex relationships among race, ethnicity, socioeconomic status, and health care access in the United States,³⁷ a larger and/or more socioeconomically diverse sample may be needed in order to detect disparities in ACP completion. Additionally, these are secondary data analyses, and the parent study was not specifically designed to examine ACP completion among women with MBC. Thus, it is possible that we were underpowered to detect differences in ACP completion by race and ethnicity. Additional larger studies, specifically designed to examine ACP, are needed to fully characterize the landscape of ACP among women with MBC.

ACP completion is a complex, multifaceted process involving multiple stakeholders. While our study design only allowed us to examine patient-level factors that may be associated with ACP completion, barriers also exist at the provider, organizational, and health care system levels. For example, fragmented communication about ACP, discomfort discussing ACP, staffing limitations/turnover, time burden for ACP, competing clinical demands, and unclear division of tasks related to ACP have all been identified as barriers to ACP completion.^38,39 Interventions aiming to increase rates of ACP completion must therefore act at multiple levels of influence, including the patient, provider, organization, and health care system.³ The Consolidated Framework for Implementation Research⁴⁰ is widely used to identify factors that affect the implementation of new practices or interventions. It organizes these factors across five key domains (intervention characteristics, outer setting, inner setting, characteristics of individuals, and process), providing a common language and structure for implementation assessment and planning. Using this framework to understand the multiple levels of influence on ACP completion would support the development of targeted interventions that address not only individual but also health care system or community level influences.

Study Limitations

First, the use of EMR data is an imperfect way to assess ACP.^41,42 We did not capture ACP data from outside of our system, and it is possible that participants had completed ACP elsewhere. Future studies might supplement EMR data with patient-reported data to fully capture ACP across clinical care settings. Second, data were drawn from a study on quality of life in MBC. Participants thus reflect a subgroup that was able and willing to participate in this study. The specific eligibility characteristics of the parent study (i.e., smartphone ownership) may have further limited our sample to individuals with higher socioeconomic status or levels of digital literacy. Finally, most participants in this study were non-Hispanic White and highly educated. Results may not generalize to the broader, socioeconomically or racially/ethnically diverse population of people with MBC.^43,44

Conclusions

Women with MBC who are younger and with lower levels of educational attainment may benefit from interventions to increase ACP completion. Furthermore, levels of ACP completion overall are low due to complex individual and system-level issues that need to be further assessed and addressed in order to complete this evidence-based practice more broadly across MBC patients.

Authors’ Contributions

All authors contributed to the study conception and design. Material preparation and data collection were performed by J.S., M.V.-S., A.B., A. M., H.K., and J.D.R. Data analysis was performed by C.C.C. and J.S. The first draft of the article was written by C.C.C. All the authors commented on subsequent versions of the article. All the authors read and approved the final article.

Footnotes

Acknowledgment

The authors would like to thank Anastacia Wahl for her assistance with EMR review.

Author Disclosure Statement

C.C.C. has received research support (to institution) from Pfizer. C.I. has participated in consulting activities for which she has received honoraria from AstraZeneca, Genentech, Seattle Genetics, Pfizer, Gilead, and Merck; royalties from Wolters Kluwer (UptoDate) and McGraw Hill (Goodman and Gillman); and research support (to institution) from Tesaro/GSK, Seattle Genetics, Pfizer, AztraZeneca, Bristol Myers Squibb, Genentech, and Novartis. S.C.O’N. has received research support (to institution) from Pfizer and Gilead Sciences. No other authors have competing interests to disclose.

Funding Information

This work was supported by the National Cancer Institute (R03 CA273489, PI: C.C.C.; T32CA261787, PI: S.C.O’N., Adams-Campbell, Schwartz; P30 CA051008, PI: Weiner). The content presented here is solely the responsibility of the authors and does not represent the official views of the National Cancer Institute.

Data Availability Statement

The data that support the findings of this study are openly available on the Open Science Framework at .

Abbreviations

References

Sudore

, Lum

, You

, et al. Defining advance care planning for adults: A consensus definition from a multidisciplinary Delphi panel. J Pain Symptom Manage, 2017; 53(5):821–832.e1.

Brinkman-Stoppelenburg

, Rietjens

, Van der Heide

. The effects of advance care planning on end-of-life care: A systematic review. Palliat Med, 2014; 28(8):1000–1025.

Bestvina

, Polite

. Implementation of advance care planning in oncology: A review of the literature. J Oncol Pract, 2017; 13(10):657–662.

Hoerger

, Greer

, Jackson

, et al. Defining the elements of early palliative care that are associated with patient-reported outcomes and the delivery of end-of-life care. J Clin Oncol, 2018; 36(11):1096–1102.

Zhang

, Wright

, Huskamp

, et al. Health care costs in the last week of life: Associations with end-of-life conversations. Arch Intern Med, 2009; 169(5):480–488.

National Comprehensive Cancer Network (NCCN). Palliative Care (Version 2.2025) 2025 March 31, 2025. Available from: https://www.nccn.org/professionals/physician_gls/pdf/palliative.pdf

US Centers for Medicare and Medicaid Services (CMS). Quality ID #47 (NQF 0326): Advance Care Plan: US Centers for Medicare and Medicaid Services (CMS). 2022. Available from: https://qpp.cms.gov/docs/QPP_quality_measure_specifications/CQM-Measures/2023_Measure_047_MIPSCQM.pdf

Mack

, Cronin

, Taback

, et al. End-of-life care discussions among patients with advanced cancer: A cohort study. Ann Intern Med, 2012; 156(3):204–210.

Clark

, Ott

, Rogers

, et al. Advance care planning as a shared endeavor: Completion of ACP documents in a multidisciplinary cancer program. Psychooncology, 2017; 26(1):67–73.

10.

Temel

, Greer

, Admane

, et al. Code status documentation in the outpatient electronic medical records of patients with metastatic cancer. J Gen Intern Med, 2010; 25(2):150–153.

11.

McDermott

, Engelberg

, Sibley

, et al. The association between chronic conditions, end-of-life health care use, and documentation of advance care planning among patients with cancer. J Palliat Med, 2020; 23(10):1335–1341.

12.

Enguidanos

, Ailshire

. Timing of advance directive completion and relationship to care preferences. J Pain Symptom Manage, 2017; 53(1):49–56.

13.

Kubi

, Istl

, Lee

, et al. Advance care planning in cancer: Patient preferences for personnel and timing. JCO Oncol Pract, 2020; 16(9):e875–e883.

14.

Liang

, Zhang

, Song

, Yang

. Metastatic heterogeneity of breast cancer: Molecular mechanism and potential therapeutic targets. Semin Cancer Biol, 2020; 60:14–27.

15.

Caswell-Jin

, Plevritis

, Tian

, et al. Change in survival in metastatic breast cancer with treatment advances: Meta-analysis and systematic review. JNCI Cancer Spectr, 2018; 2(4):pky062.

16.

Santa-Maria

, Gradishar

. Changing treatment paradigms in metastatic breast cancer: Lessons learned. JAMA Oncol, 2015; 1(4):528–534; quiz 549.

17.

Sundquist

, Brudin

, Tejler

. Improved survival in metastatic breast cancer 1985–2016. Breast, 2017; 31:46–50.

18.

Mossman

, Small

, Addington

, et al. Longitudinal quality of life in adults with metastatic breast cancer: The impact of between- and within-person changes in symptoms and positive affect. Cancer Med-US. Under Review, 2025.

19.

Mossman

, Small

, Velazquez-Sosa

, et al. Ecological momentary assessment of quality of life in women with metastatic breast cancer. American Preventive Oncology Society Annual Meeting; Pittsburgh, PA2025.

20.

Potosky

, Graves

, Lin

, et al. The prevalence and risk of symptom and function clusters in colorectal cancer survivors. J Cancer Surviv, 2022; 16(6):1449–1460.

21.

Potosky

, Ahn

, Xia

, et al. Demographic and clinical factors associated with health-related quality-of-life profiles among prostate cancer survivors. JCO Oncol Pract, 2024; 20(7):921–931.

22.

Huq

, Schwartz

, Derry-Vick

, et al. Cancer survivor preferences on the timing and content of interventions to mitigate financial toxicity associated with cancer treatment. Support Care Cancer, 2024; 32(12):1–9.

23.

Haun

, Noland-Dodd

, Varnes

, et al. Testing the BRIEF health literacy screening tool. Fed Pract, 2009; 26(12):24–31.

24.

Haun

, Patel

, French

, et al. Association between health literacy and medical care costs in an integrated healthcare system: A regional population based study. BMC Health Serv Res, 2015; 15(1):249.

25.

Hosmer

, Lemesbow

. Goodness of fit tests for the multiple logistic regression model. Comm in Stats—Theory & Methods, 1980; 9(10):1043–1069.

26.

Lovell

, Yates

. Advance care planning in palliative care: A systematic literature review of the contextual factors influencing its uptake 2008–2012. Palliat Med, 2014; 28(8):1026–1035.

27.

Inoue

. The influence of sociodemographic and psychosocial factors on advance care planning. J Gerontol Soc Work, 2016; 59(5):401–422.

28.

Korkmaz Yaylagul

, Demirdas

, Melo

, Silva

. Opinions of older individuals on advance care planning and factors affecting their views: A systematic review. Int J Environ Res Public Health, 2023; 20(10):5780.

29.

Musa

, Seymour

, Narayanasamy

, et al. A survey of older peoples’ attitudes towards advance care planning. Age Ageing, 2015; 44(3):371–376.

30.

Howard

, Bernard

, Klein

, et al. Older patient engagement in advance care planning in Canadian primary care practices: Results of a multisite survey. Can Fam Physician, 2018; 64(5):371–377.

31.

Raijmakers

NJH

, Rietjens

JAC

, Kouwenhoven

PSC

, et al. Involvement of the Dutch general population in advance care planning: A cross-sectional survey. J Palliat Med, 2013; 16(9):1055–1061.

32.

De Vleminck

, Pardon

, Houttekier

, et al. The prevalence in the general population of advance directives on euthanasia and discussion of end-of-life wishes: A nationwide survey. BMC Palliat Care, 2015; 14(1):71.

33.

Prater

, Wickizer

, Bose-Brill

. Examining age inequalities in operationalized components of advance care planning: Truncation of the ACP process with age. J Pain Symptom Manage, 2019; 57(4):731–737.

34.

Riley

, Voisin

, Stevens

, et al. Tools for tomorrow: A scoping review of patient-facing tools for advance care planning. Palliat Care Soc Pract, 2024; 18:26323524241263108.

35.

Berkowitz

, Wolf

, Troy

, Kamal

. Characteristics of advance care planning in patients with cancer referred to palliative care. JCO Oncol Pract, 2021; 17(2):e94–e100.

36.

Kelly

, Henderson

, Hyer

, Pawlik

. Intrapersonal factors impact advance care planning among cancer patients. Am J Hosp Palliat Care, 2021; 38(8):907–913.

37.

Williams

, Priest

, Anderson

. Understanding associations among race, socioeconomic status, and health: Patterns and prospects. Health Psychol, 2016; 35(4):407–411.

38.

Stevens

, Elston

, Tan

, et al. Clinicians’ experiences implementing an advance care planning pathway in two Canadian provinces: A qualitative study. BMC Prim Care, 2024; 25(1):217.

39.

Lund

, Richardson

, May

. Barriers to advance care planning at the end of life: An explanatory systematic review of implementation studies. PLoS One, 2015; 10(2):e0116629.

40.

Damschroder

, Aron

, Keith

, et al. Fostering implementation of health services research findings into practice: A consolidated framework for advancing implementation science. Implementation Sci, 2009; 4(1):1–15.

41.

Sean Morrison

. Advance directives/care planning: Clear, simple, and wrong. J Palliat Med, 2020; 23(7):878–879.

42.

Geerse

, Lamas

, Bernacki

, et al. Adherence and concordance between serious illness care planning conversations and oncology clinician documentation among patients with advanced cancer. J Palliat Med, 2021; 24(1):53–62.

43.

Gage-Bouchard

, Rodriguez

, Saad-Harfouche

, et al. Factors influencing patient pathways for receipt of cancer care at an NCI-designated comprehensive cancer center. PLoS One, 2014; 9(10):e110649.

44.

Dickson-Witmer

, Bleznak

, Kennedy

, et al. Breast cancer care in the community: Challenges, opportunities, and outcomes. Surg Oncol Clin N Am, 2011; 20(3):555–580, ix.