A Mixed Methods Analysis of Standardized Documentation of Serious Illness Conversations Within an Electronic Health Record Module During Hospitalization

Abstract

Background

Analysis of documented Serious Illness Conversations (SICs) in the inpatient setting can help clinicians align management to address patient and caregiver needs.

Methods

We conducted a mixed methods analysis of the first instance of standardized documentation of a SIC within a structured module among hospitalized general medicine patients from 2018 to 2019. Percentage of documentations that included a description of patient or family understanding of the patient’s medical condition and use of radio buttons to answer the “prognostic information shared,” “hopes,” and “worries” modules are reported. Using grounded theory approach, physicians analyzed free text entries to: “What is important to the patient/family?” and “Recommendations or next steps planned.”

Results

Out of 5142 patients, 59 patients had a documented SIC. Patient or family understanding of the medical condition(s) was reported in 56 (95%). For “prognostic information shared,” the most frequently selected radio buttons were: 49 (83%) incurable disease and 28 (48%) prognosis of weeks to months while those for “hopes” were: 52 (88%) be comfortable and 27 (46%) be at home and for “worries” were: 49 (83%) other physical suffering and 36 (61%) pain. Themes generated from entries to “What’s important to patient/family?” included being with loved ones; comfort; mentally and physically present; and reliable care while those for “Recommendations” were coordinating support services; symptom management; and support and communication.

Conclusions

SIC content indicated concern about pain and reliable care suggesting the complex, intensive nature of caring for seriously ill patients and the need to consider SICs earlier in the life course of patients.

Keywords

serious illness conversation inpatient end-of-life care palliative care caregiving hospice

Background

The purpose of conducting Serious Illness Conversations (SICs) is to build prognostic awareness, explore patient hopes and worries and align medical management to reflect each patient’s unique set of goals and priorities.¹ While serious illness communication may happen in the outpatient setting with a trusted provider 1-2 years prior to death,¹ the process is iterative and exists on a continuum.² Those conducted during hospitalization can be important for medical care depending on a patient’s acute medical issues and prognosis,^3,4 need for prompt medical decision-making or symptom management, presence of prior conversations,⁵ caregiver concern and caregiving intensity,⁶ availability of palliative care specialists in the outpatient setting,^7,8 and other factors. Many older adults with multiple chronic conditions experiencing an acute illness requiring hospitalization face complications including delirium,⁹ loss of function,¹⁰ and death¹¹ requiring multi-disciplinary and more intense and complex decision-making. Conducting SICs during hospitalization for the growing population of older adults with multiple chronic conditions supports patient-centered care, empowers patients, promotes patient dignity, and helps care teams assess and manage more complex needs.

Clinicians, including hospitalists at our institution, have previously received training in conducting SICs^12,13 and in documenting these conversations in a standardized manner using an institutionally approved electronic health record (EHR) module.^14,15

We present a mixed methods analysis with the aims of better understanding the characteristics of hospitalized patients with a first instance of standardized documentation of a SIC within a structured electronic health record (EHR) module and the content of SICs within this institutionally approved module for patients hospitalized on a general medicine service. We hypothesize that patients with standardized documentation of a SIC have high comorbidity and caregiving needs in addition to high mortality.

Methods

Study Design, Participants, Setting, Data Sources

We conducted retrospective, mixed methods analyses of adult (≥18 years old), general medicine inpatient encounters with a first instance of standardized documentation of a SIC within a structured electronic health record (EHR) module at Brigham and Women’s Faulkner Hospital (BWFH), a community teaching hospital affiliated with Mass General Brigham (MGB), a large integrated health care system in Massachusetts, from October 2018 to August 2019. All data were obtained from the enterprise data warehouse including data from our EHR (Epic Systems, Inc). The MGB institutional review board approved this study.

Quantitative Analyses: Patient Characteristics and Outcomes

We used descriptive statistics including percentage and mean (standard deviation (SD) to measure and report patient demographics: age, sex, ethnicity, race, primary language, insurance, income, comorbidity measures including Epic’s Risk of Unplanned Readmission (Readmission Risk Score [RRS]),¹⁶ Elixhauser/Van Walraven comorbidity score,¹⁷ and electronic Cardiac Arrest Risk Triage (eCART) (risk of death, cardiac arrest, or ICU transfer) score¹⁸ on admission, diagnosis related group (DRG) weight and hospitalizations in the year prior to index hospitalization. Clinical outcomes included length of stay (LOS), palliative care consult, discharge disposition, code status on discharge, and 6-month post-discharge mortality.

EHR SIC Module

The SIC EHR module is reflective of the key components of SICs¹⁹ and includes structured modules with use of radio buttons that physicians can use to describe clinicians’ “prognostic information shared,” and patient “hopes” and “worries” with the ability to include additional free-text comments. A provider can use more than 1 radio button within each structured module. As an example, a provider can choose ‘be comfortable,’ ‘be at home,’ and ‘achieve life goal’ radio buttons to document the hopes of their patient. The module also contains free-text questions to describe “patient illness understanding,” “What’s important to the patient/family?” and “Recommendations or next steps planned.” Providers in any setting, including the primary inpatient team, palliative care or geriatric consultants, and social workers, can fill out the module partially or in its entirety depending on what parts of the conversation happened during a clinical encounter. Documented modules are stored in a flowsheet data element and can be found quickly by clicking on a patient’s code status in the patient storyboard within the Epic EHR, making review of prior conversations readily accessible to all providers.

Quantitative SIC Module Entry Analysis

The number and percentage of standardized documentations of SICs within the EHR module that included a free text description of patient or family understanding of the patient’s medical condition are reported. The number and percentage of standardized documentations of SICs that used each radio button within the following structured SIC modules are reported: “prognostic information shared,” “hopes,” and “worries.”

Qualitative SIC Module Entry Analysis

Using grounded theory approach, a hospitalist (MKS) and a palliative care physician (JRL) analyzed the free text entries to 2 open-ended questions: “What is important to the patient/family?” and “Recommendations or next steps planned.” An initial coding schema was developed based on previously identified domains of serious illness conversations between patients and providers¹³ as well as a priori categories. Coding was performed independently by MKS and then reviewed with JRL. Disagreements were resolved with consensus-based discussion aimed at clarifying and refining the coding scheme. Themes were identified as they emerged. Microsoft Excel 2019 was used for thematic analysis.

Results

A total of 59 unique patients out of 5142 patients (6853 encounters) had a first standardized documentation of a SIC within the EHR module while admitted to an inpatient general medical service (Table 1) and had a mean age of 82 (standard deviation [SD] 11) years; 40 (68%) female; 43 (73%) non-Hispanic; 39 (66%) White; 43 (73%) Medicare insurance. The means (SD) for RRS,¹⁶ Elixhauser/Van Walraven comorbidity score,¹⁷ eCART score,¹⁸ and DRG weight were 22.9% (SD 15.1%), 19.3 (SD 10.2), 21.47 (SD 31.18), 2.00 (SD 1.41), respectively. Forty-six (78%) of patients had 1 or more hospitalizations in the year prior to the index admission.

Table 1.

Patient Demographics.

Total Patients	59
Age, mean (SD)	81.8 (10.9)
Female, n (%)	40 (67.8)
^aEthnicity, n (%)
Non-hispanic	43 (72.9)
Hispanic	5 (8.5)
^aRace, n (%)
White	39 (66.1)
Black	8 (13.6)
Other	8 (13.6)
^aLanguage, n (%)
English	49 (83.1)
Spanish	7 (11.9)
Insurance, n (%)
Medicare	43 (72.9)
Commercial	10 (17.0)
Other	6 (10.2)
Income, mean (SD)	75244 (21392)
b^,cRRS, mean (SD)	22.9 (15.1)
Elixhauser/Van walraven comorbidity score, mean (SD)	19.3 (10.2)
^beCart score, mean (SD)	21.47 (31.18)
DRG weight, mean (SD)	2.00 (1.41)
Hospitalizations in past year, n (%)
None	13 (22.0)
1	18 (30.5)
2-4	21 (35.6)
≥5	7 (11.9)

^aMissing Data: Ethnicity 11, Race 4, Language 3.

^bRisk of Unplanned 30-day Readmission (Readmission Risk Score) as calculated by Epic Systems, Inc

^cCalculated on admission.

These patients had a length of stay (LOS) of 11.6 (SD 10.0) days, and 45 (76%) patients had a palliative care consultation during their index hospitalization (Table 2). Five (8.5%) patients were discharged to home or self, 9 (15.3%) to home with health care services, 19 (32.2%) to a skilled nursing facility (SNF) or other facility, 13 (22.0%) to home hospice, 12 (20.3%) to inpatient hospice and 1 (1.7%) patient expired during the admission. Thirty-five (59.3%) patients had a code status of DNR/DNI on discharge. Eighteen (30.5%) patients died within 6-months of discharge from index hospitalization.

Table 2.

Patient Outcomes.

LOS in days—mean (SD)	11.6 (10.0)
Palliative care consult—no. (%)	45 (76.3%)
Discharge disposition—no. (%)
Home or self care	5 (8.5%)
Home with health care services	9 (15.3%)
Skilled nursing or other facility	19 (32.2%)
Home hospice	13 (22.0%)
Inpatient hospice	12 (20.3%)
Expired	1 (1.7%)
Code status at discharge—no. (%)
Full code	11 (18.6%)
DNR/DNI	35 (59.3%)
Comfort measures	13 (22.0%)
6-Month, post discharge mortality—no. (%)	18 (30.5%)

Fifty-six (95%) of standardized documentations of SIC within the structured module included an assessment of patient or family understanding of the medical condition(s) (Table 3). For the “prognostic information shared” module, the most frequently selected radio buttons by clinicians were: 49 (83%) incurable disease, 21 (36%) continued decline, 14 (24%) a prognosis of days to weeks and 28 (48%) a prognosis of weeks to months, respectively. The most frequently selected radio buttons for the “hopes” module were: 52 (88%) be comfortable, 27 (46%) be at home, 21 (36%) be mentally aware, and 21 (36%) be independent and for the “worries” module: 49 (83%) other physical suffering, 36 (61%) pain, 19 (32%) loss of control, and 19 (32%) being a burden.

Table 3.

Mixed Methods Analysis of Standardized Serious Illness Conversation Documentation.

Quantitative Analysis of Structured Modules
Structured modules	Clinician selected radio buttons
Patient or family understanding of medical condition assessed and documented—no. (%)	56 (95%)
Prognostic information shared—no. (%)	Incurable 49 (83%)
	Continued decline 21 (36%)
	Days to weeks 14 (24%)
	Weeks to months 28 (48%)
	Months to years 6 (10%)
	Uncertain 3 (5%)
	Missing 3 (5%)
Hopes—no. (%)	Be comfortable 52 (88%)
	Be at home 27 (46%)
	Be mentally aware 21 (36%)
	Be independent 21 (36%)
	Live as long as possible 13 (22%)
	Provide support for family 3 (5%)
	Achieve life goal 3 (5%)
Worries—no. (%)	Other physical suffering 49 (83%)
	Pain 36 (61%)
	Loss of control 19 (32%)
	Being a burden 19 (32%)
	Finances 7 (12%)
	Death 2 (3%)
Qualitative analysis of free- text entries
Open-ended questions	Themes and selected quotes
What is important to patient/family?	Being with loved ones
	“Being present for her daughter and grandson is priority.”
	“He lives for spending time with his family.”
	Comfort
	“Family is most concerned about comfort and maintaining quality care and dignity.”
	“Having pain/symptoms well controlled.”
	Mentally and physically present
	“Ability to enjoy social interactions”
	“Resuming activity outside the house”
	Reliable care
	“Providing a safe and compassionate environment, trust caregivers.”
	“Wants to make sure that her dad is safe, cared for somewhere that he won’t be scared and won’t get hurt”
Recommendations or next steps planned	Coordinating support services
	“Home with hospice upon discharge.”
	“Begin palliative care follow up at her assisted living.”
	Symptom management
	“Morphine PRN for air hunger, ativan PRN, may up titrate dose for agitation and anxiety”
	“Continue to work on pain control.”
	Patient and family support and communication
	“Support family through difficult decisions and conversations.”
	“Continue to educate about hospice.”

Themes generated from analysis of free-text responses to the open-ended question “What is important to the patient and family?” included being with loved ones; comfort; mentally and physically present; and reliable care. Coordinating support services, symptom management, patient and family support, and communication were themes generated from thematic analysis of free-text responses to the open-ended question “recommendations or next steps planned.” Illustrative quotes are included in Table 3.

Discussion

Unsurprisingly, a first structured standardized documentation of a SIC within an EHR module during hospitalization to a general medicine service occurred late in the life course of patients and frequently by or in conjunction with a palliative care consultation. Patients had elevated comorbidity burden and most had a recent prior hospitalization. Upon discharge, almost all of the patients either required social support through home health care services, skilled nursing or other facility admission or inpatient or home hospice. Approximately 30% of patients died within 6 months of discharge. This is reflected in the poor prognostic information shared by providers to patients: incurable or continued decline, days to weeks or weeks to months to live. The most frequently selected radio buttons for the “hopes” module were to be comfortable and to be at home while those for the “worries” module included other physical suffering and pain. Themes generated from analysis of free-text responses to the “What is important to patient/family?” open-ended question were being with loved ones; comfort; mentally and physically present; and reliable care. Coordinating support services; symptom management; and patient and family support and communication were themes generated from analysis of free text entries to “Recommendations or next steps planned.”

Patients admitted to medical services are among those with the highest mortality,²⁰ especially among patients aged 70 years or older.¹¹ Older patients who experience hospitalization often have functional and cognitive decline and require substantial assistance for activities of daily living (ADLs).²¹ For these patients especially, it is important to consider SICs in any setting, including during hospitalization, as the patient’s medical status changes, as their caregiving needs rise, and prior to the patient losing capacity, as most patients with serious illness are open and want to have these important conversations.^11,22 Prior to death, many patients experience hospitalization or other exposure²³ to providers, marking several opportunities to have a SIC or revisit this conversation with the patient in multiple settings and contexts to understand their priorities and provide goal concordant care. In this unique study, structured SIC documentation made it easy to aggregate results and learn the extent to which patients and their families did not want aggressive care at the end of life. We can assume that these would be the same wishes of many other hospitalized medical patients who did not have the opportunity to participate in an SIC.

The most often selected radio buttons for the “hopes” module closely mirrored many of the themes generated for the “What is important to patient/family?” open-ended question: comfort, being with loved ones, be mentally aware and be independent. These results are not surprising and are reflected in prior research.²⁴ Having “reliable care” is also important to patients and those closest to them with the ability to place trust in providers in a safe and compassionate environment.²⁵ This is especially relevant for this vulnerable population at the end of their life course and with high caregiving needs and is likely reflective of the challenges those caring for the patient at home face, including managing symptoms.^26,27 End-of-life care is complex and intensive requiring multi-disciplinary teamwork, collaboration and communication with the patient and loved ones, and optimal use of available resources to provide mental, physical and spiritual care of the dying patient.^28,29

Themes generated for the “Recommendations or next steps planned” open-ended question, including coordinating support services, symptom management and patient and family support and communication, are reflective of the challenging nature of caring for patients with high comorbidity at the end-of-life, including the difficulty in treating pain and other symptoms. It is helpful to have prior awareness of patient priorities through having SICs prior to this stage so the focus can be on medical management and alignment with these priorities. Talking with those providing care to the patient is important as the caregivers may require substantial support at home.³⁰ Additionally, most patients had a palliative care consult and these themes are also likely reflective of the additional, expert recommendations from palliative care consultants. This also draws attention to the need to have more training for primary providers to care for patients at the end-of-life.

Limitations of this study include that it is single site, low overall standardized documentation of SICs within the EHR module (thus leading to likely selection bias), and the inability to capture and analyze conversations recorded elsewhere within the EHR. In addition, documentation is only a proxy for the conversation itself and represents the clinician’s interpretation of the conversation. Lastly, the radio buttons, while useful to facilitate conversations, can also influence the choices made and documented.

In conclusion, first standardized documentation of a SIC within a structured module occurred late in the life course of patients admitted to a general medicine service and often by or in conjunction with a palliative care consultant. Analysis of SIC content indicated that most patients and families were concerned about pain or other physical suffering and having reliable care with good communication suggesting the complex and intensive nature of caring for patients with a serious illness and the need to support families and caregivers and provide additional resources and training to primary providers caring for these patients. SICs should be considered in hospitalized older adults with multiple chronic conditions, ideally when patients are still able to express their wishes, to ensure that they receive goal-concordant care. Providers should also be mindful of the burdensome nature of caregiving for a patient at the end-of-life when engaging in conversations about goals and values for seriously ill, hospitalized older adults.

Supplementary Materials: Please direct any questions and requests for data and/or models to the first author.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the a grant from the Controlled Risk Insurance Company (CRICO). CRICO had no role in the design or conduct of the study, the collection, analysis, or interpretation of data, or preparation or review of the manuscript. The conclusions in this report are those of the authors and do not necessarily represent the official position of AHR.

ORCID iD

Myrna Katalina Serna

References

Bernacki

Block

. Communication about serious illness care goals: a review and synthesis of best practices. JAMA Intern Med. 2014;174(12):1994. doi:10.1001/jamainternmed.2014.5271.

Hui

Bruera

. Models of palliative care delivery for patients with cancer. J Clin Orthod. 2020;38(9):852-865. doi:10.1200/JCO.18.02123.

Virdun

Luckett

Davidson

Phillips

, Dying in the hospital setting: a systematic review of quantitative studies identifying the elements of end-of-life care that patients and their families rank as being most important. Palliat Med. 2015;29(9):774-796. doi:10.1177/0269216315583032.

Bielinska

Archer

Obanobi

, Advance care planning in older hospitalised patients following an emergency admission: a mixed methods study. PLoS One. 2021;16(3):e0247874. doi:10.1371/journal.pone.0247874.

Heyland

. Failure to engage hospitalized elderly patients and their families in advance care planning. JAMA Intern Med. 2013;173(9):778. doi:10.1001/jamainternmed.2013.180.

Michaels

Chen

Ann Meeker

Navigating the caregiving abyss: a metasynthesis of how family caregivers manage end-of-life care for older adults at home. Palliat Med. 2022;36(1):81-94. doi:10.1177/02692163211042999.

Yeh

Urman

Besaw

Dodge

Lee

Buss

. Different associations between inpatient or outpatient palliative care and end-of-life outcomes for hospitalized patients with cancer. JCO Oncology Practice. 2022;18(4):e516-e524. doi:10.1200/OP.21.00546.

Lupu

Quigley

Mehfoud

Salsberg

. The growing demand for hospice and palliative medicine physicians: will the supply keep up? J Pain Symptom Manag. 2018;55(4):1216-1223. doi:10.1016/j.jpainsymman.2018.01.011.

Marcantonio

. In The clinic. Delirium. Ann Intern Med 2011;154(11):ITC6, ITC6-2, ITC6-3, ITC6-4, ITC6-5, ITC6-6, ITC6-7, ITC6-8, ITC6-9, ITC6-10, ITC6-11, ITC6-12, ITC6-13, ITC6-14, ITC6-15; quiz ITC6-16. doi:10.7326/0003-4819-154-11-201106070-01006

10.

Chen

Almirall-Sanchez

Mockler

Adrion

Dominguez-Vivero

Romero-Ortuno

. Hospital-associated deconditioning: not only physical, but also cognitive. Int J Geriatr Psychiatr. 2022;37(3).

11.

Bielinska

Archer

Obanobi

, et al. Advance Care Planning in Older Hospitalised Patients Following an Emergency Admission: A Mixed Methods Study. In: Cheungpasitporn

, ed. PLoS One. 2021;16(3):e0247874. doi:10.1371/journal.pone.0247874.

12.

Lakin

Arnold

Catzen

, et al. Early serious illness communication in hospitalized patients: a study of the implementation of the speaking about goals and expectations (Sage) program. Healthcare. 2021;9(2):100510. doi:10.1016/j.hjdsi.2020.100510.

13.

Lakin

Koritsanszky

Cunningham

, et al. A systematic intervention to improve serious illness communication in primary care. Health Aff. 2017;36(7):1258-1264. doi:10.1377/hlthaff.2017.0219.

14.

Lakin

Gundersen

Lindvall

, et al. A yet unrealized promise: structured advance care planning elements in the electronic health record. J Palliat Med. 2021;24(8):1221-1225. doi:10.1089/jpm.2020.0774.

15.

Wilson

Bernacki

Lakin

Alexander

Jackson

Jacobsen

. Rapid adoption of a serious illness conversation electronic medical record template: lessons learned and future directions. J Palliat Med. 2020;23(2):159-161. doi:10.1089/jpm.2019.0420.

16.

Serna

Fiskio

Yoon

, et al. Who gets (and who should get) a serious illness conversation in the hospital? An analysis of readmission risk score in an electronic health record. Am J Hosp Palliat Care. 2022;26:104990912211296. doi:10.1177/10499091221129602.

17.

van Walraven

Austin

Jennings

Quan

Forster

. A modification of the elixhauser comorbidity measures into a point system for hospital death using administrative data. Med Care. 2009;47(6):626-633. doi:10.1097/MLR.0b013e31819432e5.

18.

Churpek

Yuen

Winslow

, et al. Multicenter development and validation of a risk stratification tool for ward patients. Am J Respir Crit Care Med. 2014;190(6):649-655. doi:10.1164/rccm.201406-1022OC.

19.

Paladino

Koritsanszky

Nisotel

, et al. Patient and clinician experience of a serious illness conversation guide in oncology: a descriptive analysis. Cancer Med. 2020;9(13):4550-4560. doi:10.1002/cam4.3102.

20.

Tonelli

Wiebe

Manns

, et al. Comparison of the complexity of patients seen by different medical subspecialists in a universal health care system. JAMA Netw Open. 2018;1(7):e184852. doi:10.1001/jamanetworkopen.2018.4852.

21.

Lunney

. Patterns of functional decline at the end of life. JAMA. 2003;289(18):2387. doi:10.1001/jama.289.18.2387.

22.

Geerse

Lamas

Sanders

, et al. A qualitative study of serious illness conversations in patients with advanced cancer. J Palliat Med. 2019;22(7):773-781. doi:10.1089/jpm.2018.0487.

23.

Barbiellini Amidei

Macciò

Cantarutti

, et al. Hospitalizations and emergency department visits trends among elderly individuals in proximity to death: a retrospective population-based study. Sci Rep. 2021;11:21472. doi:10.1038/s41598-021-00648-1.

24.

Lewis

Harrison

Hanly

, et al. End‐of‐life priorities of older adults with terminal illness and caregivers: a qualitative consultation. Health Expect. 2019;22(3):405-414. doi:10.1111/hex.12860.

25.

Hanson

Danis

Garrett

What is wrong with end-of-life care? Opinions of bereaved family members. J Am Geriatr Soc. 1997;45(11):1339-1344. doi:10.1111/j.1532-5415.1997.tb02933.x.

26.

Mather

Kleijwegt

Bollens-Lund

Kelley

Ornstein

Symptom management experience of end-of-life family caregivers: a population-based study. J Pain Symptom Manag. 2022;64(6):513-520. doi:10.1016/j.jpainsymman.2022.07.017.

27.

McMillan

Small

. Symptom distress and quality of life in patients with cancer newly admitted to hospice home care. Oncol Nurs Forum. 2002;29(10):1421. doi:10.1188/02.ONF.1421-1428.

28.

Bloomer

Poon

Runacres

Hutchinson

Facilitating family needs and support at the end of life in hospital: a descriptive study. Palliat Med. 2022;36(3):549-554. doi:10.1177/02692163211066431.

29.

Harasym

Afzaal

Brisbin

, et al. Multi-disciplinary supportive end of life care in long-term care: an integrative approach to improving end of life. BMC Geriatr. 2021;21:326. DOI: 10.1186/s12877-021-02271-1.

30.

Mogan

Harrison Dening

Dowrick

Lloyd-Williams

Health and social care services for people with dementia at home at the end of life: a qualitative study of bereaved informal caregivers’ experiences. Palliat Med. 2022;36(6):976-985. doi:10.1177/02692163221092624.