Project to Improve Communication About Serious Illness--Hospital Study: Comparative Effectiveness Trial (Trial 2) (PICSI-H)

Project to Improve Communication About Serious Illness--Hospital Study: Comparative Effectiveness Trial (Trial 2)

Using the Electronic Health Record to Identify and Promote Goals-of-Care Communication for Older Patients With Serious Illness

The objective of this protocol is to test the effectiveness of a Jumpstart intervention on patient-centered outcomes for patients with chronic illness by ensuring that they receive care that is concordant with their goals over time, and across settings and providers. This study is particularly interested in understanding the effect of the intervention to improve quality of palliative care for patients with Alzheimer's disease and related dementias (ADRD) but will also include other common chronic, life-limiting illnesses. The specific aims are: To evaluate the efficacy of the Survey-based Patient/Clinician Jumpstart compared to the EHR based clinician Jumpstart and usual care for improving quality of care; the primary outcome is EHR documentation of a goals-of-care discussion from randomization through hospitalization or 30 days. Secondary outcomes include: a) intensity of care outcomes (e.g., ICU use, ICU and hospital length of stay, costs of care during the hospitalization, 7 and 30 day readmission); and b) patient- and family-reported outcomes assessed by surveys at 3 days and 4 weeks after randomization, including occurrence and quality of goals-of-care discussions in the hospital, goal-concordant care, psychological symptoms, and quality of life. To conduct a mixed-methods evaluation of the implementation of the intervention, guided by the RE-AIM framework for implementation science, incorporating quantitative evaluation of the intervention's reach and adoption, as well as qualitative analyses of interviews with participants, to explore barriers and facilitators to future implementation and dissemination.

OVERVIEW: In this comparative effectiveness trial, the investigators will examine the effectiveness of two approaches to the Jumpstart intervention designed to promote goals-of-care discussions for older, seriously ill, hospitalized patients. The trial recruits consecutively eligible patients from two UW Medicine hospitals. The Jumpstart Guide is a communication-priming intervention for clinicians, patients and their families that focuses on hospitalized patients' goals of care. There are two versions. One, the "EHR-based Clinician Jumpstart", is drawn from the electronic medical record (EHR) and lists the dates and locations of prior advance care planning documents (e.g. living wills, healthcare directives, durable power of attorney for healthcare, and Physician Orders for Life Sustaining Treatments). It also includes tips to improve this communication. This information is provided by email and in-person to patients' clinicians. The other is "Survey-based Patient/Clinician Jumpstart," and provides the same information as in the "EHR-based Clinician Jumpstart" but, in addition, goals of care information from patient's (or surrogates) self-reported surveys are summarized and shared with the patients' clinicians, the participating patient and family. This trial tests the effect of the ERH-based, clinician Jumpstart against the Survey-based Patient/Clinician Jumpstart as well as usual care. Unique to this trial is the use of the EHR to identify eligible participants, provide data for the intervention, and be the mechanism for delivering the intervention. This current study is "Trial 2" of the R01 Award funding this trial. Trial 1 was completed prior to the initiation of Trial 2. SPECIFIC AIM 1 (for TRIAL 2): Evaluate the efficacy of the Survey-based Patient/Clinician Jumpstart compared to the EHR-based Clinician Jumpstart and usual care for improving quality of care. TRIAL 2 has four components. Component 1- Subject Identification/Recruitment/Randomization: Patients who meet the inclusion criteria are screened and identified using daily screening reports and staff review. The investigators oversample patients with ADRD to include 40% of the sample. Patients are approached by study staff in person during their hospital stay to assess their interest in participating in the study. Recruitment conversations are designed to take place in the patient's hospital room. Subjects are asked to complete surveys at three time points: 1) at enrollment; 2) 3 days after randomization; and 3) 4 weeks after randomization. Follow-up surveys may be completed in-person, by paper, online, or by phone, based on respondents' preferences. If patients are not able to participate themselves (e.g., cognitive impairment, sedated or ventilated), the investigators recruit their legal surrogate decision-maker to participate. This surrogate (under Washington State Law RCW 7.70.065) provides consent on their own behalf and is a research study subject. Eligible patients are assigned to one of the three interventions in a 1:1:1 ratio. Patients are randomized using variable size blocks and stratified for hospital and ADRD vs. no ADRD. Surrogates/families follow the randomization status of the patients whom they are representing. Component 2- EHR-based Clinician Jumpstart Guide: The Jumpstart guide is developed from an automated review of the EHR. It summarizes the presence/absence of POLST, advance directives and DPOA documentation. It provides general recommendations to initiating goals of care discussions. Component 3- Survey-based Patient/Clinician Jumpstart Guide: Survey data, completed by patients or their surrogate/family at enrollment provide assessments of the following: a) preferences for discussions about goals of care; b) most important barrier and facilitator for having such discussions; and c) current goals of care. These elements are contained within the Jumpstart guides and the information is tailored to each recipient (i.e., patient, surrogate/family, or clinician). This version provides survey-response specific recommendations to initiate goals of care discussions. Component 4- Delivery of the intervention: Clinician delivery: Guides are delivered within the first few days (1-3) after randomization to the primary clinician team (attending and resident physicians and advanced practice providers) via secure email or in person. Study staff monitor the care team for the patient and ensure that any new providers also receive the Jumpstart guide. Patient and family delivery: For the survey-based Patient/Clinician Jumpstart arm, study staff will provide the patient or surrogate with a version of the guide, tailored to be appropriate for patients or their surrogates and following the same timeframe. Jumpstart guides are delivered to the patient or surrogate/family at the hospital. Comparison group: We will compare all 3 arms (survey-based Jumpstart, EHR-based Jumpstart, usual care) in three two-way comparisons. Outcome Assessment: The primary outcome is EHR documentation of a goals-of-care discussion in the 30 days following randomization. All subjects are asked to complete questionnaires at enrollment. The questionnaires may be completed as computer-assisted interviews or on-line depending on the subject's preference. The study staff member enrolling the subjects notifies them of their randomization status at this time. Subjects randomized to the survey-based Patient/Clinician Jumpstart arm complete additional questionnaire items used to create the Jumpstart Guide. The research staff will contact all subjects in person (or by telephone or email) for the evaluation/post-intervention phase. Patient and surrogate/family subjects will complete surveys at 3 days and 30 days after randomization. Surveys can be completed by telephone, mail, or online; we will contact each patient or family/surrogate at each interval using their preferred modality. Subjects may complete the surveys with study staff in person (if the patient is still in the hospital), over the phone, online using REDCap, or on paper and return via mail. Surrogate/Family subjects will complete the same measures at the same intervals as patients with a few exceptions: families will complete the SF-1 for themselves as well as by proxy for patients. Follow up contacts for subjects at 3-day follow-up will be as follows: If the patient is still hospitalized, in-person contact will be attempted (surrogates, who may not be reachable in person, may be contacted by their preferred mode); otherwise, contact will be attempted via phone, email, or mail per the subject's preference. Second and third contact attempts will be made using the subject's preferred mode 2 and 4 days later (7 and 14 days later for mail). Follow-up contacts for subjects at 30-day follow-up will be as follows: initial contact by subject's preferred mode at 4 weeks from enrollment, followed by 4 additional contacts at an interval of every 4-7 days for phone and online, and an interval of every 7-10 for mail. In all cases, only non-respondents will continue to be contacted. For surrogates of patients who have died, after a minimum of 4 weeks following the patient death, an "after death" questionnaire will be sent to the subject using their preferred mode. The after-death questionnaire includes items related to treatment preferences, psychological distress (HADS), and health-related quality of life (SF1). The after-death questionnaire will be sent one time with no additional follow-ups. At the end of study participation, study staff will collect additional information from the patient's electronic health record including intensity of care outcomes Data will be abstracted using automated EHR data collection and "gold standard" manual abstraction using standardized methods for training and quality control. Secondary outcomes also include following patient- and surrogate/family-reported outcomes assessed by survey at 3 days and 30 days after randomization iin include: occurrence and quality of goals-of-care communication in the hospital, goal concordant care, psychological symptoms, and quality of life. SPECIFIC AIM (TRIAL 2): Evaluate the factors affecting the Jumpstart intervention implementation and identify barriers and facilitators to future implementation. Patient and Surrogate/Family Subject Identification and Recruitment: Included in the patient's and surrogate/family member's consent form is a provision informing subjects that they may be contacted at the end of their study involvement to take part in a short, semi-structured interview to evaluate their study participation. Subjects will be sampled purposively to represent the following experiences: 1) participants from both intervention arms (EHR-based and survey-based); 2) participants who participated in the intervention fully as well as those who did not to better understand the reasons for less than full participation. "Full" participation is indicated by having completed all of the study materials. Clinician Subject Identification and Recruitment: Study staff will recruit clinicians who were involved with the study to participate in a short interview after the clinician's study involvement with the enrolled patient has ended. All interview participants will be selected using purposive sampling to ensure a diverse group (e.g., age, race/ethnicity, gender, specialty, year of training) in addition to the characteristics noted above. Interview: Using an interview guide developed specifically for this project, interviewers will assess respondents' experience with the intervention and gather suggestions for ways to improve the intervention's content, delivery, and implementation, including implementation outcomes (e.g., acceptability, fidelity, penetration, maintenance) that will guide future dissemination of the intervention. Assessment: Interviews are audio recorded, transcribed, and analyzed using thematic analysis. To ensure trustworthiness, interviewers perform a "member check" of the results with approximately 12 prior participants selected for diversity of participant type.

Dementia, Chronic Disease, Neoplasm Metastasis, Lung Neoplasms, Pulmonary Disease, Chronic Obstructive, Heart Failure，Congestive, Liver Cirrhosis, Kidney Failure, Chronic, Lung Diseases, Interstitial, Peripheral Vascular Disease, Diabetes With End Organ Injury, Palliative Care, Patient Care, Health Care Quality, Access, and Evaluation, Patient Care, Inpatients, Health Communication, Patient Care Planning, Quality of Life

The Survey-based Patient/Clinician Jumpstart Guide will be developed with two types of data: 1) EHR data; and 2) Survey data. Using automated methods and NLP/ML algorithms, the presence/absence of POLST, advance directives and DPOA documentation will be identified from both inpatient and outpatient notes (e.g., progress notes, specialty consult notes, alerts and care plans) preceding the current hospitalization. The survey data will be completed by patients or their surrogate/family at enrollment and will provide assessments of the following: a) preferences for discussions about goals of care; b) most important barrier and facilitator for having such discussions; and c) current goals of care. These elements are contained within the Jumpstart guides and the information is tailored to each recipient (i.e., patient, surrogate/family, or clinician).

The EHR-based Clinician Jumpstart Guide will be developed by using automated methods and NLP/ML algorithms to both inpatient and outpatient EHR notes (e.g., progress notes, specialty consult notes, alerts and care plans) preceding the current hospitalization. It will summarize the presence/absence of POLST, advance directives and DPOA documentation. It will not include survey-based information.

Patients in this arm receive usual care; neither subjects nor providers will receive either version of the Jumpstart Guide.

The Jumpstart Guide is a communication-priming intervention for clinicians, patients and their families that addresses hospitalized patients' goals of care. The intervention's goal is to prompt clinicians to provide standard of care which includes a discussion with patients or their legal surrogate decision-maker about their goals of care. The Jumpstart takes two forms. One is "EHR-based" and provides information to the clinician about prior advance care planning documentation derived from the EHR. The other is "bi-directional" and patient-specific (Survey-based Patient/Clinician Jumpstart Guide). It provides the same information as the EHR-based Clinician Jumpstart Guide but also includes information about patients' self-reported preferences for communication and care as well as primary barriers and facilitators to this communication. The Jumpstart Guides include tips to improve this communication for clinicians, patients and families.

The Jumpstart Guide is a communication-priming intervention for clinicians that addresses hospitalized patients' goals of care. The intervention's goal is to prompt clinicians to provide standard of care which includes a discussion with patients or their legal surrogate decision-maker about their goals of care. The one-page EHR-based Clinician Jumpstart Guide provides information to the clinician about prior advance care planning documentation derived from the EHR. The EHR-based Clinician Jumpstart Guide includes tips to improve goals-of-care communication.

The primary outcome is the proportion of patients who have a goals-of-care discussion that has been documented in the EHR in the period between randomization and 30 days following randomization. The proportion is the number of patients with GOC documentation over the number of patients in each study arm. Documentation of goals-of-care discussions will be evaluated using our NLP/ML methods.

Secondary outcomes include measures of intensity of care, including utilization metrics: Number of ICU admissions during the patient's (index) hospital stay will be collected from the EHR.

Secondary outcomes include measures of intensity of care, including utilization metrics: Number of days alive and out of the ICU within 30 days from randomization will be collected from the EHR. Number of ICU days from randomization to hospital discharge or death will also be collected from the EHR.

Secondary outcomes include measures of intensity of care, including utilization metrics: Number of days alive and out of the hospital within 30 days from randomization will be collected from the EHR. Number of hospital days from randomization to hospital discharge or death will also be collected from the EHR.

Secondary outcomes include measures of intensity of care, including utilization metrics: Number of hospital readmissions between randomization and 30 days following randomization collected from the EHR.

Secondary outcomes include measures of intensity of care, including utilization metrics: Number of ICU readmissions between randomization and 30 days following randomization will be collected from the EHR.

Costs for intervention vs. control will be reported in US dollars and identified from UW Medicine administrative financial databases. Costs will be reported for total hospital costs and disaggregated costs (direct-variable, direct fixed, indirect costs). Direct-variable costs will include supply and drug costs. Direct-fixed costs will include labor, clinical department administration, and overhead fees. Indirect costs represent services provided by cost centers not directly linked to patient care such as information technology and environmental services. Costs for ED (emergency department) days and ICU days will be similarly assessed.

Subjects will complete a short healthcare utilization survey at the time of the 2nd follow-up questionnaire in which they will self-report the number of visits to emergency departments, hospitals and/or outpatient clinics during the study period.

Subjects will self-report (yes or no) if they had a discussion of goals of care ("the kind of medical care you/your loved one would want") during the index hospitalization. This outcome will be presented as a proportion: the number of subjects reporting a goals-of-care discussion over the number of patients in each study arm.

Quality of goals-of-care communication is assessed with the end-of-life communication scale (QOC_eol) of the Quality of Communication (QOC) survey. The QOC_eol subscale is based on 4 to 7 items, with item scores ranging from 0 (worst) to 10 (best).

Concordance between the care patients want and the care they are receiving will be measured with two questions from the SUPPORT study. The first defines patients' preferences: "If you [patient] had to make a choice at this time, would you prefer a course of treatment focused on extending life as much as possible, even if it means having more pain and discomfort, or would you want a plan of care focused on relieving pain and discomfort as much as possible, even if that means not living as long?" The second question assesses perceptions of current treatment using the same two options. The outcome is a dichotomous variable of whether the preference matches the report of care received.

Symptoms of depression and anxiety assessed with the Hospital Anxiety and Depression Scale (HADS). The HADS is a reliable, valid 14-item, 2-domain (anxiety and depression) tool used to assess symptoms of psychological distress. Seven items evaluate anxiety and seven evaluate depression. Each item is scored on a 4-point scale (ranging from 0-3) with scores for each subscale (anxiety and depression) ranging from 0-21.

The EuroQol 5 Dimension 5 Level (EQ-5D-5L) is a self-report survey that measures quality of life across 5 domains: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression.

We are using the following question to probe patient-reported discussions of life-sustaining treatments with their doctors: "Have you ever thought about what kinds of life-sustaining treatments you would want, or not want, if you got a lot sicker? Yes/no"

Concordance between the care patients want and the care they are receiving will be measured by an investigator-developed question: "Do you think that your current medical care is in line with your goals?"

Qualitative interviews after individual participation has concluded. Interviews will be guided by the RE-AIM and Consolidated Framework for Implementation Research (CFIR) to explore the factors associated with implementation (e.g., reach, maintenance, feasibility, inner and outer settings, individuals, and processes of care.) Individual constructs within these domains were chosen to fit this specific intervention and context.

Eligibility criteria apply to three subject groups: 1) seriously ill adult patients; 2) adult surrogate legal next of kin of the patients; and 3) hospital clinicians. Inclusion Criteria: PATIENTS. Eligible patients will be those who are: 1) equal to or older than 80 years of age; or 2) equal to or older than 55 years of age with one or more chronic conditions used by the Dartmouth Atlas to study end-of-life care: malignant cancer/leukemia, chronic pulmonary disease, coronary artery disease, heart failure, chronic liver disease, chronic renal disease, dementia, diabetes with end-organ damage, and peripheral vascular disease, 3) English-speaking, 4) admitted for minimum of 12 hours/maximum of 96 hours to participating in-patient services at the participating hospitals, and 5) without documentation in the EHR of a goals-of-care discussion during this admission. Patients without decisional capacity (as documented in the EHR or as identified with a brief six-item screening tool) will be represented by a legal surrogate decision maker/legal next of kin (LNOK) in accordance with Washington State Law RCW 7.70.065. SURROGATE/FAMILY. Eligible surrogate/family subjects will be those who are 18 years of age or older, English-speaking, involved in the patient's medical care or decision-making. CLINICIANS (Interview). Eligible clinicians will be those who are 18 years of age or older, English-speaking, employed at a participating hospital, and have been the clinician of record for an enrolled patient in the trial. Exclusion Criteria: Reasons for exclusion for any patient in include: restricted status (prisoners or victims of violence); legal or risk management concerns (as determined by the attending physician or via hospital record designation); unable to complete informed consent procedures; and without a legal surrogate to participate for them. Patients who are non-English speaking (and therefore unable to complete survey materials) are excluded. Reasons for exclusion for any surrogate/family subject include: non-English speaking (and therefore unable to complete study materials), legal or risk management concerns (as determined by the attending physician or via hospital record designation); psychological illness or morbidity; and physical or mental limitations preventing ability to complete questionnaires. Patients under COVID precautions will be excluded.