Patient-Reported Outcomes in Hematopoietic Stem Cell Transplant Recipients: Design, Implementation, and Pilot Results

J Clin Pathways. 2024;10(2):17-22. doi:10.25270/jcp.2024.03.02

Hematopoietic stem cell transplantation (HSCT) involves a physically and emotionally challenging treatment with a prolonged recovery period that often leads to an inferior quality of life for patients. Real-time identification of concerns during treatment can address patient challenges, but patient-reported outcome measures (PROMs) describing symptoms, functional status, and sources of distress are often not available to inform medical decisions at the point of care during an office visit. PROs are defined as “any report of the status of a patient’s health that comes directly from the patient, without interpretation of the patient’s response by a clinician. The outcome can be measured in absolute terms (eg, severity of a symptom, sign, or state of a disease) or as a change from a previous measure.”¹ In addition to defining the patient experience, PROMs offer a collaborative opportunity between the patient and care team to improve the quality and safety of the care provided.²

Prior studies using PROMs in patients with cancer largely focus on primary or adjuvant cancer care. PROMs research in patients receiving HSCT is relatively novel and published reports are emerging.^3-8 When used during a patient’s office visit, PROMs provide information that can be used to jointly assess the patient’s evolving health outcomes, decide on next steps in the patient’s treatment plan, and monitor the impact of care on health and well-being.^9-13

The routine incorporation of PROMs from HSCT recipients into clinical care improves both patient-physician communication and patient satisfaction.^14-16 Patients demonstrate improved health when they partner with their treating clinicians to coproduce care plans, based on their preferences and concerns.^17,18 Current research demonstrates that PROMs augment the transplant physicians’ understanding of the experience of the patient receiving HSCT^3,19 since physicians tend to overestimate patients’ quality of life compared with patients’ reports.²⁰ 

Based on our experience using PROMs to improve care for patients, our interdisciplinary team, including members of the Transplant and Cellular Therapy team and health services researchers at The Dartmouth Institute for Health Care Policy and Clinical Practice and the Dartmouth Cancer Center (DCC), planned to measure baseline health-related quality of life (HRQoL) in HSCT recipients at baseline and through 12 months post-transplant period.^9,10,21-27 There were two main objectives of this study: First, we sought to determine the feasibility of routinely capturing HRQoL data in HSCT in the first year following stem cell transplantation. The second objective was to examine the trajectory of HRQoL among people receiving HSCT and to compare HRQoL between people receiving an autologous or an allogeneic HSCT. Since the allogeneic transplant experience tends to be more physically demanding with more toxicities, we expected there would be a difference in PROMs when comparing autologous and allogenic transplant recipients, but results within the literature are mixed.^28-31

Identifying Appropriate PROMs: Patient and Caregiver Forums

Our interdisciplinary team’s primary goal was to identify and evaluate HRQoL in patients receiving HSCT. After an extensive review of the literature, the research team identified the Patient-Reported Outcomes Measurement Information System (PROMIS-29) as the most appropriate tool to assess HRQoL. This tool has been validated in various populations of patients with cancer and is endorsed by the National Institutes of Health, the Center for International Blood and Marrow Transplant Research, and the Blood and Marrow Transplant Clinical Trials Network.^25,32

Prior to starting the distribution of surveys to patients, a series of four focus groups with 5-10 patients and their caregivers assessed the feasibility and acceptability of PROMIS-29, as well as potential time points for patients’ completion of the surveys. Based on recommendations from the focus groups, a pre-visit questionnaire (PVQ) with open-ended questions assessing the patient’s top concerns to address and the National Comprehensive Cancer Network (NCCN) Distress Thermometer (DT) were added to the PROMIS-29 survey. The NCCN DT has been validated in patients with cancer and HSCT recipients.^33-35

Patient Population

Patients 18 years of age or older receiving an autologous or allogeneic HSCT at Dartmouth-Hitchcock Medical Center and the DCC were eligible. After discussion with our local institutional review board (IRB), this work was considered a quality improvement project and IRB approval was not required.

Implementation of the Process

Based on recommendations obtained from patient and caregiver forums, the PVQ, which included the PROMIS-29 survey, the NCCN-DT, and agenda-setting questions, were administered to patients at five time points: 1 month prior to transplant (baseline) and at 1, 3, 6, and 12 months following transplant. The PVQ was distributed in paper format and mailed to patients or handed to patients during their clinic visit. To ensure compliance, patients were phoned 1 week prior to their scheduled appointment and reminded to complete the survey. If the survey was not completed prior to the patient’s clinic visit, a research nurse met with the patient and asked that the patient complete the survey in the waiting room. The completed forms were collected by a research nurse and each patient was assigned a unique patient number. The results were then entered into a database.

The agenda-setting question was open ended: “What is your number-one concern that you would like to discuss at today’s visit?” Agenda-setting responses were available to the provider at the time of the visit and were used to engage the patient in conversation and explore their priorities for the visit; however, due to their qualitative nature, responses were not included within our PROMs dataset. The PROMIS-29 and the NCCN DT were not available at the time of the visit but were collected and scored after the visit.

Collection of Patient Data

Each patient was provided a unique patient number and patient-specific results were entered into a spreadsheet-based software database. Data were considered evaluable if a patient completed one or more posttransplant surveys, with the pretransplant survey required for comparison.

Statistical Analyses

Primary analyses used outcomes scores at each assessment, with a linear mixed effects longitudinal regression including a random individual effect to account for correlation between repeated measurements within individuals. All models were adjusted for age, gender, type of stem cell transplant, and baseline score. We accounted for the time difference in the linear mixed models to get observed margin estimates at “time = 0.” In addition, we examined type of stem cell transplant. Subgroup analyses and outcomes were stratified by type of stem cell transplant through an interaction term between “type of stem cell transplant” and “assessment time period.”  Analyses were performed using SAS software, version 9.4 (SAS Institute Inc, Cary, North Carolina). Statistical significance was defined as P < .05 based on a two-sided hypothesis test with no adjustments made for multiple comparisons.

Patient Characteristics

From October 2019 to October 2020, 65 patients participated in the evaluation (Table 1).  Median patient age was 62 years (range: 19-73 years). Fifty-two percent were female and 48% were male. Forty-two percent received an allogeneic HSCT and 58% received an autologous HSCT. The most common patient diagnoses included multiple myeloma, acute myeloid leukemia, and non-Hodgkin lymphoma.

Survey Completion Rates

Of 67 consecutive patients completing a baseline evaluation, 65 (97%) were eligible for participation. Survey completion rates were 95% at 1 month posttransplant (n = 62 patients: autologous = 36, allogeneic = 26), 88% at 3 months (n = 57 patients: autologous = 32, allogeneic = 25), 74% at 6 months (n = 48 patients: autologous = 29, allogeneic = 19), and 65% at 1 year (n = 42 patients: autologous = 28, allogeneic = 14). Overall, 43% (n = 28) completed all five surveys, 42% (n = 27) completed four surveys, 9% (n = 6) completed three surveys, and 6% (n = 4) completed two surveys.

Identification and Comparison of HRQoL Following Autologous or Allogeneic Transplant

When HRQoL was examined among all HSCT recipients and outcomes were adjusted for age, gender, type of transplant, and baseline score, three signs and symptoms varied significantly over time, including fatigue (P < .001), physical function (P < .001), and social function (P < .001) (Figure 1; Supplementary Table 1). When compared to baseline, transplant patients’ level of fatigue increased at 1 month posttransplant for all patients and returned to baseline at 3 months following transplant. Both physical function and social function followed a similar path, with a decrease in function at 1 month posttransplant followed by a slow and gradual increase back to baseline at 12 months following transplant.

Figure 2 shows the changes in PROMs from baseline (pretransplant) until 1 year posttransplant between patients receiving an allogeneic or autologous transplant (Supplementary Table 2).  Fatigue and difficulties with physical function were more prominent within the allogeneic transplant recipients compared to autologous patients. When comparing the two cohorts, fatigue levels rose in both groups, but the level of fatigue was more pronounced in the allogeneic recipients (P = .017). An initial decline in physical function occurred in all recipients, but patients receiving an autologous transplant demonstrated a higher overall physical function than those receiving an allogeneic transplant (P = .021). A similar pattern was observed with a decline in social function in both groups, but the difference between the two groups was not statistically significant (P = .16). There were no differences between the two cohorts in the other PROMIS-29 domains or in the overall NCCN DT score (Supplementary Table 1). Within the NCCN DT problem list categories, there was a notable decline in patients receiving HSCT identifying emotional problems, from 57% at baseline to 22% at 12 months posttransplant (P < .001) (Supplementary Table 3).

The open-ended agenda-setting questions were used to guide discussion but were not routinely logged for this study. Patient and physician feedback regarding these questions was strongly favorable.

This study confirms the acceptability and feasibility of capturing PROMs longitudinally in patients receiving HSCT and supports the value of collecting PROMs in patients who have received transplants. Initially, patient acceptability of the project was high, with 97% of consecutive patients engaging in the study. Survey completion rates remained high over the year following transplant, ranging from 65% to 95%.

As expected, the results demonstrate statistically significant changes in the level of fatigue, physical functioning, and social functioning over the course of a year following transplant. We identified statistically significant differences in symptoms when comparing patients receiving autologous and allogeneic stem cell transplant, with allogeneic stem cell transplant recipients experiencing more severe fatigue and prolonged physical dysfunction when compared to patients receiving an autologous transplant.

This study provides an exploratory view of the value of agenda-setting questions for transplant patients and their providers. Based on patient and physician feedback over the time period of this study, we have expanded the open-ended questions and now include three questions: “What are the top things you’d like to discuss in your upcoming visit with the transplant team?”, “What has been going well for you in the last week?”,  and “What goals are important to you that you want the team to be aware of?” We have since conducted a thematic assessment of these responses in a subsequent patient population,¹⁸ and studies are ongoing to understand the value of agenda-setting questions and the impact of disease-related anxiety.

Due to the arduous nature of transplant and the challenging posttransplant clinical course, physical symptoms—including fatigue, sleep disturbances, and pain—are the major contributors to impairments in HRQoL following transplant.¹⁹ Fatigue is generally the most common complaint among transplant recipients.³⁶ Due to the prolonged hospital stay and subsequent course associated with allogeneic stem cell transplantation, it is more common for allogeneic patients to experience prolonged fatigue and physical dysfunction compared with autologous recipients, as our data demonstrate. Our results show that patients who received transplants report the lowest levels of physical functioning immediately following transplant, with a gradual improvement to pretransplantation levels within 3 months after transplantation. These results have been corroborated by others.^{6,19,28,37,38}

The PVQ was designed to engage the patient in shared communication and included open-ended agenda-setting questions that allow the patient to identify primary concerns they want to address during the visit. Research supports that an agenda-setting opening question enhances the physician-patient connection within the clinical encounter.^39,40 Incorporation of the PVQ surveys (including the PROMIS-29, NCCN DT, and agenda-setting questions) into the transplant clinic is now part of our routine care, demonstrating ease of use along with both clinician and patient engagement. The routine incorporation of PROMs into the clinical care of patients receiving HSCT can provide several benefits, including the potential for improved patient-physician communication, enhanced accuracy of providers’ understanding of the experience of patients receiving HSCT, and an amplified role for patient-clinician shared decision-making.^19,20,41 In addition, PROMs facilitate the prompt recognition and treatment of patient symptoms.⁴² These benefits promote a stronger alliance between patients and their providers and an overall improvement in patient satisfaction.^14-16Finally, it is important to note that the use of PROMS can provide benefits on many levels. For example, Sawatzky et al described the potential benefits of PROMs that focus on patient-specific responses and patient care, known as “micro” aims.⁴³ The use of PROMs to address the institutional delivery of care is considered a “macro” aim. Our initial focus was to design and implement a standardized process that incorporated PROMs into patients’ clinic visits (ie, a macro aim). Our long-term goal was to provide patient-specific PROMs results at the time of the patient’s visit, to assist in patient care and decision making (ie, a micro aim).

We acknowledge limitations of this study. First, as a single-center study in a rural National Cancer Institute–designated comprehensive cancer center, the results may not be applicable to other institutions. Second, due to published reports, national committee recommendations, and patient/caregiver recommendations, we selected the PROMIS-29 and NCCN DT as part of our PROMs surveys. Other surveys or measures may provide additional beneficial or complementary information. For example, tools that capture financial toxicity or symptoms related to graft-vs-host disease may add significant value to our current PROMs surveys. Third, the paper format of the PVQ limited our ability to intervene in real time. The agenda-setting questions were available to the provider at the time of the visit, but the DT and scored PROMIS-29 were not. All PROMs were reviewed by the treating physician and the medical team in aggregate and retrospectively. This process demonstrated the limitations of the paper format and the urgent need for real-time PROM results that could be reviewed with the patient, as was done with the agenda-setting questions. Based on knowledge gained from this pilot study, the PVQ is now completely electronic, which allows the care team to review the results with the patient during their clinic visit and track changes over time by viewing current and past responses within the electronic medical record (EMR). Fourth, as an exploratory study, we did not mandate or track interventions for individual patients based on PRO values. Future studies are underway that allow us to use PROMs to activate clinical pathways (eg, providing referrals based on food insecurity, transportation barriers, or PROMIS-29 function scores) and to examine the relationship between PROMs and outcomes.

Our data demonstrate that PROMs can reliably monitor areas that impact quality of life of HSCT recipients over time. We continue to monitor PROMs in our patients and we have tailored interventions to improve symptom management in these patients, which may improve patients’ quality of life and long-term outcomes. Based on these results, we have begun to pilot the use of an electronic PROMs survey collected prior to office visits. These results are incorporated within the patient’s EMR. A patient-specific dashboard displays the patient’s agenda-setting responses and PROMs alongside their clinical data, including laboratory results, vital signs, and medications. Other studies have shown that such a dashboard facilitates discussion about the relationship between laboratory results, medications, and PROMs and improves patient understanding of their health status over time.^9,10,43-46 Patients’ responses and primary concerns will also identify areas requiring additional support. We believe that the routine collection of PROMs in patients receiving HSCT and the review at each clinic visit will serve as a springboard for improved discussion and collaboration between patients and providers. Additionally, we are examining ways to utilize PROMs to improve patients’ financial concerns, quality of life, and long-term outcomes of patients receiving HSCT.

This article has supplementary material, which can be accessed here: Supplementary Table 1, Supplementary Table 2, Supplementary Table 3.

Authors: Christi Ann Hayes MD¹; Aricca D. Van Citters MS²; Wenyan Zhao PhD²; Kate L. Caldon¹; Charlotte M. Coughenour¹; Tor D. Tosteson PhD²; Anna N. A. Tosteson PhD²; Kenneth R. Meehan MD¹

Affiliations: ¹Transplant and Cellular Therapy Program, Section of Hematology, Dartmouth-Hitchcock Medical Center and the Dartmouth Cancer Center; ²The Dartmouth Institute for Health Policy & Clinical Practice; Geisel School of Medicine at Dartmouth, Level 5, Williamson Building, One Medical Center Drive, Lebanon, NH 03756

Funders: This study was sponsored by Dartmouth Cancer Center.

Address correspondence to: 

Kenneth R. Meehan, MD

Transplant and Cellular Therapy Program

Dartmouth-Hitchcock Medical Center and the Dartmouth Cancer Center

One Medical Center Drive

Lebanon, NH 03756

Email: kenneth.meehan@hitchcock.org

Disclosures: The authors reported no financial or other conflicts of interest.

Preliminary results of this research were reported in poster format at the Northern New England Clinical Oncology Society meeting (an ASCO state-affiliate) in November 2022 and at the ASH Annual Meeting & Exposition (December 2022).