Inadequate survivorship care after allogeneic hematopoietic cell transplantation : A retrospective chart review

Hematopoietic cell transplant (HCT) survivors are at risk of Background developing long-term complications. Guidelines for survivorship care of HCT recipients were published in 2012; however, the degree to which these guidelines are incorporated into clinical practice is unknown. The purpose of this study was to determine whether providers utilize the 2012 guidelines and analyze whether survivorship-focused providers, provider gender, or provider year of practice influenced adherence to these guidelines. Adult allogeneic HCT recipient’s medical records were reviewed at Methods: the University of Minnesota between 2010 and 2012; only patients who survived without relapse to their 2-year follow-up visit after HCT were included. A semi-quantitative scoring system was developed providing 1 point for each of the 13 organ systems assessed by the 2012 survivorship care guidelines. Data was collected on history, clinical exam, laboratory tests, preventive measures, and counseling. The primary endpoint was the overall score for adherence to the survivorship care guidelines. Wilcoxon rank-sum tests for continuous and Chi-square tests for categorical factors were used to compare the overall score between provider groups (survivorship-focused providers vs others), provider gender, and provider year of practice (≥10 years vs <10 years). Fifteen providers (9 male, 3 survivorship-focused, 7 with <10 years of Results: practice) provided follow-up care to 77 HCT survivors. Survivorship-focused providers had a higher median overall score than other providers (median 10 vs 8, p<0.01).  Female providers had a higher median overall score than male providers (median 9.0 vs 8, p<0.01). There was no difference in median overall score based on provider year of experience (p=0.43). In conclusion, survivorship-focused providers were more likely Conclusions: to achieve long-term screening recommendations. However, even within this group, adherence to the 2012 screening and preventive practice guidelines was incomplete.  Further efforts to automate and standardize the survivorship assessments in HCT survivors are necessary.


Introduction
Patients who undergo allogeneic hematopoietic cell transplantation (HCT) are typically exposed to chemotherapy and radiation therapy as part of their cancer treatment and transplant conditioning regimens. Two-thirds of HCT survivors have at least one chronic medical illness, and one-fifth of survivors have at least one severe health condition 2 . Their life expectancy is estimated to be 30% lower than the general population because of excess risk of death from second malignancy, cancer relapse, infection, chronic graft-versus-host disease (GVHD), and pulmonary and cardiovascular diseases 3 . Follow-up care after HCT is essential to screen for long-term complications and institute preventive therapies. To standardize management of complications after HCT, a collaborative group led by the Center for International Blood and Marrow Transplant Research (CIBMTR) published guidelines for survivorship screening and preventive practices for long-term survivors after HCT 4 . The guidelines provide specific monitoring recommendations categorized into 13 organ systems commonly affected in transplant survivors. Although the guidelines were published in 2012, adherence to these guidelines among transplantation oncologists and other clinical providers has not been evaluated.
To determine whether providers utilize HCT survivorship care guidelines, we conducted a retrospective, single-institution case series of provider adherence to the guidelines at the 2-year follow-up visit after HCT. In addition, we analyzed whether survivorship-focused care, provider gender, or provider years of practice influenced adherence to the recommended guidelines for managing HCT survivors.

Study design and eligibility criteria
A retrospective chart review was performed to identify adult survivors after allogeneic HCT who attended their 2-year posttransplant visit at the University of Minnesota Medical Center between 2012 and 2014. Inclusion criteria included age ≥18 years and receipt of allogeneic HCT from a matched related, matched unrelated, or umbilical cord blood donor between 2010-2012. Exclusion criteria included disease relapse before the 2-year visit or lack of a follow-up visit at our center. Data was collected on history, clinical exam, laboratory tests, preventive measures, counseling, and provider. This study was approved by the Institutional Review Board of University of Minnesota Medical Center (1412M58923).
Scoring provider adherence to survivorship care guidelines For semi-quantitative assessment of adherence to the HCT survivorship care guidelines, a scoring system was developed that provided 1 point for each quality measurement implemented (Table 1). Scores were determined for 13 individual organ systems: immunization status, ocular system, oral system, respiratory system, cardiovascular system, hepatic system, renal system, musculoskeletal system, central nervous system, endocrine system, mucocutaneous system, second cancer screening, and psychological system. Individual system scores were added to determine an overall score for follow-up care. There were 41 questions under the 13 categories. Each individual organ system score was determined by adding one point for each question addressed. The questionnaire used is available as Supplementary File 1. The overall score assessment was determined by adding one point for each organ system if at least one question in the category for that organ system was addressed. Points were only assigned if there was documentation in the medical chart that a question was addressed; missing data were analyzed as not being addressed.

Study endpoints and definitions
The primary endpoint was the overall score for adherence to the HCT survivorship care guidelines. Secondary endpoints involved analyzing the association between guideline adherence and potential modifiers of follow-up care: provider gender, provider year of experience, and self-reported specialization in survivorship care. Providers were transplantation oncologists who cared for the patients during their 2-year follow-up visit. The time allotted on 2-year visit for providers was 30 minutes. A survivorship-focused provider was a self-identified individual conducting education and research on survivorship.

Statistical analysis
The Wilcoxon rank sum test was used to compare the overall score between provider groups (survivorship-focused providers vs others), provider gender (male vs female), and provider year of practice (≥10 years vs <10 years). For the 13 individual system scores, statistical comparisons between factors were completed using Wilcoxon rank-sum test for continuous and Chi-square test for categorical factors. All analyses were performed using SAS 9.4 (Institute Inc., Cary, NC, USA).

Patient and provider characteristics
Of 111 adult patients surviving to 2 years after HCT, 34 patients were excluded due to disease relapse before the 2-year visit or lack of a follow-up visit at our center. 77 patients who completed a 2-year visit at the University of Minnesota Medical Center were included (Table 2). 15 providers delivered care for these patients. Of these, 6 (40%) were female providers, 3 (20%) were survivorship-focused providers, and 7 (47%) were providers with <10 years of experience.
Overall and individual scores for adherence to survivorship care guidelines The median overall score for adherence to survivorship care guidelines was 8.0 (6.0-13.0; Table 3). Notably, providers did not perform assessments in a substantial percentage of patient organ systems, including immune (36.4%), ocular (59.7%), musculoskeletal (62.3%), central nervous (39.0%), endocrine (27.3%), mucocutaneous (27.3%), and psychological (92.2%). In addition,a large majority of providers did not perform second cancer counseling and screening (84.4%). Responses to all 41 questionson the questionnaire are shown in Supplementary Tables S1 through S3. All patients received a perfect score on the renal system assessment, which included blood pressure screening and routine laboratory tests for blood urea nitrogen and creatinine. Laboratory test for hepatic function score was also completed in the entire cohort of patients.

Factors Influencing Adherence to Survivorship Care Guidelines
Survivorship-focused providers and other providers had significant differences in completeness of screening assessment. Survivorship-focused providers had a higher median overall score than other providers (median 10 vs 8; p<0.01; Table 3). Survivorship-focused providers were more complete in assessment of the ocular system, oral system, cardiovascular system, musculoskeletal system, central nervous system, and mucocutaneous system (Table 3 and Supplementary Table S1). In addition, survivorship-focused providers were more likely to perform counseling on second cancer awareness and screening than other providers. However, most survivors had no documented assessment of their psychological and behavior health by either survivorship-focused providers or any other providers (85.7% vs 91.2%, p = 0.52).
Female providers had a higher median overall score than male providers (median 9.0 vs 8, p<0.01; Table 3). Female providers covered more organ systems than male providers with higher median scores in the ocular system, cardiovascular system, endocrine system, and second cancer screening and counseling (Table 3 and Supplementary Table S2).
There was no statistically significant difference in median overall score based on provider year of experience (Table 3; p=0.43); however, providers with <10 years of experience performed better in assessments of the immune system, while providers with ≥10 of experience performed better in assessments of the endocrine system (Table 3 and Supplementary Table S3). published in 2012; however, the degree to which these guidelines are incorporated into clinical practice is unknown. In the present study, we demonstrate that adherence to the 2012 guidelines is sub-optimal and identified areas needing diligent assessment attention in the long-term care of HCT recipients. While all providers followed certain guidelines of survivorship care, such as liver function screening, renal screening, and blood pressure screening, there were several areas identified where extra training in survivorship care would enhance the implementation of screening measures. Our study showed that survivorship-focused providers performed more extensive screening assessments and counseling in 7 organ systems (ocular, oral, cardiovascular, musculoskeletal, central nervous system, mucocutaneous, and second cancer screening) as compared to other providers. Given the discrepancy in these areas, further oncology education in HCT survivorship care for all providers is needed to improve the health of HCT survivors.
Caring for HCT survivors requires understanding their longterm morbidity and probability of high mortality. The challenge for clinicians in caring for HCT survivors is tremendous and a major responsibility of the transplant team. In our case series, there were several important areas where clinicians could improve their assessment of HCT survivors, including immunization assessment and completing immunizations on schedule, assessment of endocrine dysfunction, cardiovascular disease, cognitive function, sexual function, psychosocial and quality of life, inquiry of family functioning, substance abuse, counseling on physical activity, fall prevention, supplementation with calcium/vitamin D, and awareness of second malignancies. All of these areas were noted to be significantly deficient in many patients in our series. Having a provider with a background and interest in survivorship care improved the number of long-term health screenings an HCT recipient received; however, there is still room for improvement as less than a quarter of HCT survivors received a truly comprehensive clinical assessment during their 2-year visit.
Over a decade ago, the Institute of Medicine proposed that all cancer survivors have a survivorship care plan (SCP), which should include a treatment summary and follow-up care plan 5,6 . The American Society of Clinical Oncology and other professional organizations put tremendous effort into developing various survivorship care models, such as the SCP, treatment summaries, and implementation strategies to attain improved quality care for all cancer survivors; however, the implementation of these SCPs has not been satisfactorily accomplished in oncology practices [7][8][9][10] . There are several practical issues: lack of time and staff, lack of coordination and communication among providers, and potentially reimbursement. In addition, evaluation of the benefits in terms of improved outcomes of survivorship care is ongoing 11,12 .
The post-transplant visit is allocated 30 minutes for each patient, with labs studies completed prior to the provider visit. Addressing complex medical issues and potential complications during 30 minutes is challenging for providers. A high-quality posttransplant visit can only be completed with expert time management and thorough preparation and coordination of care among providers. Using electronic methods or automation to ensure the thirteen identified areas are covered during the 2-year visit would ensure that all HCT patients receive the recommended care.
One proposed strategy to overcome the above-mentioned barriers is a risk-stratified shared-care model with delegation of roles and responsibilities between the oncologist and primary care provider. For example, subspecialists such as a dentist, an ophthalmologist, and a psychologist could be closely involved in coordination of long-term follow-up care 13 . It has been shown that the frequency of health promotion and health behavior discussions are suboptimal by clinicians because of the fact that approximately 25% of cancer survivors are not engaged in any such discussions 14 . Cancer survivors are likely to receive appropriate interventions for their comorbid conditions if they routinely follow up with their primary care providers 15 . Educating HCT survivors with ongoing care by their primary    1 6(7.8%) 4(7.1%) 2(9.5%) 4(9.8%) 2(5.6%) 3(9.1%) 3(6.8%) care providers will be an important element in post-transplant care. In general, coordinating the shared care of a cancer survivor between oncology and primary care appears insufficient because of primary care providers' knowledge, attitudes and comfort in caring for cancer survivors 16,17 . With effective and efficient communication, this strategy is likely a feasible option to achieve.
Several transplant centers have adopted the concept of a dedicated clinic of providers that focus on the long-term effects of cancer and previous treatment and deliver education on healthy lifestyle behaviors to reduce complications and lower the risk of additional cancers, while the non-transplant oncologist focuses on cancer surveillance. Our study suggests the potential benefit of having dedicated providers for HCT survivors as the providers' interest, enthusiasm and additional training would likely lead to a comprehensive review of all important clinical assessment in a well-organized fashion.
Several professional oncology organizations have proposed a SCP in a denotable template and web applications with data storage (e.g., The American Society of Clinical Oncologists, Journey Forward and OncoLink) to assist in developing a survivor's own care plan and delivering health education. It is critical to allow the survivors to be actively involved in their SCP. One other potential approach would be to have patients complete their clinical assessment questionnaire prior to their visit; this questionnaire would allow the provider to focus on clinical assessment and counseling rather than data collection. In addition, the electronic medical record could be customized to ensure a comprehensive assessment by providing clinicians with a SCP to-do list.
Our analysis also supports the notion that some categories in assessment (eg, laboratory testing for liver function, renal function and blood pressure measurement) were completed 100% of the time as simply a result of routine practice. Further work is needed to examine how the electronic medical record can more efficiently automate these measures.
Our analysis may be limited in that it is a retrospective chart review at a single transplant center a few years after guidelines were published. It captures one time point in the management of HCT survivors who are likely followed by their primary care clinic in addition to our transplant clinic; important clinical issues during other visits or by their primary care providers may have occurred. Given there is currently no standard for being an expert in survivorship care, providers also self-reported their expertise in survivorship care. Another factor that has to be taken into account in this chart review is the comprehensiveness of the documentation of topics addressed at these visits. For example, mental health screening may have occurred; however, if not documented, the screening was not captured. Nonetheless, this study is the first step to recognize how effectively HCT survivorship care is being provided. Despite these limitations, this study demonstrates that providers need more education in long-term complications of cancer treatment, as well as strategies and tools about how to implement these screening recommendations in a timely, efficient manner given the other demands of their clinical practice.
In conclusion, post-transplantation care by survivorship-focused providers was more likely to achieve the long-term screening recommendations set forth by experts from international transplant professional societies; however, there is room for improvement in adherence to the guidelines even within this group of providers. Our study highlights the critical need of survivorship education for providers as well as better tools to automate this process. Given the complexity of caring for HCT survivors,

Supplementary material
Supplementary File 1: Study questionnaire.
Click here to access the data.
Supplementary Click here to access the data.
Supplementary Table 2: Overall and individual system scores by provider's gender.
Click here to access the data.
Supplementary Table 3: Overall and individual system scores by provider's experience.
Click here to access the data.
new comprehensive and efficient tools to improve adherence to these guidelines are needed to provide optimal survivorship care.

Grant information
The author(s) declared that no grant were involved in supporting this work.
The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Mehrdad Hefazi
Division of Hematology, Mayo Clinic , Rochester, USA This manuscript is aimed at answering an important question in the field of survivorship care after allogeneic HCT. Authors have used a novel and interesting approach to quantify the rate of compliance with recommended guidelines. However, there are several major issues in their methodology: First, authors have used provider gender, provider years of practice, and providers' focus of practice (survivorship-focused or not) as independent variables. However, considering that survivorship care is provided longitudinally and often by different providers over time, categorizing subjects based on these variables is almost impossible. Which category do we put a patient in if they were seen twice by a female provider and three times by a male provider over the course of two years? Unless patients were seen exclusively by the same providers for the entire period, I do not think we can categorize them accurately based on these characteristics. We cannot take only one visit into account either as survivorship care involves more than a single visit.
Second, for overall score assessment, authors have added one point for each organ system if at least one question in the category for that organ was addressed. The questions grouped together in each category are of very different importance though, and they cannot be a substitute for each other. For example, in the Respiratory System, there are four questions, and one of them is "Clinical Exam," which is very likely to be documented in almost any clinicians' note. Taking this alone as an overall assessment of adherence to survivorship guidelines for the respiratory system will be rather misleading as we can have a group of patients with 0% adherence to PFTs, who are marked as 100% compliant with overall respiratory survivorship care just because they all had a clinical exam.
Third, missing data were analyzed as not being addressed, even though this is not an accurate assumption. Some aspects of survivorship care such as "inquiry about family functioning" or "Inquiry about sexual functioning" are inherently less likely to be documented in the records, whereas other components such as renal function or liver function tests almost always get recorded and can be captured retrospectively. Authors should distinguish these from each other, and either report missing data separately or state that assessment of adherence in certain areas were not possible.
At the end, an alternative scoring system (or a different reporting methods) that is based on more objective and specific measures instead of lumping together a group of questions with different clinical significance and rates of documentation would be a more reliable and informative approach. For instance, separately reporting on the rate of compliance with PFTs, measurement of bone densitometry, colonoscopy, or referral to ophthalmology is far more informative and reliable than providing a summative score for how many questions in each organ system were addressed.

Is the work clearly and accurately presented and does it cite the current literature? Yes
Is the study design appropriate and is the work technically sound? No

Are sufficient details of methods and analysis provided to allow replication by others? Partly
If applicable, is the statistical analysis and its interpretation appropriate? Yes