A. BACKGROUND AND SIGNIFICANCE

In advanced cancer patients, cancer and its treatment cause multiple physical and psychological symptoms (1). These devastating symptoms are major aspects of morbidity in cancer care and impact overall survival. In patients receiving  palliative  treatment, as the intensity and severity of symptoms increase so does their impact on QOL. In addition to the increased symptom burden and functional decline, patients frequently have spiritual and financial concerns. Their family members may experience physical and emotional distress. At the same time, patients and their families need to discuss the goals of care and to participate in advance care planning (1). These issues are even more urgent in medically underserved regions.

With fewer than 800 doctors nationwide in Zambia, expert consensus is that the country’s medical system remains ill-equipped to take care of the needs of cancer patients who usually are diagnosed late when they are symptomatic and are at advanced stage(2-5). Similar challenges exist in underserved areas of Kenya, Nigeria, Ghana, South Africa, and India (2-5). This system is significantly under budgeted and many needs on health assistance remain uncovered (7). Among the unmet needs include limited access for palliative care. One of the main reasons is the lack of health professional education, training and support to health care providers taking care of this vulnerable patients.

Another special challenge is the size of geographical area of the country and the distance between small cities and the big centers where, usually, there is good quality care. This issue creates an additional barrier to the development of any health policy, including Palliative Care. However, training select health care providers has been challenging due a high turnover of trained health care providers, and problems maintaining and measuring the quality and consistency among those who receive training. There is a great need to train the physicians, nurses, volunteers helping the patients with life limiting illnesses with skills required to provision of quality palliative care. Additionally, there is need to evaluate the impact of the training on improvement of patient outcomes.To assist with these challenges, we are proposing a collaborative project using Project ECHO between the Cancer Diseases Hospital, Lusaka, Zambia, Aga Khan University, Nairobi Kenya, Korle Bu Teaching Hospital, Accra, Ghana, University College Hospital, Ibadan, Nigeria, Tata Memorial Hospital in India, The African Cancer Institute, Stellenbosch University (ACI) South Africa, and Tygerberg Hospital, Capetown, South Africa.

Project ECHO (Extension for Community Healthcare Outcomes) is a collaborative model of medical education and care management that empowers clinicians everywhere to provide better care to more people, right where they live.

The goal of Project ECHO is to improve healthcare workforce capacity and increase access to specialty care for the world’s rural and underserved populations. A low-cost, high-impact intervention, Project ECHO links expert multidisciplinary specialist teams with frontline community healthcare providers via ongoing videoconference-enabled sessions. In these teleECHO clinics, specialists share their expertise and community providers share their experience with individual patients via case-based learning and telementoring. Overall knowledge is enhanced as cutting-edge research, treatments, and best practices from academic and research centers are tested and refined through ongoing discussion and application within community- and culturally-specific contexts on the ground. The ECHO model builds and strengthens communities of practice through latitudinal learning and the free exchange of knowledge: all participants teach, and all participants learn from one another. Community providers develop the capacity to care for patients with complex conditions where they live.

Our long-term objective is to improve access to quality Supportive/Palliative Care (SPC) services for all patients with life-limiting illness. In the current study, we will examine the effects of a Telementoring intervention based on the Project ECHO (Extension for Community Healthcare Outcomes) platform for teaching Palliative Care to primary care clinicians in underserved areas of I,Zambia, Kenya, Ghana, Nigeria, India and South Africa,  and explore its effects on patient outcomes. The data will help to conduct a larger Randomized control trial using NIH funding to determine the effectiveness on patient outcomes.

SIGNIFICANCE

Patients with advanced cancer face severe symptoms stemming from the disease and its treatment, including physical symptoms, like pain, and psychosocial symptoms including depression and anxiety. Palliative Care (PC) providers work provide an interdisciplinary and holistic approach to relieve symptoms and improve quality of life. However, throughout the world many regions lack access to adequate Palliative Care. We have identified an opportunity to improve Palliative Care in underserved regions of Africa and India through our Sister Institution Network partnerships. Specifically, the PC team at MD Anderson Cancer Center in Houston will collaborate with Tata Memorial Hospital in India, the Cancer Diseases Hospital, Lusaka, Zambia, Aga Khan University, Nairobi Kenya, Korle Bu Teaching Hospital, Accra, Ghana, University College Hospital, Ibadan, Nigeria, The African Cancer Institute, Stellenbosch University (ACI) South Africa, and Tygerberg Hospital, Capetown, South Africa using a Telementoring approach to train generalist providers in state-of-the-art PC techniques and methodologies via the Project ECHO model. This project will be highly significant because it will leverage the power of the Sister Institution Network to address a critical need throughout the developing world: provision of quality Palliative Care to the sickest patients.

PRELIMINARY STUDIES

Our team at MD Anderson in Houston, guided by our partner PC providers in Sub Saharan Africa, recently completed a successful pilot Palliative Care ECHO clinic, ECHO Palliative Care Africa (ECHO-PACA) that included a standardized curriculum based on PC needs in the region. Participants were recruited and joined our monthly telementoring sessions consisting of case presentations, discussions, and didactic lectures, beginning in July 2016, and included personnel from 14 teaching hospitals and clinics in Ghana, Kenya, Nigeria, South Africa, and Zambia. We surveyed participants at the beginning, mid-point and end of the 16 month program to evaluate changes in attitudes, and self-perceived efficacy in pain assessment and management, identification of signs/symptoms of imminent death, and identifying and addressing challenging communication issues related to end of life. Median participation per session was 30, the highest of any ECHO clinic at MD Anderson. Among the 33 of 40 initial participants (83%) who completed the survey, there was significant improvement in appropriate use of non-opioid analgesics for persistent pain (p = .03), titrating opioids to optimize pain control (p = .03), ability to identify signs/symptoms of imminent death (p = .05), and ability to identify and address challenging communication issues related to end of life (p = .02). Our results suggested that ECHO was a feasible, cost effective, pragmatic approach to disseminate PC knowledge without the need for participant travel, and enabled us to refine our didactic lecture series to best serve participants in underserved regions.

Relevance to Sister Institution Network: The leadership teams at our Sister Institutions will serve as a Co-PIs of the project and co-mentors of the primary care clinicians. Importantly, these co-mentoring teams will provide expertise on the specific needs of their regions and provide support to local providers and their patients. Despite limited access to palliative care for advanced cancer patients in the Low and Middle Income Countries (LMICs) of these Sister Institutions, all Co-PIs are actively involved in research and model palliative care education in their respective country. Through constant engagement, this project will enrich the research collaborations and educational partnerships between MD Anderson and our Sister Institution Network colleagues in LMICs.

Recent evidence strongly suggests early referral to PC can improve the quality of cancer care and possibly even survival. However, the majority of patients with cancer in underserved regions are diagnosed late, and suffer from more severe physical and psychological symptoms, negatively affecting their quality of life and ability to receive cancer treatment. Training providers in PC, from community health workers to cancer specialists is needed to fill this gap, and our proposed ECHO telementoring intervention is specifically designed to enable non-specialist providers in the community to improve their skills and self-efficacy. Importantly, ECHO training will likely have a lasting impact and we anticipate that improved patient outcomes will be sustained long after the project is completed. Ultimately, this model educational intervention can be easily replicated to improve access to quality palliative care by both PC specialists and primary care providers throughout the Network and beyond, thereby reducing the profound suffering of advanced cancer worldwide.

 B. OBJECTIVES

Primary Objective:

To evaluate the effects of Project ECHO on patient-reported quality of life [QOL] (as assessed by the change in Functional Assessment of Cancer Illness Therapy FACT-G) in seriously ill patients of providers receiving ECHO Model-Based comprehensive educational and telementoring intervention (ECHO-Palliative Care, or ECHO-PCT).

We expect that patients treated by providers trained and mentored in palliative care via ECHO-PCT for 12 months will have improved quality of life.

Secondary/ Exploratory Objectives:

1. To examine the effects of ECHO-PCT on patients’ symptom distress as assessed by the Edmonton Symptom Assessment Scale (ESAS), Symptom Distress Scores (SDS), as well as patient and caregiver experience and satisfaction as assessed by the FAMCARE–P-16-patient and FAMCARE-caregiver questionnaires.

We expect that patients treated by providers trained in Palliative Care via  ECHO-PCT will have improved ESAS SDS and FAMCARE–P-16 scores. Similarly, we expect that the caregivers of patients whose providers received ECHO-PCT will also report an improved experience.

2. To explore the effects of ECHO-PCT on providers’ efficacy in, knowledge of, and competence to deliver Palliative Care to Patients with serious life limiting illness, as well as their satisfaction with providing Palliative Care services as assessed by Palliative Care Participant Initial Survey (Appendix Q), Palliative Care Participant Follow Up Survey(Appendix R), and Feedback Survey (Appendix U); (ii) increase patient access to quality palliative care in these regions (morphine consumption); and (iii) increase in the number of palliative care procedures performed by the ECHO participants (subcutaneous hydration, opioid rotation). iv) increased patient access to PC measured by the number of patient visits (new and follow-ups); v) provider turnover; vi) adherence, measured by the number of ECHO sessions held, the number of ECHO participants at each session, and the number of cases reviewed; and vi) comfort with the ECHO platform and ability to learn/engage via distance technology, assessed by the Palliative Care Participant Initial Survey (Appendix Q) and Palliative Care Participant Follow Up Survey (Appendix R).

We expect that providers trained in Palliative Care via ECHO-PCT will report improved efficacy, knowledge, competency, and satisfaction.

DESIGN AND PROCEDURE

All ECHO sessions will be provided via the Zoom platform, a secure, web-based videoconferencing system that is approved by MD Anderson and proven to be functional in low-resource settings. Under the Project ECHO licensing agreement, Zoom software is provided free of charge to all ECHO participants, and will work with any computer, smartphone, or tablet with a camera and internet access. Project ECHO utilizes a “hub and spokes” model. The Palliative ECHO telementoring program will consist of one-hour, twice monthly teleECHO clinics with ECHO participants (spokes) from the regions served by the partnering Sister Institutions. The interdisciplinary PC team at MD Anderson Cancer Center will act as the Hub, leading and coordinating the ECHO clinic, and collecting and analyzing outcomes and benchmarks. The Hub team includes PC physicians, nurses, pharmacists, and other PC team members. Experienced PC clinicians from the Sister Institutions themselves will act as expert co-mentors for their respective regional areas. At each session, ECHO participants will take turns presenting de-identified patient cases, including pertinent patient history, exam findings, and workup, for approximately 45 minutes during each session. In addition to discussion among ECHO participants, guidance and feedback regarding diagnosis and management will be provided by the specialists at MD Anderson and the co-mentors from the Sister Institutions, who will then present a 15-minute didactic lecture. The didactic curriculum, tailored to the needs of primary care clinicians in rural and underserved areas, conveys the latest evidence-based techniques in PC. Throughout the 12 month program, the ECHO participant primary care clinicians will have the ongoing support and expertise of the Hub and Expert Co-Mentors in the ECHO sessions and through direct communication by e-mail and telephone. We will continuously survey the participating clinicians to address their needs and knowledge gaps and improve ECHO-PC. In addition to the ECHO clinic, in-country workshops will be held two times per year. Members of the MD Anderson Hub and the co-mentors at the Sister Institutions will travel to the primary care centers of the Sister Institution Network to perform hands-on training and workshops. All ECHO participants will also receive 5 core lectures through MD Anderson’s online Professional Oncology Education (POE) program, an internet-based professional education curriculum that ECHO participants can access at no cost through the MD Anderson POE website (www.mdanderson.org/poe). The core lecture topics will include: 1) Assessment and Management of Pain and Symptoms; 2) Interventions in Palliative Care; 3) Assessment and Management of Delirium; 4) Hydration and Nutrition, including Cachexia; and 5) End of Life Care, including Management of Symptoms in the Last Week of Life.

ECHO Participant Outcomes: Outcomes that will be used as benchmarks for the success of the program, and will be used to adjust the intervention and maximize the impact of the program, include 1) an increase in reported self-efficacy in assessment and management of pain and related symptoms, communication, and ability to perform palliative care procedures relevant to end of life in advanced cancer, measured by a custom survey; 2) increase in the number of PC procedures performed (subcutaneous hydration, opioid rotation) by ECHO participants; 3) morphine consumption by patients, a surrogate measure of access to PC; 4) increased patient access to PC measured by the number of patient visits (new and follow-ups); 5) provider turnover; 6) adherence, measured by the number of ECHO sessions held, the number of ECHO participants at each session, and the number of cases reviewed; and 7) comfort with the ECHO platform and ability to learn/engage via distance technology, assessed by the Palliative Care Participant Initial Survey (Appendix Q) and Palliative Care Participant Follow Up Survey (Appendix R). We will also assess the satisfaction and feedback following the twice a month ECHO clinic during the intervention period (Appendix U).

Eligibility Criteria

Inclusion Criteria

Patient:

1. Scheduled to receive treatment at participating primary care practice for advanced cancer

2. 18 years or older

3.  Willing to be accessible to the research team for follow-up by telephone or in person

4. Prognosis of at least 1 year as ascertained by the treating physician

5. Able to read and understand English as assessed by the treating physician.

Caregiver:

1. Identified or self-identified as primary caregiver of the patient

2. Willing to be accessible to the research team for follow-up by telephone or in person

3. 18 years or older

4.  Able to read and understand English as assessed by the treating physician.

ECHO Participant:

1. Participant must be a health care provider interested to participate in the study for 2 years and understand the protocol and provide consent.

2. Able to read and understand English as assessed by PI

Exclusion Criteria

1. Unable to understand and sign consent.

Patient Recruitment: Our research staff will access the patient schedule of all participating primary care practices at all study sites each weekday (excluding holidays) and approach all potentially eligible patients. If patients and caregivers meet our inclusion and exclusion criteria as described above, we will obtain informed consent. Each primary care practice sees approximately 25-35 new patients with advanced cancer per month, and we anticipate 12 primary care providers (2 per center) will be ECHO participants in our program. Our highly experienced research staff will diligently follow up with all patients to ensure completion of all study surveys and instruments. Study assessments occur only four times per year for up to 2 years, collected by telephone at the patients’ convenience; we anticipate that data collection will constitute minimal burden to patients and their caregivers. The following specific outcomes will be assessed quarterly for up to 2 years.(Table 1)

The 2 provider centres/organizations for Tata Memorial Hospital in India are

1.      Department of Palliative Medicine in HBCH and MPMMMC, Varanasi

2.      Palcare India, Mumbai

The palliative care physicians from these centres will be the ECHO participants.

HBCH and MPMMMCC will be a provider centre site for this study.

Patient Outcomes: At each time point, we will measure 25 patients per site. The Edmonton Symptom Assessment Scale Distress Score (ESAS-SDS) will be used to measure symptom distress. The ESAS is a validated questionnaire measuring 10 symptoms (pain, fatigue, nausea, depression, anxiety, drowsiness, shortness of breath, appetite, sleep, and well-being). It has good reliability and can be completed within minutes. The ESAS-SDS will be calculated by the sum of the symptom scores of measures ESAS Pain, ESAS Fatigue, ESAS nausea, ESAS Depression, ESAS anxiety, ESAS drowsiness, ESAS Dyspnea, ESAS Appetite, ESAS Well-being. Health related quality of life (HRQOL) will be assessed using the Functional Assessment of Cancer Therapy-General (FACT-G) questionnaire, a validated instrument examining HRQOL over the past 7 days. It consists of four domains and each question is assessed using a Likert scale ranging from 0 to 4. The FACT-G total score is the sum of the domain scores. Lower scores are indicative of increasing symptom burden and lower QOL, and the minimal clinically important difference (MCID) is 4 to 7 points in cancer patients. This questionnaire was chosen over other similar assessments because it has been validated extensively, is specific for life limiting illnesses, has been used in other PC studies, and has an established MCID.

Caregiver Outcomes: We will measure 25 caregivers per site. Experience with care will be assessed using the FAMCARE, which has been validated in caregivers of patients with advanced cancer.

Assessments:

The following tools are validated for all patients who can read and write at a

Demographic and Clinical Characteristics (Appendix S)

Demographic variable include age, gender, country, region, cancer type, education, Zubrod performance status score, type of work, marital status, and primary care diagnosis, metastatic site status, initial anti-cancer therapy (platinum-based, single-agent chemotherapy, oral epidermal growth factor-tyrosine kinase inhibitor, radiation, radiation and chemotherapy, no chemotherapy), and type of initial chemotherapy regimen (standard vs. clinical trial) will be recorded.

After informed consent is obtained by a research staff member, the patient and caregiver will complete the study assessments.

Edmonton Symptom Assessment Scale (ESAS) (Appendix A) is a widely used and validated scale for symptom assessment in seriously ill people. Patients rate the intensity of 10 symptoms on a 0-10 rating scale ranging from 0 (no symptom) to 10 (worst possible symptom). Symptoms include pain, fatigue, nausea, depression, anxiety, drowsiness, appetite, dyspnea, sleep disturbance, and overall wellbeing.

Health Related Quality of Life:

The FACT-G (Appendix B) is a 27-item compilation of general questions divided into four primary domains: Physical Well-Being, Social/Family Well-Being, Emotional Well-Being, and Functional Well-Being. It is appropriate for use in patients with any form of cancer, and extensions of it have been used and validated in other chronic illness conditions as well.

Patient and Caregiver Experience

FAMCARE P16 (Appendix C) is a self-report scale assessing patient experience and satisfaction with outpatient palliative oncology care, which is composed of 16 items rated from 1 (very dissatisfied) to 5 (very satisfied). The items are not specific for a particular tumor type or symptom, but are broadly relevant for outpatients with advanced cancer; the summed items produce a single satisfaction score. A preliminary analysis indicated that the measure had good psychometric properties when used with advanced cancer patients in an outpatient palliative care clinic.

FAMCARE –cg (Appendix D) FAMCARE scale, which measures the degree of experience and satisfaction of the caregivers with health care in terms of information provided, availability of care, physical individual with advanced cancer care, and psychological care for patients.

Distress Plan: During the assessments or during the study period if the patient or caregiver express distress or high ESAS distress total score of 55, the site PI will be immediately informed and based on the assessment appropriate care or referral will be made.

Data Collection: All project data will be collected using REDCap (Research Electronic Data Capture) electronic data capture tools hosted at MD Anderson (http://www.project-redcap.org). REDCap is a secure, compliant, web-based application with controlled access designed to support data capture. All protected health information (PHI) will be removed when data are exported from REDCap for analysis.

EXPECTED OUTCOMES: Based on preliminary data from our ECHO-PACA pilot project, we anticipate that the ECHO participants in our ECHO clinic will find this educational opportunity highly beneficial, resulting in a successful training and education experience.

*ECHO telementoring consists of one-hour, twice monthly teleECHO clinics with ECHO participants. Satisfaction and Feedback (Appendix U) will be completed after each teleECHO clinic.

DATA ANALYSIS PLAN

Primary Objective: Descriptive statistics such as mean, median, standard deviation (SD), range, frequency and confidence interval (CI), as well as graphical presentations such as boxplot and histogram will be used to assess patients’ demographics, clinical characteristics, and the pattern of patient reported quality of life measured by FACT-G before the start of intervention and after 12 months of ECHO-PCT intervention, where appropriate. To compare the change in QoL measurements between different patient groups, such as gender and primary cancer diagnosis, we will use t-test and ANOVA. If underlying distributional assumptions are not met non-parametric tests (e.g. Mann-Whitney and Kruskal-Wallis) and appropriate data transformation (e.g., log) will be employed. Pearson or Spearman correlation coefficient will be applied to assess the correlation between two continuous variables (e.g., change in FACT-G score and change in morphine dose). We will also use generalized linear mixed model to assess the QoL measures using the longitudinal data adjusted for relevant demographic and clinical covariates (e.g., sex, primary cancer diagnosis, etc.), where the within-patient (repeated measures across time points during and after intervention) will be accounted for. Multiple comparisons will be adjusted for where appropriate. Other statistical methods, when appropriate, may be utilized.

Missing Data: If missing data become an issue, we will assess the mixed model assumption by assessing the predictors of missingness. We will examine whether patients who drop out of the study differ from those who do not and adjust for covariates that are found to be related to missingness. Additional techniques such as pattern mixture models will be explored if there is a strong belief that the missingness is nonignorable. Sensitivity analyses under various missing data mechanisms will be conducted to verify if assumptions are properly made.

Sample Size Justification: Given 25 patients per center for 6 centers totaling 150 patients, assuming a 20% drop-out rate and assuming the intracluster-correlation coefficient (ICC) equals 0.01 (0.02), the effective sample size equals 97 (81) (effective N equals sample size adjusted for presence of clustering). Therefore, we can estimate the mean change in FACT-G score with a 2-sided 95% confidence interval half-width of 0.199 (0.218) units of the standard deviation. The sample size justification is calculated using nQuery4.0 software.

Secondary Objective 1: The objective is to examine the effects of ECHO-PCT on patients’ symptom distress as assessed by the ESAS, SDS, as well as patient and caregiver experience and satisfaction as assessed by the FAMCARE–P-16-patient and FAMCARE-caregiver questionnaires. Descriptive statistics such as mean, median, standard deviation (sd), range, frequency and confidence interval (CI), as well as graphical presentations such as boxplot and histogram will be used where appropriate. To assess the change in the aforementioned endpoints generalized linear mixed models adjusted for relevant demographic and clinical covariates (e.g., sex, primary cancer diagnosis, etc.), where the within-patient (repeated measures across time points during and after intervention) will be accounted for, will be employed. Multiple comparisons will be adjusted for where appropriate. Pearson or Spearman correlation coefficient will be applied to assess the correlation between two continuous variables (e.g., change in FAMCARE–P-16-patient and FAMCARE-caregiver scores). Other statistical methods, when appropriate, may be utilized.

Sample Size Justification: Given 25 patients per center for 6 centers totaling 150 patients, assuming a 20% drop-out rate and assuming the intracluster-correlation coefficient (ICC) equals 0.01 (0.02), the effective sample size equals 97 (81) (effective N equals sample size adjusted for presence of clustering). Therefore, we can estimate the mean change in ESAS SDS and FAMCARE–P-16 and FAMCARE-caregiver scores with a 2-sided 95% confidence interval half-width of 0.199 (0.218) units of the standard deviation. The sample size justification is calculated using nQuery4.0 software.

For Tata Memorial Hospital in India, 25 patients and 25 caregivers (for the 25 patients recruited) in all will be recruited for the project by the 2 provider centres – Palliative Care department in HBCH Varanasi and Palcare India, Mumbai.

HBCH and MPMMMCC will be a provider centre site for this study.

Secondary Objective 2: The objective is to determine the effect of ECHO-PCT on providers’ (2 per site) efficacy in knowledge of and competence to deliver Palliative Care to Patients. For this objective the data will be summarized by descriptive statistics at baseline, year 1 and 2 of ECHO-PCT by completing the Palliative Care Participant Initial Survey (Appendix Q) and Palliative Care Participant Follow Up Survey (Appendix R).

DATA CONFIDENTIALITY PROCEDURES

All research staff has completed training concerning confidentiality of protected health information.

Collection of identifiers: Participating sites will collect and securely store identifiers (including patient names and telephone numbers). E-mail addresses will be used to send the invitation letter and REDCap survey link for ECHO participants who wish to participate in the ECHO clinic. REDCap requires that an opt-out message be included in all e-mail invitations. ECHO participants will be able to opt out of further e-mails by checking a box on the email indicating they do not wish to receive any other emails from our group.

Data Storage: Protection of paper and electronic records will be maintained to the best of our ability. Paper records will be stored in a locked file cabinet inside a locked office. All electronic records will be stored on password-protected institution servers behind the institution firewall. Data may be maintained indefinitely, aggregated in the future, and used for future IRB-approved research studies.

REDCap: Study data will be collected and managed using REDCap (Research Electronic Data Capture) electronic data capture tools hosted at MD Anderson.(Harris et al., 2009) REDCap (http://www.project-redcap.org) is a secure, web-based application with controlled access designed to support data capture for research studies, providing: 1) an intuitive interface for validated data entry; 2) audit trails for tracking data manipulation and export procedures; 3) automated export procedures for seamless downloads to common statistical packages. In the case of multi-center studies REDCap uses Data Access Groups (DAGs) to ensure that personnel at each institution are blinded to the data from other institutions. REDCap (https://redcap.mdanderson.org) is hosted on a secure server by MD Anderson Cancer Center’s Department of Research Information Systems & Technology Services.

REDCap has undergone a Governance Risk & Compliance Assessment (05/14/14) by MD Anderson’s Information Security Office and found to be compliant with HIPAA, Texas Administrative Codes 202-203, University of Texas Policy 165, federal regulations outlined in 21CFR Part 11, and UTMDACC Institutional Policy #ADM0335. Those having access to the data file include the study PI and research team personnel. Users are authenticated against MDACC’s Active Directory system. The application is accessed through Secure Socket Layer (SSL). All protected health information (PHI) will be removed from the data when it is exported from REDCap for analysis. All dates for a given patient will be shifted by a randomly generated number between 0 and 364, thus preserving the distance between dates. Dates for each patient will be shifted by a different randomly generated number. Following publication, study data will be archived in REDCap.

M.D. Anderson’s Research Information Systems and Technology Services (RISTS) group provides deployment and integration support for research data management using the REDCap (Research Electronic Data Capture; http://www.project-redcap.org) application.(Harris et al., 2009) REDCap (https://redcap.mdanderson.org) allows researchers at M.D. Anderson to create projects with data collection instruments and share those projects with colleagues. Researchers can capture and analyze data through a simple web-based interface. REDCap is a two-tier pHp-based web application that can be hosted on a variety of hardware and operation systems.

The application also relies on web server software and a database server to function properly. RISTS works with the Data Center Operations Team (DCOT) to deploy and configure the servers which host REDCap. The application is deployed on a RedHat Linux server that is maintained and monitored by DCOT. The open source Apache web server is used in the REDCap deployment. The servers are backed up nightly and have well-defined processes for disaster recovery. The database used to store project related information is maintained in a MySQL database cluster that is supported by a RISTS Database Administrator (DBA). The database server, like the application server, also has a well-defined process for disaster recovery. The REDCap installation also conforms to the stringent requirements for security and protected health information (PHI). Users are authenticated against MDACC’s Active Directory system. The application is accessed through Secure Socket Layer (SSL).

Training of personnel: Only MDACC personnel trained in maintaining confidentiality, the principal investigator, collaborators, data manager, and research staff will have access to study records. Data may be maintained indefinitely, aggregated in the future, and used for future IRB-approved research studies.

Data sharing: Study data will not be shared with any individuals or entities without an IRB-approved protocol. Study data will be collected and managed using RedCap.

Final disposition of study records: Data may be maintained indefinitely, aggregated in the future, and used for future IRB-approved research studies. Telephone numbers will be deleted after each participant completes the last follow up assessment. Participating sites will maintain a key that links the participants’ name to the REDCap ID, this link will not be shared with MDACC. ECHO participants’ email addresses will be deleted upon receipt of the survey response by the data manager.

REFERENCES

1. Bruera E, Yennurajalingam S: Palliative Care in Advanced Cancer Patients: How and When? The Oncologist 17:267-273, 2012

2. Logie DE. An evaluation of a public health advocacy strategy to enhance palliative care provision in Zambia. BMJ Support Palliat Care 2012;2:3 264-269 doi:10.1136/bmjspcare-2012-000232

3. Powell RA, Ali Z, E Luyirika, Harding R, Radbruch L, Mwangi-Powell F N. Out of the shadows: non-communicable diseases and palliative care in Africa. BMJ Support Palliat Care bmjspcare-2014-000751Published Online First: 21 September 2015 doi:10.1136/bmjspcare-2014-000751

4. Wairagala W. Working to improve access to palliative care in Africa. Lancet Oncol. 2010; 11: 227–228.

5. Lynch T, Connor S, Clark D. Mapping levels of palliative care development: a global update. J Pain Symptom Manage. 2013; 45: 1094–1106

6. Jemal A, Bray F, Center MM, et al. Global cancer statistics. CA Cancer J Clin. 2011; 61: 69–90.

7. Floriani C. Palliative care in Brazil: a challenge to the health-care system. Palliative care. Res Treat. 2008; 2: 19–24.

8. Arora, S., et al., Partnering urban academic medical centers and rural primary care clinicians to provide complex chronic disease care. Health Aff (Millwood), 2011. 30(6): p. 1176-84.

9. Arora, S., et al., Expanding access to hepatitis C virus treatment--Extension for Community Healthcare Outcomes (ECHO) project: disruptive innovation in specialty care. Hepatology, 2010. 52(3): p. 1124-33.

10. Arora, S., et al., Expanding access to hepatitis C virus treatment--Extension for Community Healthcare Outcomes (ECHO) project: disruptive innovation in specialty care. Hepatology, 2010. 52(3): p. 1124-33.

11. Katzman, J.G., et al., Innovative telementoring for pain management: project ECHO pain. J Contin Educ Health Prof, 2014. 34(1): p. 68-75.

12. Colleran, K., et al., Building capacity to reduce disparities in diabetes: training community health workers using an integrated distance learning model. Diabetes Educ, 2012. 38(3): p. 386-96.

13. Yennu S, Jr CEA, Weru J, et al: ECHO palliative care in Africa (ECHO-PACA): Improving access to quality palliative care. Journal of Clinical Oncology 36:6545-6545, 2018