Table 1

Summary table of study metrics

Implementation setting

Three initial sites were included in this early implementation phase: the University of Iowa, the University of California at Los Angeles (UCLA), and the Medical University of South Carolina (MUSC). Each location identified a clinic with physician champions for the initial rollout. The University of Iowa Hospitals and Clinics (UIHC) is the tertiary care, academic medical institution in the state of Iowa, which serves the majority of the state of Iowa as well as many patients from western Illinois and other surrounding midwestern states. Holden Comprehensive Cancer Center (HCCC) is the matrix cancer centre at UIHC. HCCC provides patient care on the main hospital campus is delivered across five adult and one paediatric outpatient clinics, two infusion suites, a stem cell transplant and cellular therapy unit and inpatient nursing units as well as through outreach clinics across the state. In FY23, patients were seen from all 99 counties in Iowa as well as 36 US states and had over 100 000 annual visits to the clinical cancer centre and infusion suites. The ACS CARES programme will be implemented in the Gynecologic Oncology Clinic. The UCLA Health Jonsson Comprehensive is a designated Comprehensive Cancer Center by the National Cancer Institute (NCI) and part of the National Comprehensive Cancer Network (NCCN). UCLA Health Jonsson Comprehensive Cancer Center serves the entire cancer population of Los Angeles County (>20 million) from children through adulthood through its four hospital locations and over 100 ambulatory clinics. The ACS CARES program at UCLA will be piloted in three separate and highly diverse ambulatory clinics: (1) Adult Neuro-Oncology, (2) Pediatric Bone and Soft Tissue Sarcoma and (3) Pediatric Leukemia and Hematopoietic Stem Cell Transplant. The latter two clinics see a roughly even split between Medicaid (MediCal) and Preferred Provider Organization (PPO) insurance with roughly 20%–30% of families with Spanish as their preferred language. The MUSC is the only NCI-designated cancer centre in South Carolina. They have a strong focus on the care of underserved patients and those in rural areas, serving as 1 of only 14 NCI Minority/Underserved Community Oncology Research Program sites and have a strong existing nurse navigation programme. The ACS CARES programme will be piloted at MUSC in the head and neck radiation oncology, genitourinary medical oncology and thoracic surgery clinics.

Volunteer non-clinical navigation

Navigation will be implemented as standard of care for patients receiving care with a participating oncologist offering the intervention, thus no consent will be required. The navigation process and activities are shown in figure 1.

The oncologist or other healthcare team member will introduce the programme to new patients as an approach for routine patient support. ACS CARES volunteers then approach new patients in-person to introduce the programme, discuss potential benefits and capture the best method for future contact (encrypted messaging vs phone). The volunteers guide the patients through any technical aspects of participation, such as signing up for secure messaging. Patients will be allowed to opt out of participation. After enrolment, the volunteer will complete a social determinants of health screen (table 2), which will guide navigation activities including connection to other members of the clinical team, health system resources and community resources. Initial contact will be predominantly in person, but phone contact will be permitted. After the initial engagement, patients will be contacted by the volunteer weekly, monthly or every 6 months based on a severity score calculated based on number of barriers and type of barrier. Volunteers will document using a navigation software all encounters with patients, screening surveys administered, barriers encountered and resolved, and resource referrals. Clinical teams have access to dashboards of activities completed both for individual patients and in aggregate.

Staff training

We offer a two-part training curriculum for volunteers. The initial training is self-paced, online and includes the following modules: privacy training, IT security awareness for volunteers, introduction to phishing, IT security and acceptable use, ACS programmes and services, ACS CARES programme overview, introduction to patient navigation, history of navigation, cancer 101, identifying and addressing needs across the cancer continuum, financial basics, health equity, clinic student volunteer role expectations, introduction to cancer caregiving, introduction to childhood and AYA cancer. On completion of this online training, an in-person training is held at each site that includes a tour of the clinic space, appropriate documentation practices, role-playing practice scenarios and integration of all aspects of the role. Finally, the volunteers learn and practice documentation. The lead oncologist and administrators are invited to training and are responsible for educating other oncologists, nurses, social workers and clinical staff about the programme. Materials are provided to support this effort, including a Clinic Flow Sheet describing volunteer responsibilities and key opportunities for clinic integration, and an In and Out of Scope Document detailing potential opportunities for volunteer engagement. The opportunity to customise clinic integration and volunteer engagement to the needs and preferences of the clinical team is offered at this time. The number of volunteers completing each training component is recorded. A super-user from ACS completes weekly check-ins and continuing education with the volunteer teams and the ACS CARES team provides support on an ongoing basis, including bi-weekly meetings with the pilot site teams.

Implementation frameworks

The Updated Consolidated Framework for Implementation Research (CFIR) by Damschroder and colleagues is optimal to guide the evaluation of the ACS CARES programme. The updated CFIR focuses on key domains: innovation, outer setting, inner setting, individuals, implementation process and outcomes.15 16 Proctor’s Implementation Outcome Framework17 will be used in the assessment of implementation outcomes and inform future implementation effort expectations.

Intervention context

Sample size considerations

The sample size for this portion of the study will be based on the volume of new patients seen by participating oncologists; thus, no formal sample size calculation is needed.

Design

This component will include secondary, descriptive analysis of existing data capture both within the electronic medical record and electronic navigation platform. No active recruitment will occur as part of this study component. Pilot sites will abstract data from the electronic medical record (EMR) and provide ACS with a limited dataset for analysis from two time periods.

Populations of interest

Patients (1) age ≥18, (2) with an ICD-10 diagnosis codes for an eligible cancer at participating institution. Each site will define their pilot population based on the oncologists participating in the pilot programme. Sites will differentiate between new eligible patients and all eligible patients. The baseline patient population will include patients seen by participating oncologists from 1 January 2022 to 31 December 2022. The intervention patient population will include patients seen from 1 September 2023 to 31 August 2024. The volunteer population will include all volunteers participating in the programme.

Data abstraction

The following patient data from the EMR will be abstracted: age, sex, race and ethnicity, insurance status, home address (to determine rurality), cancer type, cancer stage, diagnosis date, date of new patient appointment, treatment types (surgery, chemotherapy, radiation therapy, targeted therapy, immunotherapy, stem cell or bone marrow transplant, hormone therapy), total clinic appointments, # clinic appts missed/cancelled, total infusion appointments, # infusion appointments missed/cancelled, total radiation treatment appts, # radiation treatment appointments missed/cancelled, enrolled in a clinical trial (yes/no), start date of clinical trial participation, emergency room (ER) visits, hospitalisations, ICU admissions, total cost of care. Sites will use home addresses to characterise patients as living in urban or rural residences (RUCA codes) and in areas of high or low area deprivation as a measure of socioeconomic disadvantage (Area Deprivation Index scores).19 Chart review will be completed as needed to minimise missing data. Navigation data will be abstracted from the NurseNav system (navigation platform technology). From this system, data will be abstracted monthly and will include the number of patients/caregivers approached, enrolled and declined ACS CARES participation; the number of patients/caregivers supported by ACS CARES who received psychosocial distress and social determinants of health screening; the number of follow-ups conducted (in person and remote); the number of shifts completed per volunteer; the number of referrals to the programme received.

Outcomes

Access will be defined by the following metrics for the above populations: (1) time to treatment=the time from initial appointment with the clinical team to first treatment; (2) time to volunteer navigation=time from new patient appointment with the clinical team to first navigation contact; (3) time from volunteer navigation to first treatment=time from first volunteer navigator appointment to first treatment; (4) receipt of clinical trial during study period. This will be captured overall and by type of treatment (surgery, medical therapy, radiation). Healthcare utilisation will be defined by the per cent of patients with each type of event (ED, hospitalisation, ICU admission, no show to clinic appointment). Each of the event types will be described as events per month. The average total cost of care will also be calculated monthly. All outcomes will be assessed for all patients (navigated, not navigated, both) and for new patients (navigated, not navigated, both) from June 2023 to September 2024 from the target population.

Statistical analysis

We will summarise sociodemographic characteristics, adherence and utilisation measures by navigated/not navigated with means and SD for continuous variables, counts and proportions for categorical variables and compare with t tests and χ² tests. We will use multivariable logistic regression or multivariable generalised linear regression with a beta distribution and logit link, depending on the outcome for each analysis. Site and volunteer will be treated as fixed effects when needed, and we will consider two-way interaction terms for all covariates.

Power and sample size considerations

The three participating sites will select the clinics for implementing ACS CARES; thus, power and sample size will be dependent on the enrolment at the sites and not predetermined. We anticipate each site to enrol at least 200 patients, for a minimum of 600 patients across sites.

Population

Volunteers participating in ACS CARES at the three pilot sites. We will be working closely with the volunteers implementing the ACS CARES programme throughout the pilot project. Each of these volunteers committed to at least 1 year of participation as part of their selection for participation in the project.

Data collection

For evaluation of volunteer experience, all volunteers will be asked to participate in a 15 min, 30-question Volunteer Satisfaction Index survey20 in May 2024 assessing perspectives on the programme. Participation will have no impact on the ability to continue volunteering for the programme.

Outcomes

Volunteer satisfaction will be measured by the Volunteer Satisfaction Index,20 which consists of 26 items in four subscales for organisational support, participation efficacy, empowerment and group integration. Item responses are given using a 7-point Likert scale ranging from strongly dissatisfied to strongly satisfied. It has previously demonstrated satisfactory alpha coefficients for all subscales, ranging from 0.75 to 0.91.

Sample size considerations

While we expect high participation (>75%) from the volunteers based on prior volunteer engagement, we note that the sample size for volunteers will be small (estimated 5–10 per site) due to the limited volunteer pool and thus will be hypothesis-generating for future studies.

Distress and social determinant of health analysis

Cross-sectional survey

Synthesis into implementation blueprint

Within the implementation of ACS CARES, we will develop a formal implementation blueprint,30 31 which will include processes for volunteer navigation, expected outcomes, information on the context and training. This blueprint will provide guidance to future sites interested in implementing ACS CARES. This will not include any patient-specific or provider-specific data.