Multi-site feasibility of a web-based and telephone navigation intervention to connect lung cancer caregivers in community oncology settings with resources: Protocol for the WF-2301CD Caregiver Oncology Needs Evaluation Tool (CONNECT) trial

Abstract

Background:

Family caregivers of patients with lung cancer experience high levels of unmet needs, burden, and distress, yet systematic processes to identify and connect caregivers with resources are lacking in community oncology. The Caregiver Oncology Needs Evaluation Tool (CONNECT) is a hybrid intervention delivered through a combination of web-based components and telephone-based navigation to inform caregivers about resources, assess their needs, and connect them to tailored supportive care services. Initial pilot testing at a single academic center demonstrated feasibility and acceptability, but feasibility in community settings remains untested. In this study, we aim to assess the multi-site feasibility of CONNECT.

Methods:

This multi-site, randomized controlled pilot trial (WF-2301CD) is conducted through the Wake Forest National Cancer Institute Community Oncology Research Program (NCORP) Research Base. Lung cancer caregiver-patient dyads (N = 120) are recruited from 12 community oncology practices and randomized to: CONNECT, generic resource list, or usual care. The primary objective is to assess multi-site feasibility, measured by caregiver retention at 12-weeks. Secondary objectives include evaluating caregiver accrual, participation, retention at 24-weeks, and acceptability, as well as process metrics. Exploratory objectives assess patient accrual, participation and retention. Data are collected via surveys at baseline, 12-, and 24-weeks, with additional process tracking by site staff and navigators.

Discussion:

This trial will provide critical data on the multi-site feasibility of implementing CONNECT in community oncology practices to inform protocol and design refinements for a future efficacy trial. If feasible, future efficacy testing will focus on the impact of CONNECT on caregiver burden and distress.NCT06383988

1. Introduction

Lung cancer is the second most common cancer in the US, with an estimated 226,650 new cases and 124,730 deaths in 2025 [1]. Treatment regimens commonly have high physical and psychosocial symptom burden, translating to an intense and complex caregiving experience for family (or “family-like”) caregivers [2–7]. Lung cancer caregivers report significant unmet needs, surpassing those of many other cancer caregiver populations [8–10]. Furthermore, lung cancer caregivers commonly experience anxiety, depression, and caregiver burden, defined as the subjective feeling of strain, which may have adverse implications for the well-being of their care-recipients [11–17].

Although caregiver interventions are typically designed to improve caregiver well-being, their potential to influence patient outcomes is of growing interest. Improvements in caregiver psychological health, skills, and capacity may translate into downstream benefits for patients, including better symptom management and engagement with care [18]. Therefore, evaluating patient outcomes alongside caregiver outcomes enhances the clinical relevance of caregiver-focused interventions and helps determine whether such interventions have broader implications for oncology care delivery [19].

Supportive care interventions, such as psychosocial support, self-care/wellness classes, and medical task training decrease burden and depression and improve well-being, satisfaction, knowledge, and skills among informal or familial cancer caregivers [20]. Our research, along with others’, shows that lung cancer caregivers are interested in supportive care interventions [21–23]. Yet, despite broad availability of supportive care resources at cancer centers and at the community and national level, uptake of these supportive care services among lung cancer caregivers is low [24–26]. Major barriers to connecting cancer caregivers to appropriate services include lack of systematic identification, assessment, and referral of caregivers by oncology providers [20,27,28].

To address these barriers, our team developed the Caregiver Oncology Needs Evaluation Tool (CONNECT) which informs caregivers about the benefits of supportive care services and identifies resource preferences to connect lung cancer caregivers with a tailored list of supportive care resources [29]. We demonstrated strong feasibility and acceptability of CONNECT in a pilot randomized clinical trial at one academic medical center in which caregiver-patient dyads were recruited. Although CONNECT is promising, we need to ensure the observed single site feasibility and acceptability translates to the multi-site community setting where trial success is more dependent on study staff with multiple competing priorities. In this study, we are collaborating with the Wake Forest National Cancer Institute Community Oncology Research Program Research Base (WF NCORP RB) to test the multi-site feasibility of CONNECT in preparation for a future, larger efficacy trial. The NCORP is a national network of community oncology practices that provides the infrastructure needed to enroll patients in clinical trials [30]. In this study, CONNECT is augmented to include brief telephone-delivered navigation sessions as a hybrid intervention, in response to caregiver feedback in the pilot trial.

2. Methods

2.1. Study overview

In this NCORP trial, lung cancer caregiver-patient dyads are recruited from 12 oncology practices and randomized to the CONNECT intervention (n = 40), a generic resource list (n = 40), or usual care group (n = 40) (Fig. 1). Although CONNECT is a caregiver intervention, patients are enrolled in the study and complete measures. This study is conducted as a planning trial (U34CA283478) and data are collected to inform a future, fully powered efficacy trial. A key component of this study is stakeholder engagement; we assembled an advisory panel that meets quarterly throughout the study to provide input and review study progress. The panel includes NCORP research staff, a lung cancer caregiver advocate, a social worker, and a cancer caregiving investigator with NCORP experience [31]. This study has been approved by the National Cancer Institute Central Institutional Review Board (WF-2301CD) and is registered as NCT06383988.

Fig. 1.

Study schema for the multi-site randomized pilot trial for CONNECT.

2.2. Objectives

2.2.1. Primary objective

To assess the multi-site feasibility of the CONNECT Intervention as measured by retention of caregivers at 12-weeks via a randomized pilot trial enrolling 120 lung cancer caregiver-patient dyads (CONNECT, n = 40; generic resource list, n = 40; usual care, n = 40).

2.2.2. Secondary objectives

2.2.2a To further assess the multi-site feasibility of the CONNECT Intervention as measured by accrual rate, participation, and retention of caregivers at 24-weeks.

2.2.2b To describe caregiver acceptability within each arm (CONNECT, generic resource list, usual care).

2.2.2c To obtain data on time needed for practice participation and central navigation, referral processes, and communication processes between the Local Practice Referral Coordinator and Central Caregiver Navigator (defined in section 2.4.1; Oncology Practices and Study Team Members).

2.2.3. Exploratory objective

To further assess the multi-site feasibility of the CONNECT Intervention as measured by accrual rate, participation, and retention of patients at 12 and 24-weeks.

2.3. Setting

This protocol is being carried out through the WF NCORP RB and includes oncology practices throughout the US that are affiliated with the WF NCORP RB. For this study, an NCORP practice may comprise of one or more affiliates/sub-affiliates that share providers and a staff member responsible for processing referrals for caregivers in the CONNECT intervention arm, referred to as “Local Practice Referral Coordinators” in this trial. We aimed to open the trial at 8–12 practices and successfully opened the study at 12 practices.

2.4. Eligibility

2.4.1. Oncology practices and study team members

NCORP oncology practices that participated in a prior feasibility survey study (WF-2300CD), received highest consideration for participation. Further, because more than 12 practices expressed interest, we prioritized site participation based on larger estimated accrual numbers, ability to contribute to a representative sample, and inclusion of rural and urban practice locations. Oncology practices were eligible if they met the following criteria: (1) provide care for ≥50 newly diagnosed (new or recurrent) stage II-IV lung cancer patients annually; (2) willing to complete an entry in the screening project for all dyads screened; (3) able to identify 1–2 Local Practice Referral Coordinators willing to assist with the CONNECT intervention delivery; and (4) able to identify a research and/or clinical champion for the study, distinct from the Local Practice Referral Coordinator. Practices able to identify only one practice referral coordinator were required to specify a back-up plan in case the practice referral coordinator became unavailable. We have included a list of participating sites that comprise the twelve practices, as well as their associated NCORP (Table 1).

Table 1.

Participating Sites for Each of the 12 Practices

NCORP Program	Corresponding Sites
ChristianaCare	Beebe Health Campus; Beebe South Coastal Health Campus
GA Cares	Augusta University Medical Center
Geisinger	Community Medical Center; Geisinger Wyoming Valley/Henry Cancer Center; Geisinger Cancer Center Dickson City
Hawaii	Hawaii Cancer Care Inc- Waterfront Plaza; Hawaii Cancer Care- Westridge
Montefiore	Montefiore Medical Center- Einstein Campus
Aurora	Aurora Saint Luke’s Medical Center; Aurora BayCare Medical Center; Aurora Cancer Care- Kenosha South
Kaiser	Kaiser Permanente- Franklin; Kaiser Permanente- Rock Creek; Kaiser Permanente- Lone Tree
Ozarks Lake Regional	Lake Regional Hospital
Ozarks CoxHealth	CoxHealth South Hospital
Prisma	Prisma Health Cancer Institute- Eastside; Prisma Health Cancer Institute- Butternut
ThedaCare	ThedaCare Regional Cancer Center; ThedaCare Cancer Care Shawano; ThedaCare Cancer Care Oshkosh; ThedaCare Cancer Care Neenah; ThedaCare Cancer Care New London; ThedaCare Cancer Care Waupaca; ThedaCare Cancer Care Berlin
Upstate	Spartanburg Medical Center; Gibbs Cancer Center Pelham; MGC Hematology Oncology Union; Gibbs Cancer Center Gaffney

Local Practice Referral Coordinators are defined as staff members from the NCORP practice site who are knowledgeable, or able and willing to learn, about available local supportive care resources and referral processes for caregivers at their practice (e.g., social worker, navigator, therapist, nurse, research staff). Local Practice Referral Coordinators are responsible for adding local resource information to the CONNECT resource database, though they may enlist support from other local site staff. They are also responsible for facilitating referrals to connect caregivers participating in the CONNECT intervention with services. They interface with the Central Caregiver Navigator (located at Wake Forest University School of Medicine [WFUSM]) for caregivers randomized to the CONNECT intervention and report on study training feedback, participation metrics and details, and suggestions to improve the future trial. See Section 2.5.1 for description of the Central Caregiver Navigator role. Study Champions are defined as research or clinical personnel with high interest and agency to support the project who can communicate its importance to various stakeholders. The Study Champion is distinct from the Local Practice Referral Coordinator. Local practice study team members completed study-related training before onboarding.

2.4.2. Caregiver-patient dyad eligibility

2.4.2.1. Caregivers.

Eligible caregivers must self-report providing informal (i.e., not professional) care during cancer treatment for a patient who meets the criteria below and who is willing to participate. Caregivers must self-report being the primary caregiver when there is more than one caregiver for a patient. They must be at least 18 years of age, as this study is focused on the experience of adult caregiver-patient dyads. They must have access to the internet or be willing to use a CONNECT study tablet in the clinic. Additionally, caregivers must have access to a telephone to complete sessions with the Central Caregiver Navigator. Caregivers are excluded if they self-report currently receiving cancer treatment themselves or are unable to read and communicate in English.

2.4.2.2. Patients.

Eligible patients must have a current diagnosis of new (i.e., initial diagnosis) or recurrent stage II-IV lung cancer. Patients must be enrolled after the start of anticancer systemic therapy, with or without radiation therapy, and have at least nine weeks of any planned anticancer treatment remaining to ensure adequate exposure to the intervention. They must be ambulatory and up (i.e., not bedridden) more than 50% of waking hours. Patients must be at least 18 years of age and receiving informal care from a caregiver who meets the eligibility criteria outlined above and who is willing to participate. Patients with stage I lung cancer are excluded due to the typically limited duration and intensity of caregiving associated with surgically treated early-stage disease. Patients are excluded if they have already completed treatment for their lung cancer at the time of study enrollment or are enrolled in hospice, as these contexts involve different care goals and caregiving needs than those observed during active treatment. Additionally, they are excluded if unable to read English and are not willing to have someone read surveys for them.

2.4.3. Caregiver and patient screening, consent, and enrollment

Local NCORP study staff typically utilize clinic schedules to identify potentially eligible patients to screen, and working closely with the oncology treatment team, approach patients in-person, with or without the caregiver, or by phone, mail, or patient portal, to confirm eligibility and interest. If the caregiver is not present, the patient is asked to nominate a caregiver and provide contact information. Eligible caregivers and patients are provided a copy of the NCI CIRB approved informed consent form to review and if interested, they are asked to provide informed consent in person or by phone. Site staff record reasons for ineligibility or study decline by adding stage II-IV lung cancer patients and caregivers to a Research Electronic Data Capture (REDCap) screening log. During quarterly advisory panel meetings, members provide feedback on screening, consent, and enrollment procedures.

2.4.4. Caregiver randomization

Following the completion of both patient and caregiver baseline surveys, caregivers are randomized 1:1:1 into one of three study groups (CONNECT, generic resource list, usual care) using a permuted block randomization scheme with random block sizes. Patients are assigned to the same study group as their caregiver but do not directly receive an intervention.

2.5. Interventions

2.5.1. CONNECT

CONNECT is a hybrid intervention delivered through a combination of web-based components and telephone-based navigation support. The intervention is designed to inform caregivers of patients with lung cancer about supportive care services by providing education about the benefits of resources, identifying their resource preferences, and connecting them with tailored resources. The multi-component intervention includes: (1) a brief educational animated video that seeks to inform and educate caregivers about the importance of self-care and benefits of supportive care resource use; (2) an assessment of supportive care preferences; (3) a tailored resource list with local and national resources corresponding to caregivers’ resource preferences; and (4) navigation services delivered by telephone to help address barriers and facilitate connection to resources. CONNECT was built by the Wake Forest Center for Healthcare Innovation to be accessed by tablet, computer, or mobile device. Tablets are available in the clinic, managed by local NCORP study staff, for participants to access the CONNECT website. When feasible, caregivers complete the web-based portion of CONNECT on a study tablet at the oncology clinic, though they can also access the web-based portion on their own device away from the clinic.

Despite evidence that caregivers express interest in supportive care resources, multiple barriers limit engagement, including limited awareness of available services, difficulty identifying resources that align with individual needs and preferences, uncertainty about how to access services, and logistical or psychosocial barriers to follow-through. CONNECT is designed to address these gaps through complementary components. The educational video increases awareness of supportive care resources and reinforces their value; the preference survey identifies individual needs and modality preferences to enhance relevance; the tailored resource list improves access by curating appropriate local and national resources; and navigation support addresses practical and psychosocial barriers, facilitates referrals, and promotes follow-through with resource use.

CONNECT components:

1. Educational Video. The 1.5-min animated video describes the multidimensional challenges caregivers face and discusses the benefits of utilizing supportive care resources.

2. Preference Survey. The preference survey includes 18 items to assess caregiver interest in accessing resources in the following domains: cancer education; emotional and social support; communication; finances, transportation and lodging; spiritual and religious; health and wellness; decision making; employment challenges; patient care needs; and respite care. If a caregiver expresses interest in a resource domain, follow-up questions ask about service modality preferences (e.g., one-on-one counseling versus group; in-person versus virtual).

3. Tailored Resource List. The tailored resource list includes local and national resources matched to items selected by a caregiver in the preference survey. Each item on the preference survey has local and/or national-level resources, with local resources varying across practice groups. In the absence of a local resource for a specific caregiver interest, only the relevant national level resource(s) displays. All resources are stored in a Resource Database within CONNECT. The tailored resource list also includes information pertaining to modality (e.g., accessible in person, by telephone, or both) and cost (e.g. free, sliding scale, accepts insurance), when available, to support decision making. Modality and cost information are especially important for caregivers with transportation or financial barriers [32–34]. After completing the preference survey, caregivers can view their tailored resource list in real time and also receive it by email and/or mail.

4. Navigation Services. Navigation sessions are provided by a member of the WFUSM team who received a certificate of completion in the George Washington Patient Navigation Training. The Navigator contacts all caregivers enrolled on the CONNECT arm by telephone within 48 business hours of caregivers receiving their tailored resource list. The Navigator reviews the resource list with the caregiver, assist with problem-solving around potential barriers to accessing resources, and works with the Local Practice Referral Coordinator to process referrals. The Navigator will guide the caregiver in setting an actionable time-bound plan for accessing resources that are available “on demand” (e.g., web-based resources) without a referral. A navigation guide, developed by the study team and with input from the advisory board, is used for the navigation calls. A letter is emailed or mailed, based on the caregiver’s preference, approximately 2 weeks after the initial call to maintain contact and remind them of the upcoming follow-up call. The Navigator follows up with the caregiver by telephone approximately 4 weeks after the initial navigation call to assess whether the resource(s) have been accessed, determine the degree to which it has met the caregiver’s need, recommend additional resources to align with caregivers’ potentially fluctuating needs, and problem solve if they have not accessed the resource or if it did not meet their need. Caregivers will be provided with contact information for the Navigator and instructed to call if they have a new need or if they would like additional assistance with accessing a resource before the 4-week call; these additional contacts are tailored to caregiver needs, while a minimum of two standardized navigator-initiated calls ensures a consistent intervention dose.If a caregiver requests information about new resources during any navigation call, information regarding these resources may be provided to the caregiver via email, mail, or text message, pending caregiver consent for text messaging.

2.5.2. Generic resource list

Caregivers randomized to the generic resource list group are provided with a comprehensive and untailored (generic) resource list via mail and/or email by the CONNECT Study Team. This list will consist of resources consistent with all domains assessed as part of CONNECT as described in section 2.4 (Preference Survey). Local NCORP site staff are instructed to interact with caregivers participating in the generic resource list group as they normally would.

2.5.3. Usual care

Caregivers assigned to usual care will be treated no differently than a caregiver not on the study. Caregivers randomized to the usual care group will receive standard care for the local NCORP practice which may include offering supportive care services such as counseling, support groups, and wellness classes, though this will vary across practices. They will interact with practice supportive care staff, as normal. Usual care practices are recorded at each practice. Caregivers in the usual care group will be provided with the generic resource list via mail and/or email by the CONNECT Study Team after completing the 24-week assessment.

2.6. Assessments and outcome measures

A description of study surveys is included in Table 2. Feasibility metrics are tracked throughout the study. Although primarily focused on feasibility, we are also collecting caregiver and patient data on potential primary and secondary outcomes for the future efficacy trial that this study supports. In addition, we will evaluate the feasibility of collecting these data in caregivers and patients. Although caregivers and patients are enrolled as dyads, accrual, participation, and retention will be examined separately for each group to characterize feasibility at the individual level. This approach allows identification of distinct barriers to participation among caregivers and patients and will inform decisions about the potential inclusion of patient outcomes in the future trial.

Table 2.

Assessment plan

Assessment Domain	Time	Description
Multi-site feasibility	Baseline, 12- & 24-week follow-ups	Accrual rate, participation, and retention (primary feasibility outcome). Data will be captured in standard WF NCORP RB forms capturing screening, registration, and on/off study status.
Caregiver acceptability	12-week follow-up	Self-report survey developed for this study to assess the degree to which caregivers found different aspects of CONNECT, the generic resource list, or usual care helpful (using a Likert response from 0 (Not at all) to 5 (Very much)). For the CONNECT arm, the assessment is divided into 2 parts: Part 1 is delivered immediately after the web-based portion of the intervention within CONNECT to mitigate recall bias and Part 2 is delivered at 12-weeks.
Preliminary data to assess multi-site feasibility of collecting caregiver and patient data to inform future trial
Caregiver & patient sociodemographics, health-related social needs (HRSN; caregivers only), and health behaviors	Baseline	Age, sex, race/ethnicity, education, employment, income, marital status, and insurance status. We will assess the following HRSN domains from the AHC HRSN Tool [35] (housing, food, transportation, utilities, and financial strain). Additional questions about health behaviors (e.g., tobacco use, physical activity, sleep) are included.
Patient clinical information	Baseline	Clinical data abstracted include AJCC Stage, tumor site, cancer history, and treatment history
Caregiving characteristics	Baseline	We will assess relationship to patient, days per week of care provided, average daily care hours, other caregiver responsibilities (child/parent), and respite care availability.
Participant-reported outcomes Caregivers	Baseline, 12- & 24-week follow-ups	Zarit Burden Interview: 12-item instrument assessing stresses experienced by caregivers; demonstrated validity and reliability and has been used widely with cancer caregivers, including lung cancer caregivers [36–40]. PROMIS Emotional Distress-Anxiety-Short Form 8a: 8-item instrument that assesses fear, anxious misery, hyperarousal, and somatic symptoms related to arousal; widely used in oncology populations [41]. PROMIS Emotional Distress- Depression Short Form 8a: 8-item instrument that assesses sadness, guilt, self-criticism, worthlessness, loneliness, interpersonal alienation, as well as loss of interest, meaning, and purpose [41]. Caregiver Quality of Life Index-Cancer (CQOLC): 35-item instrument assessing dimensions of caregiver quality of life (burden, disruptiveness, positive adaptation, financial concerns). The CQOLC has demonstrated validity and reliability [42]. Supportive Care Resource Use (12 & 24 week only): Self-reported supportive care service use (e.g., counseling, support groups, wellness classes) will be recorded using a structured questionnaire developed in the pilot study with additional questions to capture adherence to potential intervention-related referrals for caregivers participating in the CONNECT Intervention arm [29].
Participant-reported outcomes Patients	Baseline, 12- & 24-week follow-ups	FAMCARE (Perceived care quality): 20-item adapted instrument with 4 subscales to assess patient satisfaction with information giving, availability of care, physical care, and symptom control [43]. Cancer symptoms: 16-item MDASI-LC instrument to assess lung cancer specific and general symptoms and the impact on daily functioning [44,45], plus 7 items related to 3 additional symptoms) from the National Cancer Institute Patient Reported Outcomes version of the Common Terminology Criteria for Adverse Events (NCI-PRO-CTCAE) [46,47]. These assessments were added to assess the potential severity, frequency, and interference of immune-mediated effects that may be experienced by patients on immunotherapy or chemo-immunotherapy (arthralgia, myalgia, pruritus). These items were chosen based on their frequency and severity as reported by patients with lung cancer receiving immunotherapy in routine clinical practice, but will be asked of all patients for consistency [48].
Unplanned healthcare utilization Patients	Baseline, 12- & 24-week follow-ups	Three questions assessing number of visits since the last assessment (emergency room [ER], urgent care, and oncology and non-oncology healthcare providers). For ER and urgent care, the location of care will be solicited. For provider visits, patients will indicate the number of unplanned visits and provider type. Self-reported questionnaires are a valid method of data collection for healthcare utilization. We will cross-validate with data from the EHR chart abstraction. Using standard methods patterned on the Quality Oncology Practice Initiative [49–51], a research nurse at each participating practice will use a structured data entry tool to abstract data from oncology medical records.
Local Practice Referral Coordinator
Referral Coordinator Practice Intervention Start-up Time and Processes	Prior to first participant accrual	Time (minutes) required for: (1) completing study-related training; (2) identifying local resources, and (3) entering resource information into the CONNECT database. Satisfaction with and suggestions to improve study training will also be collected. Because more than one person may contribute to 1–3 above, the referral coordinator who was most engaged in these steps will complete this survey and consult with other personnel who contributed as needed.
Referral Process Tracking (CONNECT Arm Only)	Throughout randomized pilot trial	For each of the caregivers randomized to the CONNECT Intervention arm, data will be reported on referral strategies, number of contacts with caregivers and the Central Caregiver Navigator, and methods/ min used to communicate with caregivers and the Central Caregiver Navigator (e.g., telephone, virtual web-based conversations, email). Because it is possible that more than one referral coordinator may interact with a caregiver, the referral coordinator who was the most engaged will complete this survey and consult with the other coordinator (if available), as needed.
Referral Coordinator Improvements in the Future Trial	At the conclusion of practice intervention activities	Survey to assess suggestions for improvements in the future fully-powered trial such as: (1) entering and updating resources within the CONNECT resource database; (2) accessing the tailored resource list for each caregiver; (3) communication processes with the Central Caregiver Navigator; (4) ease of, and process for conducting participant referrals; (5) capacity to integrate this role into current clinic workflow; (6) perceptions of the trial among clinic staff; (7) perceived impact of intervention on caregivers and patients; and (8) unanticipated challenges with study staffing and implementation of new caregiver screening or services at the clinic. This survey will be administered to the 1–2 Local Practice Referral Coordinators that are available at each practice.
Central Caregiver Navigator Data
Navigation Tracking	Throughout randomized pilot trial	Referrals made (number and types), number of minutes interacting with caregivers, and number of contacts with caregivers.

The study advisory panel members reviewed and provided feedback about the study measures. Caregivers and patients complete a survey at baseline to assess sociodemographics, health behaviors, and health-related social needs (this latter survey is completed by caregivers only). At baseline, 12-weeks, and 24-weeks, caregivers complete additional surveys to assess caregiver burden, emotional distress, quality of life, supportive care resource use (12-and 24-weeks only), and acceptability of the intervention (12-weeks only). Patients complete surveys at baseline, 12-weeks, and 24-weeks to assess satisfaction with care, cancer symptoms, and unplanned healthcare utilization. Site Local Practice Referral Coordinators complete a survey after finishing study start-up processes and throughout the trial to collect process data, and at the conclusion of site intervention activities. Patient clinical characteristics are abstracted from the electronic health record within 14 days following baseline and at 12-weeks and 24-weeks. The preferred method of survey completion is to have the participants complete them directly into REDCap while in the clinic or at home; alternatively, participants may complete paper surveys in the clinic, by phone, or mail. All participants receive a $20 gift card upon completion of surveys at each time point (baseline, 12-weeks, and 24-weeks).

2.7. Design and analysis

This is a multi-site randomized controlled pilot trial to assess the multi-site feasibility of CONNECT to prepare for a future, larger efficacy trial.

2.7.1. Sample size, accrual and feasibility

Our sample size of 120 caregivers is based on the primary feasibility outcome of caregiver retention at 12-weeks. As the study enrolls dyads, 120 patients will also be accrued. We anticipate accruing at least 10 dyads per month over a 12-month period to reach our anticipated target of n = 120 dyads. If the caregiver retention proportion at 12-weeks is less than 75%, we will deem a future study infeasible in this current form. If the true probability of caregiver retention at 12-weeks is 75%, we expect 82 or more caregivers to be retained (i.e., the probability of fewer than 82 being retained if the true probability of retention is 0.75 is less than 5%).

2.7.2. Primary objective, endpoint, and analysis plan

Our primary outcome is caregiver retention at 12-weeks, which will be defined as the number of caregiver participants who complete at least 75% of the study measures at 12-weeks divided by the number of caregivers randomized. All measures defined in Table 2 for caregivers at 12-weeks will be used in the calculation of the retention proportion. Retention will be estimated and reported along with an exact 95% confidence interval. If the retention proportion at 12-weeks is significantly lower than 75%, we will deem a future study infeasible in this current form. Retention at 12-weeks will also be estimated separately for each randomization arm and will be compared between arms using Fisher’s exact test.

2.7.3. Secondary objectives, endpoints, and analysis plans

In addition to the primary feasibility measure of caregiver retention at 12-weeks, we will also measure average accrual rate per month (total number of caregivers accrued divided by number of months spent accruing) and participation (proportion of eligible caregivers who agreed to participate). We will also compute retention at 24-weeks (number of caregivers who complete at least 75% of the 24-week study measures divided by the number randomized; all measures given to caregivers at 24-weeks as defined in Table 2 will be used in the calculation). We anticipate that 50% of eligible caregivers will agree to participate, and that we will retain 80% of the starting sample of participants at 24-weeks. We will examine heterogeneity in caregiver accrual, participation, and retention by practice and heterogeneity in retention by randomization arm. Missing data will be described for each caregiver measure at each time point to help inform data collection and outcome measures for the future trial. Acceptability of the intervention will be summarized quantitatively. Percentages of caregivers reporting that different aspects of CONNECT, the generic resource list, or usual care were quite a bit or very much helpful will be calculated.

Time required (in minutes) for: (1) completing study-related training; (2) identifying local resources (3) entering resource information into the CONNECT database, and (4) interactions with the Central Caregiver Navigator and caregivers will be collected from Local Practice Referral Coordinators. Number of contacts with each caregiver, modality of contact, and type of referral facilitated will also be recorded. In addition, Local Practice Referral Coordinators will report on frequency of needed resource updates (e.g., replacing resources no longer available) during the study period, and their perspectives on improvements for the future trial. Minutes, number of contacts, and number and type of referrals will also be collected from the Central Caregiver Navigator to evaluate time spent on each caregiver. Time metrics will be summarized across practices using mean, median, standard deviation, and interquartile range. Preferred modality of training and method of communication with the Central Caregiver Navigator and caregivers, types of referrals facilitated by the Central Caregiver Navigator and Local Practice Referral Coordinators, and other survey responses will be summarized using appropriate descriptive statistics.

2.7.4. Exploratory objectives, endpoints, and analysis plan

Average patient accrual rate per month (total number of patients accrued divided by number of months spent accruing), patient participation (proportion of eligible patients who agreed to participate), and patient retention (the number of patients who complete at least 75% of the study measures at 12-weeks and 24-weeks divided by the number randomized) will be calculated. Patient retention proportions will be calculated using all measures defined for patients in Table 2. We will also examine heterogeneity in patient accrual, participation, and retention by practice and heterogeneity in retention by randomization arm. Missing data will be described for each patient measure at each time point to help inform data collection and outcome measures for the future trial. Descriptive statistics will be used to characterize changes in both caregiver and patient measures over time within each arm to further inform the future trial.

3. Discussion

In this randomized controlled pilot trial, we are assessing the multi-site.

feasibility of the CONNECT intervention for lung cancer caregivers across twelve community oncology clinics in the NCORP network. We are also collecting practice-level data to guide refinements and inform planning for a future efficacy trial. Lung cancer caregivers have expressed unmet needs and interest in supportive care interventions, though a process to assess caregivers’ resource preferences and connect them with resources has been lacking in community oncology practices [20–23,27,28]. This trial is an important first step to determining the feasibility of delivering this type of intervention to address lung cancer caregiver needs in the community, where most cancer patients receive oncology treatment. To our knowledge, CONNECT is the first intervention to facilitate identification of lung cancer caregiver supportive care resource preferences and potential connection to resources both locally and nation-wide; further, we believe this type of intervention approach has been underutilized in cancer caregivers.

Determining the feasibility of a multi-site trial is a critical step before undertaking a larger efficacy study that requires substantial additional time and resources, particularly when the intervention is being implemented in a new context such as a broad network of community oncology practices. Although the primary feasibility outcome is fixed, we recognize that a more comprehensive assessment of feasibility will strengthen decision-making regarding progression to a future trial. For example, caregiver adherence to the CONNECT intervention will be an important factor and will be carefully evaluated alongside other metrics related to feasibility (e.g., participant engagement time with the intervention) when determining whether it is appropriate to proceed to a larger-scale trial. In addition, because participation in NCORP requires established research infrastructure, NCORP practices may not be fully representative of community oncology practices more broadly. Consequently, if findings suggest CONNECT is feasible in NCORP community oncology clinics, it cannot be assumed that this intervention would be similarly feasible in non-NCORP community settings.

CONNECT was designed to be adaptable to routine oncology care by leveraging existing personnel and resources. While certain roles in this study (e.g., champions) are specific to the research context, other functions, such as identifying and facilitating access to supportive care resources, may be integrated into existing roles (e.g., social workers, navigators, or therapists). Additionally, caregiver navigation support may be provided or supplemented by external organizations (e.g., nonprofit organizations such as the American Cancer Society). Together, these considerations support the potential scalability and sustainability of CONNECT beyond the research setting.

This trial is currently enrolling caregivers and patients. Considering the relatively limited NCORP practice research experience with engaging with and recruiting dyads, our study team provided site training for these activities prior to study activation [52]. Although we have observed steady recruitment of dyads into this trial, our recruitment goal of 120 dyads in a limited time frame is ambitious and has required ongoing training, site support, and championing the study through various channels such as the WF NCORP RB broadcast emails, bi-monthly meetings, and the annual research base meeting. Our screening and participation data provided in this trial will allow us to adjust our eligibility criteria, if needed, for a future trial to support more robust recruitment.

The CONNECT intervention is currently offered only to caregivers who can speak and understand English, thereby limiting its reach to a population of potentially underserved caregivers [53]. While the CONNECT intervention materials could be culturally and/or linguistically adapted in future work, the CONNECT intervention includes a database of existing resources that are primarily only available in English. More advocacy and funding is warranted to support the development of resources available to non-English-speaking caregivers. Further, although we are testing CONNECT with lung cancer caregivers in this trial, the CONNECT intervention was developed for cancer caregivers broadly and has the functionality to tailor some resources by cancer disease type. In this trial, we have included resources that are lung cancer caregiver specific, as well as disease agnostic.

In summary, the CONNECT intervention has significant potential to increase connection to resources, both locally and nationally, for lung cancer caregivers during patient’s systemic treatment. If multi-site feasibility is determined in community oncology practices, an important next step is an NCORP Phase III efficacy trial to determine the impact of CONNECT on key caregiver outcomes (e.g., caregiver burden) that will be identified in this feasibility trial. At the conclusion of this trial, we will engage the advisory panel to review findings and help refine the protocol for future testing.

Glossary

CONNECT

Caregiver Oncology Needs Evaluation Tool

WF NCORP RB

Wake Forest National Cancer Institute Community Oncology Research Program Research Base

REDCap

Research Electronic Data Capture

HRSN

Health-related social needs

ER

Emergency Room

Data availability

No data was used for the research described in the article.