Stakeholder perspectives on overcoming barriers to cascade testing in Lynch syndrome: A qualitative study

Abstract

Cascade testing (i.e., genetic testing of family members of individuals with disease) among families affected by hereditary cancer disorders, such as Lynch syndrome (LS), is suboptimal and thus represents a missed opportunity in cancer prevention. We aimed to fill a gap in the literature by exploring multilevel barriers and facilitators to the implementation of cascade testing for LS. We conducted semi-structured, in-depth interviews guided by the Consolidated Framework for Implementation Research and the Integrated Behavioral Model among key stakeholders (n=60): LS patients and relatives (n=20), providers (n=20) and administrators (n=20). Transcripts were double-coded (20% sample) using template analysis in ATLAS.ti. Barriers identified included (1) low awareness about LS, (2) psychosocial barriers, (3) lack of provider follow-up, (4) accessibility to genetic counseling, and (5) fear of discrimination. Facilitators included (1) motivation to engage in cascade testing, and (2) free genetic testing offered to relatives. Stakeholders also recommended strategies to overcome implementation barriers in the short-term (increasing education, preparing patients for communicating with relatives), medium-term (optimizing clinical workflow and staffing resources) and long-term (nation-wide standardization). These findings indicate that modifiable, multilevel barriers to the implementation of cascade testing in LS are experienced across stakeholders. Understanding and targeting implementation barriers is imperative to achieving public health impact of precision health interventions such as cascade testing.

Introduction

Lynch syndrome (LS), also called hereditary nonpolyposis colorectal cancer (HNPCC), is an inherited genetic condition arising from variants in DNA mismatch repair (MMR) genes that increases the likelihood of developing several cancers, including colorectal, endometrial, gastric and ovarian cancers. One in 279 of the population carry these variants(1), and bear elevated lifetime risks for cancers: up to 82% for colorectal and 60% for endometrial cancers(2), compared to 4% and 3% in the general population, respectively.(3,4)

Identifying individuals with LS has become a national priority for cancer prevention efforts including priorities set by Health People 2020(5) and the Cancer Moonshot Blue Ribbon Panel(6), as timely identification is critical for preventing and initiating surveillance for LS-associated cancers. However, an estimated 98% of individuals with LS are unaware of their condition.(7) Cascade testing, the process of testing family members of individuals with confirmed variants for the genetic condition, is a key strategy for increasing the rate of identification of individuals with familial cancers, and the National Comprehensive Cancer Network (NCCN) guidelines recommend that family members of individuals with LS receive genetic counseling and testing.(8) One study estimated that over half of colorectal cancer deaths attributable to LS in the US (~6,500) could be averted if family members were identified through cascade testing and received subsequent risk management.(9) Yet, 52% or fewer first-degree relatives of individuals with LS receive cascade testing and uptake is even lower among more distantly-related family members.(10)

Prior studies on understanding factors affecting cascade testing for LS in the US have focused on the association between baseline characteristics of probands (index patients) and communication with family members about LS.(11,12) However, to gain a better understanding of how cascade testing can be improved to enhance management of LS-related cancers, it is necessary to understand the multilevel factors that enable or hamper cascade testing in clinical practice. The objective of this work was to identify barriers and facilitators to cascade testing for LS using qualitative research methods informed by the Consolidated Framework for Implementation Research (CFIR)(13) and the Integrated Behavioral Model (IBM).(14) We sought perspectives from diverse stakeholders who have been affected by the implementation of cascade testing for LS, including stakeholders at the patient-, provider- (i.e., genetic counselors, specialists, primary care providers), and organizational- (i.e., genetic testing labs, genetics clinic managers, patient advocates and payers) levels. Data from these interviews will inform a future multilevel intervention to improve cascade testing uptake among families with LS.

Materials and Methods

Participants and Recruitment

Research procedures were conducted in accordance with the U.S. Common Rule, following approval by the University of North Carolina at Chapel Hill Institutional Review Board (IRB). This study adhered to the Consolidated Criteria for Reporting Qualitative Research (COREQ) reporting guidelines (see Supplemental Methods).(15) Semi-structured interviews were conducted with stakeholders at the patient-, provider- and organizational- levels (n=60) to understand barriers and facilitators to cascade testing for LS, and a flow diagram of the study design is presented in Figure 1. Multiple strategies were utilized the recruit the diverse types of participants in this study.

Figure 1. Flow diagram of qualitative study design:

recruitment, interviews and analyses procedures used for generating barriers, facilitators and recommendations for cascade testing in LS

Patients.

LS patients and relatives of patients (n=20; 17 were positive for LS variants) were recruited through the Facebook page of Lynch Syndrome International (LSI), a web-based patient advocacy organization with a strong social media presence.(16) Respondents who were 18 years or older, had LS or were related to patients with LS, resided in the US, and were English-speaking were included in the study. We used purposive sampling to recruit a demographically diverse sample of patients (gender, race/ethnicity).

Providers.

Given the wide range of providers and administrators that may be involved in cascade testing throughout the care continuum, sampling was focused on the most salient types of providers and administrators such that we attained saturation of themes across patient, provider and administrator stakeholder groups. Twenty providers were recruited using both a recruitment email (n=12) and purposive sampling (n=8). A recruitment email was sent to the listserv of the Collaborative Group of the Americas on Inherited Gastrointestinal Cancers (CGA-IGC)(17), a professional organization focusing on inherited gastrointestinal cancers including LS. Purposive sampling(15) was used to ensure that a broad range of medical specialties that may see LS patients in clinical practice were represented in this stakeholder group: genetic counselors (n=10) and non-genetics providers, including gastroenterologists (n=2), colorectal surgeons (n=2), oncologists (n=2), primary care physicians (n=2) and gynecologic oncologists (n=2).¹⁵ Genetic counselors were oversampled, as discussion about familial risk and disclosure about genetic information is traditionally embedded in genetic counseling practice.

Administrators.

To gather perspectives from administrators of the diverse organizations which play a role in cascade testing, representatives were recruited from genetic testing laboratories, patient advocacy groups, payer organizations and genetics clinics (n=20). Genetic testing laboratories (n=5) and patient advocacy and support groups for LS (n=5) were first identified through purposive sampling, and relevant representatives (staff such as directors or managers) were made available for interview by each organization. Medical directors or pharmacy directors of payer organizations (n=5) with knowledge of hereditary cancer conditions were identified and made available for interviews by Medical Marketing Economics, LLC (MME), a healthcare research and consulting firm with payer relationships, through an internal process.(18) Finally, program managers or directors of genetics clinics (n=5) were identified through the national directory of professionals available through the National Society of Genetic Counselors (NSGC).(19)

Semi-structured Interviews

The IRB determined that the study design justified a waiver of written (signed) consent according to section 45 CFR 46.117(c)(2) of the U.S. Common Rule, which denotes that the research presents no more than minimal risk of harm to subjects and involves no procedures for which written consent is normally required outside of the research context. Verbal informed consent was obtained from each participant prior to the interview, and no participant withdrew from the study after giving their consent. The purpose of the study was shared with participants in this process. A semi-structured interview guide was developed, informed by the CFIR, which comprises five domains: (1) characteristics of individuals involved, (2) characteristics of the intervention, (3) inner setting, or the organizational context within which the implementation of the intervention will occur, (4) outer setting, or the economic, political and social environment within which the organization is situated, and (5) the implementation process, allowing a comprehensive examination of factors that influence the implementation of cascade testing in clinical practice. Other than the construct of “Other Personal Attributes” in the individual characteristics domain, CFIR does not provide a way to capture nuanced individual-level behavioral constructs. Therefore, to obtain a richer analysis of patient behavior and attributes, the patient interview guide was further informed by constructs from the IBM, which informs individuals’ intention or readiness to perform cascade testing.

The guide was developed in consultation with key stakeholders (patient advocates and clinical experts), using open-ended questions focused on understanding the individual’s views on and experiences with the process of cascade testing for LS. These guides elicited barriers and facilitators to cascade testing and gathered perspectives on improving the implementation process of cascade testing for LS in clinical practice. The interview guide was adapted to each stakeholder group by tailoring semi-structured interview prompts (e.g. patients were asked: “What resources or support has helped you contact family members to initiate cascade screening?”, whereas administrators at genetic clinics were asked: “What resources does your practice use to facilitate implementation of cascade testing for patients with LS?”). Interview guides were reviewed by each relevant stakeholder prior to utilization in the study.

Relationships with participants were not established prior to the study. The interviews were conducted privately via telephone by University of North Carolina at Chapel Hill staff who have training and experience in qualitative research. All interviews were completed in one session and were digitally audio recorded between December 22, 2018 and June 26, 2019. On average, patient interviews lasted 60 minutes and provider and administrator interviews lasted 30 minutes. All participants received a gift-card on completion of the interviews as compensation for their time ($50 per hour with the exception of payers who received a higher rate). The audio-recordings were de-identified and professionally transcribed. Transcripts were not shared with participants for comments or corrections.

Data Analysis

Interview transcripts were loaded into ATLAS.ti v8 (Berlin, Germany) software for analysis. We used template analysis(20), an approach to qualitative analysis that incorporates both inductive and deductive approaches. A codebook with a priori codes consisting of CFIR constructs (and IBM constructs for patient transcripts) was first applied to 20% of the transcripts from each of the three stakeholder groups by two coders (M.C.R. and S.S.). Next, the codebook was revised by adding emergent codes separately from each stakeholder group, creating three final codebooks. Twenty percent of transcripts from each stakeholder group were double-coded (M.C.R. and S.S.) for consistency using the revised codebooks and discrepancies were resolved using discussion. Intercoder agreement of above 85% was established before the final codebooks were applied to all transcripts in the respective stakeholder groups, including those used for consensus-building (M.C.R and/or S.S). Both coders’ analytic files were merged. Themes were reviewed, grouped and organized by the number of participants who discussed each theme at least once during their interview. Thematic saturation was achieved for each major stakeholder group at the level of patients, providers and administrators. Participants did not share feedback on the findings from this study.

Results

Sixty stakeholders participated in this study (66.67% female, 79.63% white/non-Hispanic, median age: 40 [range: 23–72] years) and thematic analysis identified barriers, facilitators and recommendations that cut across CFIR domains and constructs. Table 1 presents the major themes that emerged as barriers and facilitators mapped to the relevant CFIR and IBM constructs, along with the sub-themes and the number of participants from each stakeholder group who discussed each theme. Similarly, Table 2 presents themes that emerged as recommendations for overcoming barriers. Major themes are discussed below, and exemplar quotes by patients, providers and administrators are denoted by PA, PR and AD respectively.

Table 1.

Number of participants who described barriers and facilitators

Major themes	• CFIR Domain (Constructs) / ▴ IBM Constructs	Sub-themes (if any)	Frequency
			PAa	PRa	ADa	Total
BARRIERS
Low awareness about Lynch syndrome	• Characteristics of Individuals (Knowledge and Beliefs about the Intervention)	Patient education on LS and understanding the benefits of genetic testing for LS	6	8	7	21
		Knowledge among non-GC providers (PCPs, specialists)	9	4	10	23
		Provider understanding of role in the process	0	2	0	2
Psychosocial barriers	• Characteristics of Individuals (Other Personal Attributes) • Process (Planning, Engaging, Reflecting and Evaluating) ▴ Attitude	Fear, avoidance	6	6	7	19
		Apathy, invincibility	2	2	0	4
		Guilt, stigma	3	3	4	10
		Privacy concerns	3	3	4	10
Lack of provider follow-up with relatives and probands	• Inner Setting (Structural Characteristics) • Intervention Characteristics (Complexity) • Process (Planning, Engaging, Executing, Reflecting and Evaluating) ▴ Personal Agency		10	8	4	22
Accessibility to genetic testing services	• Inner Setting (Networks and Communications) • Process (Planning, Engaging, Executing, Reflecting and Evaluating) • Outer Setting (Patient Needs and Resources, External Policy and Incentives)	Logistical issues	9	14	5	28
		Genetic counseling workforce shortages	9	8	9	28
Fear or employment and insurance discrimination	• Characteristics of Individuals (Knowledge and Beliefs about the Intervention, Other Personal Attributes) • Outer Setting (External Policy and Incentives) ▴ Attitude ▴ Perceived Norm		16	10	5	31
FACILITATORS
Motivation to engage in cascade testing	• Characteristics of Individuals (Other Personal Attributes) ▴ Attitude ▴ Personal Agency	Duty, concern for family	11	3	2	16
		Life planning	11	0	0	11
		Family support and encouragement	3	0	1	4
Free testing offered by genetic testing labs for relatives of probands	• Inner Setting (Networks and Communications) • Process (Planning, Engaging, Executing, Reflecting and Evaluating) • Outer Setting (Patient Needs and Resources, External Policy and Incentives)		1	12	7	20

^aPA = patient, PR = provider, AD = administrator

Table 2.

Number of participants who described strategies for overcoming barriers

Major themes	• CFIR Domain (Constructs) / ▴ IBM Constructs	Sub-themes (if any)	Frequency
			PAa	PRa	ADa	Total
Improve patient education	• Characteristics of Individuals (Knowledge and Beliefs about the Intervention, Individual Stage of Change) • Process (Planning, Engaging, Executing, Reflecting and Evaluating) ▴ Attitude ▴ Perceived Norm	Education initiatives, trainings and awareness events	7	0	4	11
		Limiting information overload	1	1	2	4
		Explaining benefits of genetic testing	6	2	1	9
		Tailoring to health literacy levels	4	0	3	7
		Clarifying costs and coverage	1	1	2	4
Improve provider education	• Characteristics of Individuals (Knowledge and Beliefs about the Intervention, Individual Stage of Change) • Process (Planning, Engaging, Executing, Reflecting and Evaluating) ▴ Attitude ▴ Perceived Norm		3	6	4	13
Preparing patients to communicate with family members	• Characteristics of Individuals (Self-Efficacy, Individual Stage of Change) • Process (Planning, Engaging, Executing, Reflecting and Evaluating) ▴ Personal Agency	Offering patients materials	10	0	1	11
		Identifying relatives at risk	2	1	0	3
		Assessing patients’ understanding and emotional readiness	2	0	0	2
		Brainstorming approaches for communicating with relatives	3	1	0	4
Implement clinical workflow solutions	Intervention Characteristics (Intervention Source, Complexity, Adaptability) • Process (Planning, Engaging, Executing, Reflecting and Evaluating) • Inner Setting (Readiness for Implementation)	Automating where possible (record release, screening reminders, etc.)	0	5	2	7
		Facilitating provider engagement via technology & EHR integration	5	3	0	8
Optimize staffing resources within clinical practice	Intervention Characteristics (Intervention Source, Complexity, Adaptability) • Process (Planning, Engaging, Executing, Reflecting and Evaluating) • Inner Setting (Readiness for Implementation)	Dedicating resources specific to cascade testing	0	6	2	8
		Care coordination	1	2	1	4
Standardize cascade testing at the national level	Intervention Characteristics (Intervention Source, Complexity, Adaptability) • Outer Setting (External Policy and Incentives)		2	1	0	3

^aPA = patient, PR = provider, AD = administrator

Barriers

Low awareness about Lynch syndrome

Participants across stakeholder groups indicated that the lack of awareness about LS was a barrier for patients for engaging in cascade testing as both probands and relatives usually had low knowledge. According to both providers and patients, the benefits of genetic testing for familial conditions such as LS was not readily perceived by relatives, particularly those who were elderly or in young adulthood. These participants reported that older relatives did not see any use for themselves in confirming a genetic condition, while younger relatives were indifferent to the idea of screening due to feeling invincible at their current stage of life. Participants, particularly patients and administrators, noted that non-genetic providers were largely unaware about LS and related patient needs. Some providers suggested that more clarity on the role of non-genetics providers in the process of cascade testing for LS was needed:

“Who owns that [cascade screening]? Is it the primary care provider of the proband patient? Is [it] the family member’s primary care provider? Is it the specialist if you’ve got, then, something [mutation] that’s identified?”

[PR12]

Psychosocial barriers

Numerous emotional barriers for LS cascade testing emerged among patients. For example, participants across stakeholder groups felt that fear and information avoidance played a significant role in relatives of probands declining genetic testing. Some participants reported that anticipated stigma (sometimes arising from religious or cultural norms) and guilt was a barrier for relatives of patients to consider cascade testing:

“…you know, in our culture, it’s taboo to be sick. You know, we don’t like to advertise the fact that we’re sick. Everybody’s supposed to be healthy and productive and so, a lot of people hide stuff.”

[PA2]

Participants across stakeholder groups also reported that a reluctance to reveal mutation status or discuss LS with relatives due to privacy concerns was another barrier for patients to engage in cascade testing.

Lack of provider follow-up with relatives and probands

Participants across stakeholder groups suggested that provider follow-up with probands and/or relatives was a barrier, as relying solely on patients to communicate genetic information to their relatives was inadequate. Providers noted that several patients were in the difficult position of trying to communicate genetic information to relatives while also grappling with a positive result themselves and/or a cancer diagnosis and treatment. Patients suggested that relatives were more likely to be attentive to information from a medical expert than to a lay person and that provider involvement was needed to reinforce the necessity for genetic evaluation. However, providers and administrators described several deterrents for provider follow-up, including administrative resources for follow-up, priorities during office visits and the lack of clinical workflow protocols to routinely follow-up with probands on whether relatives were notified.

“Well, it takes resources to go back to these folks, and I don’t know whether there’s any way to bill for those services and so, figuring out how to garner resources to allow follow-up would be a huge barrier, actually.”

[PR10]

Health Insurance Portability and Accountability Act (HIPAA) laws on contacting family members or sharing results of probands were also described as a hurdle by participants across stakeholder groups.

Accessibility to genetic counseling services

Participants across stakeholder groups reported several accessibility barriers to genetic counseling services including variability in genetic counseling licensure across states and waiting time to see a genetic counselor in areas with shortages. A lack of practicing professionals in languages other than English and low health literacy were other patient access barriers that emerged. Participants across stakeholder groups also reported that relatives of patients face logistical difficulties in cascade testing including identifying genetics professionals, traveling long distances to academic centers or clinics with genetic services, and taking time off work when services were not available outside of normal work hours. Finally, participants from all three stakeholder groups also voiced concerns about the workforce shortages in the genetic counseling field and care coordination between genetic counselors and other medical specialties:

“Well, I think as we already understand, there is a shortage of genetic counselors and genetics professionals in general, so even once a patient is on board with getting this family variant testing that they personally need, you know, finding someone that they can do that with can be a challenge.”

[AD1]

Fear of employment and insurance discrimination

Participants, particularly patients and providers, suggested that reluctance in pursuing genetic testing was due to concerns about the impact of a positive result on employment and eligibility for health or life insurance. Fears about obtaining employment or career advancement following a positive result, particularly in organizations like the military, were discussed:

“I’ve had some people who, like, they are career army, and they do not want their mutation status to be recorded anywhere within their VA medical records or army medical records because that could impede their progression in their army career…”

[PR2]

Patients reported that uncertainty about the future of coverage for pre-existing medical conditions and concerns about potential changes to the Affordable Care Act (as it relates to coverage of individuals with pre-existing conditions) were also barriers in seeking genetic testing both for probands and their relatives.

Facilitators

Motivation to engage in cascade testing

Motivation to inform their family members was a key facilitator for cascade testing. Participants, particularly patients, felt that motivation was driven by a sense of duty or commitment to family. Patients reported that were especially motivated to inform relatives to facilitate life planning:

“…Family members deserve the most information so that they can make their decisions regarding, you know, family planning and other aspects of their lives. The cascade screening is very important for that.”

[PA6]

Patients and administrators suggested that support and encouragement from family to seek testing, and connecting to patient advocacy organizations, online peer support groups and other resources such as genetic counselor directories were beneficial in motivating patients to get tested.

Free testing offered by genetic labs for relatives of probands

Patients reported that cost or insurance coverage was a barrier for relatives to seek genetic testing. In contrast, other stakeholders emphasized that cost was a perceived (but not actual) barrier, citing free testing services offered by genetics laboratories for relatives of probands within a window of time (45–90 days). Providers and genetic testing laboratories indicated that the advent of these services has significantly boosted the rates of cascade testing in clinical practice. Free testing was perceived by providers and administrators to be a carrot for those individuals whose hesitations were largely around cost and insurance coverage of genetic testing:

“…what really made it much easier because as soon as people hear “free testing,” like, that has really been something that has made it easier to get people to come in within that time window.”

[PR2]

Recommendations

Stakeholders offered several recommendations for improving the process of cascade testing described below.

Improve patient education

Participants, primarily patients, suggested that providers use approaches that mitigate fear in patients by emphasizing the benefits of genetic testing for cancer prevention and early detection. According to stakeholders across groups, ensuring clarity about costs and coverage, and limiting information overload were important aspects of patient-provider education. Patients and administrators also emphasized the importance of tailoring information to the health literacy levels of patients, and discussed other strategies for increasing patient education and promoting screening for LS including community testing events, awareness campaigns and training modules on encouraging relatives to receive testing.

Improve provider education

Participants, particularly providers and administrators, suggested LS education for providers could be improved, especially for primary care physicians and for specialists who encounter patients with LS in clinical practice using internal training initiatives, grand rounds, continuing medical education (CMEs) and office literature.

“The other thing would be to impress upon other providers that do genetic testing the importance of the cascade testing, so ongoing CME - if you will - education with other providers that do genetic testing for hereditary cancer.”

[PR5]

Prepare patients to communicate with family members

Participants, primarily patients and providers, recommended that providers could offer patients materials that could be passed on to relatives to facilitate cascade testing (e.g., information about LS, concrete next steps, contact information for providers, etc.):

“Well, I think if there’s a method; if there’s some type of documents, pamphlets, web pages that I could have used to really strengthen my letter that I sent to them. So, I’m thinking that if there’s something that could equip those of us who are attempting to reach out to support our efforts.”

[PA5]

Patients and providers also suggested that providers could facilitate cascade testing by providing assistance in identifying relatives at risk, assessing patients’ understanding of LS and emotional readiness before patients relay information to relatives, and preparing patients to share information by brainstorming approaches for relatives of varying age, cultural upbringing and degree of closeness.

Implement clinical workflow solutions

Providers and administrators suggested that proband identification could be improved by technology such as embedding a management tool in a hospital electronic record (EHR) system to trigger notifications for screening patients for LS. Providers also suggested that that non-EHR software could be used for creating pedigrees, family history-taking, and follow-up with probands on notifying relatives:

“…it would be really useful if our programs that - like, I use CRA which is a web-based risk analysis pedigree-drawing program - if it had the ability to - based on HIPAA and, of course, PHI and things like that - to sort of automatically, you know, send some type of communication to family members, you know, regarding people who are appropriate for cascade testing.”

[PR5]

Automating elements of the clinical workflow wherever possible to reduce clinician burden was recommended by providers and administrators

Optimize staffing resources within clinical practice

Providers and administrators suggested that dedicating resources specific to cascade testing such as hiring nurse practitioners, patient navigators, or other support staff in large practices would prevent patients from falling through the cracks:

“If I had a team of not only genetic counselors but also admin support, then, yeah - I would love to have a dedicated person and that’s their only job. I mean, that would be perfect ‘cause they could send out letters, contact patients and make sure that relatives are being contacted; figure out how to contact them and coordinate them…”

[PR4]

For specific strategies involving dedicated resource utilization for increasing overall genetic testing rates for LS, these participants recommended offering monthly, single-site testing clinic days in genetics clinics, increasing availability of genetic counseling during weekends and weeknights, and to incorporate a high-risk genetics component in multidisciplinary clinics. A few participants across stakeholder groups also indicated that improving care coordination between different types of providers (genetics and non-genetics) seen by a patient could improve cascade testing rates for LS.

Standardize cascade testing process at the national level:

A few patients and one provider felt that implementing a standardized procedure at the national level would prove most effective at increasing cascade testing for LS:

“…You know, my pie in the sky dream is some sort of a family communication portal that’s standardized for any family with hereditary, you know, whatever to be able to share this information…”

[PR7]

Discussion

Themes that emerged in this study cut across multiple CFIR domains, reflecting the complexity of cascade testing in the US healthcare system and underscoring the need for multi-level interventions to improve this process. Broadly, interviewed stakeholders indicated that gaps remain in cascade testing for LS in the US. Barriers that predominantly map to the individual characteristics domain (patient and provider awareness, impact on employment and insurance) overlapped with barriers reported in the literature for uptake of genetic testing and counseling for cancers more broadly, suggesting that improving acceptance of genetic testing is essential for impacting downstream cascade testing in LS.(21–24) Psychosocial barriers regarding cascade testing were also concordant with barriers reported in the literature on disclosure of genetic information within families.(25–27) Importantly, 50% of the patients in this sample wanted materials to help prepare them for communicating with relatives. Prior research suggests that these materials are often provided as part of genetic counseling in clinical practice(28), and patient-facing resources are also available from a Centers for Disease Control resource webpage.(29) Our results suggest that although such materials are desired by patients, current materials may be insufficient or not easily accessible. Therefore, examining the accessibility of current materials and identifying what additional materials and support would be meaningful to patients should be an area of focus for future research.

Participants, particularly patients and providers, also cited lack of provider follow-up with both proband and relatives as a barrier. This finding is consistent with studies that suggest that greater cascade testing uptake is achieved when trained providers directly contact relatives instead of probands(30,31), indicating that identifying ways to increase provider engagement is central to improving the process. Providers and administrators suggested that staffing resources within a clinical practice could be reorganized and optimized to facilitate cascade testing, suggesting that organizational readiness for implementation (a CFIR inner setting construct) is also necessary to improve cascade testing. Participants across stakeholder groups also perceived that regulations on contacting relatives prevented provider involvement to a large extent, with patients and providers proposing improvements to clinical workflows that centered on supporting probands with cascade testing. Our results also suggest that patients and providers perceive value in leveraging EHR systems for supporting cascade testing.

External policies and incentives (a CFIR outer setting construct) played a key role in the barriers and facilitators identified in this study. Participants across stakeholder groups discussed the influence of federal laws protecting against employment/insurance discrimination(32), including the Genetic Information Nondiscrimination Act, as well as broad health privacy laws, such as HIPAA privacy laws. Thus, a national conversation on specific policies that affect cascade testing in familial cancers like LS such as HIPAA rules on disclosures of familial cancer risk to relatives is warranted.(33) On an organizational policy level, participants discussed genetic testing company policies that provide access to free genetic testing for relatives of probands in their experiences with cascade testing.

This study also revealed a sharp disconnect between patient and provider/administrator perceptions in aspects of cascade testing. For example, only one patient described free testing offered by genetic testing labs for relatives of probands as a facilitator, in contrast to several providers and administrators who described free testing as a key facilitator for cascade screening. Similarly, several patients suggested that offering materials that could be passed on to relatives would be an effective strategy for improving cascade testing, while only one administrator suggested this strategy. Clinical workflow solutions, such as automating health record release, were suggested only by providers/administrators in comparison to zero patients, as may be expected. These results indicate that strategies used by providers/administrators are often based on their perceptions around logistical feasibility and may not always align with patient needs. Thus, exploring this disconnect further in future research, and involving all stakeholders when developing interventions, will be essential in efforts to overcome barriers to cascade testing in LS.

Limitations

The results of this study should be interpreted in light of several limitations. First, patients were recruited through LSI, a prominent patient advocacy organization, and may be more activated with respect to cascade testing and management of LS than general LS patients. Similarly, providers recruited through a professional organization focused on hereditary cancers (CGA-IGC) may have had more interest in cascade testing. Second, we aimed to maximize diversity of stakeholders in this study and as a result have lower subgroup sizes within the provider and administrator stakeholder groups. Thus, saturation of themes was attained not among these subgroups (e.g., primary care providers, gynecologists), but rather was attained among the stakeholder (patient, provider, administrator) levels. As a result, we present cross-cutting themes among these main stakeholder groups and do not draw conclusions on themes specific to subgroups. Third, all but one of the patients in our sample had pursued genetic evaluation for LS, and barriers perceived or experienced by individuals who have elected not to pursue testing may not be captured in our study. Finally, in our semi-structured interview approach, we used prompts on hypothesized barriers and facilitators if participants had trouble organically discussing such factors in the course of their interviews. While this approach allowed us to effectively extract information, acquiescent participants may have endorsed themes that did not play a significant role in hindering or facilitating cascade testing.

In conclusion, cascade testing is a vital tool for improving detection of LS in individuals and initiating preventative measures and surveillance for improving their overall health. This study identified multiple barriers and facilitators for cascade testing in LS and elicited recommendations for improving this process. Policy recommendations from our results, such as implementation of a national LS cascade testing program as well as policies to facilitate other models of non-patient directed cascade testing have been proposed. Indeed the most successful example of cascade testing to date comes from a nationally sponsored, non-patient directed approach used for familial hypercholesterolemia cascade testing in the Netherlands.(34) Variation in genetic privacy laws in the US may make it difficult to implement direct provider communication with relatives; however, findings from our study provide a roadmap for designing interventions that address multilevel barriers as probands communicate with family members about pursuing genetic testing for LS by increasing provider engagement and patient support in this process. Future work should focus on devising and testing multi-level interventions that drive behavioral change in patients, providers and organizations for increasing uptake of LS cascade testing.