Yorkshire Lung Screening Trial (YLST) pathway navigation study: a protocol for a nested randomised controlled trial to evaluate the effect of a pathway navigation intervention on lung cancer screening uptake

Introduction

Lung cancer is the most common cause of cancer death globally, causing approximately 35 000 deaths per year in the UK.1 The low survival rate (around 85% of people die within 5 years of a diagnosis1) is largely attributed to the majority of people being diagnosed at a late stage when treatment is less likely to be curative.2 Survival rates are lower still for people experiencing socioeconomic deprivation3 yet their incidence rates are highest.4 Globally, trials of lung cancer screening using low-dose CT scanning (LDCT) to detect asymptomatic, early-stage disease, have demonstrated a reduction in lung cancer mortality compared with controls in high-risk populations (adults over 55 with a smoking history).5–11 A meta-analysis of nine trials found LDCT screening was associated with a 16% relative reduction in lung cancer mortality compared with non-screened controls.11 Based on this combined evidence, the UK National Screening Committee has recommended implementation of a nationwide targeted lung cancer screening programme.2 This programme is being rolled out across England via the expansion of the Targeted Lung Health Check programme, which currently offers LDCT lung cancer screening to people at high risk of lung cancer.

While lung cancer screening offers great promise for reducing lung cancer deaths, care must be taken not to widen the existing lung cancer mortality deprivation gap.3 Socioeconomic and smoking-related inequalities in uptake have consistently been observed across trials and services internationally, with uptake significantly lowest among people living in the most deprived areas, and those who currently smoke.12–16 There is also emerging evidence of lower participation by ethnic minority groups, such as those of white ethnicity other than British.16 This means those communities with the highest lung cancer incidence are least likely to be screened, reducing both the equity and effectiveness of screening. The ongoing Yorkshire Lung Screening Trial (YLST), which invites adults with a smoking history aged 55–80 to a telephone-based risk assessment to determine eligibility for a Lung Health Check screening appointment (subsequently offered at a mobile unit), found just 50.8% of those invited took up the offer of the telephone assessment in the first round of screening.17 Crucially, uptake was skewed towards those from more affluent areas, and those people who had quit smoking (compared with those who continue to smoke). This was despite the use of evidence-based interventions from other screening programmes, including general practitioner (GP) endorsement, a low-burden leaflet and reminder invitations.18–23 These figures compare favourably to other UK trials and pilots, which have a 20.4%–52.6% range in response.23–25

Pathway navigation (PN), sometimes referred to as patient navigation, aims to help individuals overcome barriers to early and effective diagnosis and treatment, particularly among populations who experience inequalities in cancer outcomes. It can be split into three phases: (1) navigation to screening, (2) diagnostic evaluation and (3) treatment.26 Research in the USA suggests navigation programmes are effective at improving participation in different types of cancer screening27 and an American Thoracic Society statement28 concluded navigation should be integrated within lung cancer screening programmes as a strategy to reduce disparities in uptake. In a randomised trial (n=1200) comparing navigators with usual care among a lower socioeconomic population, uptake was significantly improved in the navigator arm (23.5%) relative to usual care (8.6%).26

To date, there have been no published studies specifically testing PN for individuals invited to lung cancer screening in the UK. One small feasibility study tested whether navigation by specialist screening practitioners could improve the uptake of bowel scope screening.29 However, in that context, it was not feasible to successfully implement the intervention as telephone numbers were not available from health records and participants had to explicitly consent to take part in the study before providing their telephone number to receive navigation. The authors reflected this resulted in a selection bias, whereby only participants engaged in the screening process (and arguably those least likely to require navigation) would consent to take part. However, many cancer alliances in the UK (which do not require consent prior to patient contact) have demonstrated the feasibility of using telephone-based interventions to improve the uptake of cancer screening where numbers are available. For example, a case study from West London RM Partners Cancer Alliance reports successful contact of 13 000 individuals by telephone of which 25% subsequently participated in bowel cancer screening.30

While the potential of PN to reach high-risk non-responder populations is promising, adoption of this type of intervention within a national screening programme in the UK would involve considerable resources. It is, therefore, vital that its effect is analysed within the UK health system to demonstrate efficacy and cost-effectiveness. Likewise, it is crucial to understand any barriers to intervention efficacy that may lay in implementation and service design. The YLST recently received funding for a third round of screening to continue the provision of biennial Lung Health Checks. This offers an opportunity to further develop an understanding of barriers to participation and to test if PN can maximise engagement in a lung cancer screening programme in the UK context.

Therefore, the aim of the study is to determine whether PN results in more individuals participating in the telephone risk assessment and, if eligible, the LDCT screening scan as part of their Lung Health Check, compared with usual written invitation within a previous non-responder population. In addition, a mixed-methods process evaluation with people who have repeatedly not responded to invitation will provide an avenue to understand issues preventing individuals from considering or being able to take part in LDCT lung cancer screening, as well as how PN might work, its acceptability and how it could be protocolised at scale, to inform screening invitation and delivery.

Methods and analysis

Study design

A two-arm randomised controlled trial (RCT) with mixed-methods process evaluation nested within the YLST. The methods of the YLST are published elsewhere.18 The study flow is described in figure 1. The protocol is reported in line with the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) statement (online supplemental file 1).31

Supplemental material

Figure 1
Figure 1

YLST pathway navigation study flow chart. COM-B, capability opportunity motivation-behaviour; GP, general practitioner; LDCT, low-dose CT; LTH, Leeds Teaching Hospitals; YLST, Yorkshire Lung Screening Trial.

Outcomes

For the nested RCT, the outcomes are stated below. These outcomes will also be reported for the process evaluation.

Coprimary outcomes

  1. The number undergoing initial telephone assessment of lung cancer risk.

  2. The number undergoing a LDCT screening scan.

Secondary outcomes

  • Demographic, clinical and risk parameters of people undergoing telephone assessment of lung cancer risk.

  • The number undergoing telephone assessment of lung cancer risk who were found eligible for screening (irrespective of whether they subsequently attended for LDCT).

  • The number of screen-detected lung cancers diagnosed.

  • Costs of delivering the PN intervention.

  • Quantitative and qualitative process evaluation (assessed through observations of strategies and barriers; acceptability; intervention fidelity; and semistructured interviews).

Demographic, clinical, risk and lung cancer outcomes will also be compared with the original YLST baseline responder group, and the group that responded following written reinvitation during the second round of screening in YLST.

Participants

Repeat non-responders (nested RCT and process evaluation participants): Participants for the RCT will be those who have not responded to a postal invitation to participate in a telephone assessment of lung cancer risk (Lung Health Check eligibility) within the YLST, either as part of the initial invitation round or after being reinvited approximately 2 years later.

Disengaged responders (process evaluation participants): Participants in this group will not be randomised within the nested RCT but will all receive PN and be recruited to the process evaluation. This group is made up of YLST participants who have responded to at least one of the first two rounds of Lung Health Check invitations, undergone the telephone triage and were found to be high risk, but did not attend for their most recent LDCT screening scan. This includes people first found to be eligible through telephone assessment of lung cancer risk during:

  1. The first screening round, those who attended their first scan, remained eligible at their subsequent eligibility reassessment during the second round but did not attend for their second LDCT screening scan appointment.

  2. The second screening round but did not attend their LDCT screening scan appointment.

The group does not include people who were risk assessed during the first screening round, found to be eligible, but did not attend their first scan (since they were already recontacted during the second round).

Eligibility criteria

For the nested RCT, the full eligibility criteria are stated below. Eligibility criteria for the process evaluation are as described above, including both repeat non-responders and disengaged responders.

Inclusion criteria

  1. Aged 55–80 (inclusive) at the time of the planned screening block (all non-responding participants will be assigned an intended screening block between March 2023 and October 2024; participant age will be calculated on the first day of screening during that block to determine eligibility).

  2. Assigned to the intervention group (lung cancer screening arm) of YLST.

  3. No previous YLST telephone triage assessment.

Exclusion criteria

  1. Previous registration of dissent from data usage or further contact from YLST team.

  2. National data opt-out was recorded prior to the screening round.

  3. Lung cancer diagnosis within 5 years prior to the screening round.

  4. Any metastatic cancer diagnosis.

  5. No telephone number on Leeds Teaching Hospitals Electronic Patient Record.

Randomisation

Repeat non-responders will be randomised to PN or usual care (control) by household so that cohabiting individuals are assigned to the same arm of the study to avoid contamination. Randomisation will take place after completion of the second round of screening but before reinvitation to the third round of screening. All disengaged responders will receive PN (ie, no randomisation) and will not be randomised as part of the nested RCT, to avoid contaminating the sample (since disengaged responders represent a different subpopulation to repeat non-responders).

Sample size

Nested RCT

The sample size for the RCT is calculated based on the number of people who have not responded to the invitation during the first screening round of YLST. Of the 22 128 people who did not respond during the baseline round, 20 074 people were aged ≤76 years at the time of original GP data extraction (and thus would be aged ≤80 years at the time of proposed PN randomisation) and remained alive at the time of designing the PN study. After accounting for an anticipated 10% response rate to written invitation during the second round, we estimated 18 067 potentially eligible participants for PN randomisation, of which 74.1% had contact numbers recorded, leaving a total eligible population of 13 388.

An 8% response rate to written invitations is anticipated in the control arm, increasing to 12% in the PN arm, based on previous randomised navigator trials in lower socioeconomic populations.26 Baseline data from YLST indicate approximately 35% of telephone responders were deemed high risk. Assuming a similar rate in the repeat non-responders in this nested PN substudy and further, considering that only 84% of those assessed as high risk had an LDCT scan (as was observed during the first round of screening in YLST), would infer an increase from 2.4% (standard invitation arm) to 3.5% (PN arm) of those invited responding, being high risk and receiving an LDCT scan. Based on these assumptions, and with an anticipated 13 388 participants, as can be seen from the sample size calculation provided in table 1, there would be ample power (>90%) to detect an increase in both response with telephone assessment and attendance in terms of undergoing an LDCT scan using two-sided tests at the 2.5% significance level to account for coprimary outcomes and using a continuity correction, and adjusting for centre effects by household as appropriate.

Table 1

Sample size considerations for coprimary outcomes in the nested RCT

Process evaluation

The sample size for each component of the process evaluation is as follows:

  • Quantitative evaluation of barriers and strategies to undergoing a risk assessment and/or attending a screening appointment discussed during PN calls. These will be recorded using a standardised observation proforma (online supplemental file 2) after every PN telephone call (up to five per participant) and after every telephone triage telephone appointment conducted within the control arm.

  • Semistructured interviews (n<60): Purposive sampling will be used to recruit a diverse sample of participants who have undergone the PN intervention with respect to demographic characteristics, smoking status and intention to attend a lung screening appointment. Up to 30 participants will be interviewed within each group (ie, non-responders and disengaged), in line with norms for qualitative research, and to ensure adequate representation of those subsequently attending and declining a screening appointment.

  • Fidelity observations: A subset of 200 PN calls will be observed. The number has been selected to ensure variability in the calls being observed (with regard to screening day, team member delivering the call, etc) while being possible to conduct within the time frame and in line with other similar studies of intervention fidelity.32

Supplemental material

Intervention

Development

A theory-and-evidence-based intervention development process was followed.33 The navigation-to-screening manual was informed by a bowel screening navigation intervention.34 Strategies and barriers were based on existing research investigating factors influencing lung cancer screening uptake15 35–38 and through discussion with YLST trial practitioners and patient and public involvement (PPI) representatives. Theoretically, the intervention is informed by the Capability Opportunity Motivation-Behaviour (COM-B) model, which is the central behaviour system of the behaviour change wheel.39

PN intervention protocol

Component 1: advanced notification

Participants receive an advanced notification letter (online supplemental file 3) ~3 weeks before a scheduled telephone appointment. The notification frames the appointment as an ‘Introduction to Lung Health Checks’ telephone call. It includes a low-burden leaflet (online supplemental file 4) with basic information about the Lung Health Check (similar to that sent in the control arm, but with instructions to call the team to book a risk-assessment removed) and an A5 flyer with dissent information.

Supplemental material

Supplemental material

Component 2: protocolised ‘Introduction to Lung Health Checks’ telephone call

At the prearranged time, a pathway navigator telephones the participant to introduce and discuss the lung screening offer. If contact is unsuccessful, the pathway navigator attempts two more telephone calls at different times on varied days.

The PN team (pathway navigators) comprise trained senior clinical trials assistants, administrative officers and research nurses. Figure 2 shows the phased telephone approach as protocolised in the PN training manual (online supplemental file 5). There are four key stages, with step 3 (strategies and barriers discussion) taking place flexibly throughout the call:

Supplemental material

  1. Introducing the offer of lung screening (Lung Health Check).

  2. Assessing eligibility for LDCT lung cancer screening.

  3. Eliciting and addressing any barriers to capability, opportunity and motivation for attending screening using core communication competencies, problem-solving, behaviour change techniques and practical arrangements.

  4. If eligible, arranging either a Lung Health Check appointment, another navigation telephone conversation (to offer further support/deliver the intended strategy as per protocol), or respecting the individual’s choice not to engage.

Figure 2
Figure 2

Phased pathway navigation ‘Introduction to NHS Lung Health Checks’ telephone appointment. NHS, National Health Service.

Component 3: reminder telephone calls

If the participant books a screening appointment, they will also receive reminder calls approximately 1 week and 1 day before the booked appointment.

PN intervention training

All pathway navigators receive the protocolised training manual (online supplemental file 5). This covers the structure of the telephone call; principles of informed choice and types of non-responder (informed, misinformed, disengaged, inclined); common barriers and suggested solutions for navigators to offer for individuals to overcome them; and behavioural science-informed communication techniques (motivational interviewing, use of simple language, implementation intentions and teach-back). All pathway navigators attend a training course in motivational interviewing,40 and a 1-day interactive session led by behavioural scientists on the contents of the training manual before delivering the intervention. Check-ins are arranged 1 month after beginning the calls, and then every 3–6 months for feedback on experiences and to exchange knowledge.

Control arm (usual care)

The control arm will receive the same invitation materials used to reinvite non-responders 2 years after the initial screening round of YLST (a GP-endorsed invitation letter, a low-burden reinvitation information leaflet, a dissent information statement and a follow-up reminder letter if they do not respond to the initial reinvitation).18 People without an available telephone number (who are therefore excluded from the PN arm) would be reinvited using the same method but not included in the analysis for this substudy, as there may be systematic bias in the characteristics of those who do not have a telephone number recorded. The letters invite the individual to telephone the YLST team to undergo a telephone-based risk assessment for a Lung Health Check appointment where LDCT lung cancer screening is offered.

Ethics and dissemination

REC and other regulatory review

The YLST was approved by the Greater Manchester West Research Ethics Committee (REC) (18-NW-0012) and the Health Research Authority following review by the Confidentiality Advisory Group (CAG). The PN substudy was approved by the REC and CAG as a Substantial Amendment on 5 October 2022.

Trial monitoring and governance

Trial governance is overseen by the trial management group, the independent data monitoring committee and an independent trials steering committee. Full details are provided in the YLST protocol.18

Informed consent

The YLST and nested PN study follow a service demonstration design to provide a realistic indication of uptake following PN in a real-world clinical context, improve ecological validity and avoid the Hawthorne effect,43 following a precedent set by earlier lung cancer screening trials.23 This means participants will be unaware of the research nature of the PN intervention or YLST until they attend for a Lung Health Check screening appointment, where fully informed consent will be collected to participate in the YLST. Individual consent will, therefore, not be sought before randomisation to the PN study. However, the opportunity and mechanism for dissent will be clearly described in the advance notification letter sent with details of the scheduled telephone appointment so individuals can opt-out if they do not agree to the use of this information. The opportunity to dissent from the wider YLST was also provided with all previous invitations. This approach was REC and CAG approved, and deemed acceptable by PPI representatives.

Patient and public involvement

A PPI consultation exercise was undertaken with 12 individuals (5 male/7 female) aged between 50 and 80 years, including individuals with a smoking history who had lived experience of cancer, including lung cancer (n=7), as well as those without (n=5), to gain insight from a broad range of perspectives. The feedback was used to inform the development of the PN intervention, data collection tools and to understand the acceptability of the observational methods used for the process evaluation. All 12 individuals were supportive of the research.

Dissemination

Study findings will be submitted for publication to relevant peer-reviewed journals in accordance with the Consolidated Standards of Reporting Trials (CONSORT) Statement44 and presented at conferences. A summary of the results will be provided for participants on the study website.

This post was originally published on https://bmjopen.bmj.com