Targeted encouragement of GP consultations for possible cancer symptoms: a randomised controlled trial

Study design and setting

A two-arm randomised controlled trial (RCT) — called Writing to Encourage Late Consultation Outpatients to Make Engagement with their GP (WELCOME-GP) — was carried out to test the effectiveness of a postal intervention that was targeted at certain vulnerable groups and intended to promote awareness of cancer symptoms and earlier help seeking at general practice. It was designed to:

• promote awareness of six cancer ‘red-flag’ symptoms; and

• allay fears of wasting the doctor’s time.

The intervention was targeted at a population that had not attended their general practice in the previous 12 months, with factors that suggest this may be incongruous with their clinical need and attributes known to be risk factors for late-stage presentation of cancer.^13–17 A total of 24 practices from three areas in England — rural and urban areas of the North East, Greater Manchester, and Greater London — agreed to take part; one practice subsequently withdrew.

Sample and randomisation

Patients were eligible for inclusion if they:

• were aged 50–84 years;

• were registered with a participating practice;

• had not had a consultation at their registered practice in the previous 12 months; and satisfied at least two of the following:

  • — lower socioeconomic status (in either of the two lowest quintiles of the Index of Multiple Deprivation 2015, based on post code);¹⁸

  • — missed last scheduled screening for breast cancer, bowel cancer, cervical cancer, or abdominal aortic aneurism;

  • — history of use of emergency or out-of-hours (OOH) services instead of primary care;

  • — missed last appointment for chronic-disease monitoring/management;

  • — lived alone (indexed by being the only person registered with the practice at their address) as a marker of social isolation; and

  • — smoker (ever).

Eligibility was assessed from patients’ primary care electronic health records. It was not possible to identify, with confidence, patients who had used emergency or OOH services from the practice databases so no patients were recruited on this basis.

Persons were ineligible if:

• they already had a diagnosis of cancer;

• the GP considered recruitment inappropriate in view of the patient’s state of health or cognitive capacity; or

• the GP felt that the patient would not wish to be included in the research.

Sample-size calculations are described in Supplementary Appendix S1. Each practice was provided with a computer-generated randomisation schedule. Practice staff identified eligible patients and randomly allocated them to the intervention or to usual care; there was no concealment of allocation. At the end of the 6-month observation period, staff at each practice reported the following for each randomised patient:

• the number of consultations by each patient who continued to be registered throughout the period;

• the month of the consultation(s);

• the reason for the consultation(s);

• whether any investigations or physical examinations were arranged and completed; and

• whether referrals to secondary care were made.

De-identified data were stored on encrypted databases that remained at practices. Intention-to-treat (ITT) analyses were carried out by the data analysis team on anonymised data for all eligible patients for whom data were available at the end of the observation period. Due to the nature of the intervention, it was necessary that patients in both groups were unaware that they were taking part in a research study; however, it would have been impractical for practice staff and the data analysis team to operate on this basis so they were not blinded to the allocation. Censored, anonymised data output was generated from the encrypted database at each practice and securely transmitted to the research team.

Intervention

The Model of Pathways to Treatment provides a framework for understanding the pathways between an individual first detecting a bodily change and subsequently undergoing treatment,¹⁹ beginning with ‘appraisal’ and ‘help seeking’ intervals. Crucially, the individual must believe there to be a reason to present with their symptom; this is often motivated by heuristics, such as concern about a worsening symptom or its interference with their daily activities.²⁰

The present intervention can be contextualised within these intervals, in that it was intended to:

• prompt appraisal through symptom awareness;

• provide individuals with a ‘cue to action’ for visiting their GP; and

• use messaging to counteract known attitudinal barriers to help seeking.

Furthermore, the personalised, GP-led approach drew on the success of primary care endorsement in the cancer-screening context.^21,22

The intervention consisted of a letter (see Supplementary Appendix S2), signed by one of the GPs at the patient’s practice, noting that the recipient had not been seen at the practice for some time and reassuring them that consulting with any symptoms would not be considered wasting the doctor’s time. The letter also drew attention to an enclosed leaflet (see Supplementary Appendix S3), which detailed six symptoms that are considered to potentially raise suspicion for cancer:

• blood in urine;

• blood in stool;

• persistent cough;

• haemoptysis;

• difficulty swallowing; and

• unexplained weight loss.

These symptoms were chosen as they feature in the National Institute for Health and Care Excellence’s recommendations for referral for suspected cancer.²³ The first practice mailed the intervention materials to all patients allocated to the intervention arm on 11 November 2016, and the last did so on 31 May 2017.

Outcomes

The primary endpoint was the equivalent annual rate of consultations with clinicians from the patient’s registered practice in the 6 months following randomisation (that is, the date that letters were mailed), whether the consultations were at the practice itself, home visits, or telephone consultations. This rate was calculated as the total number of consultations divided by the total period of exposure (half a year per patient; only those patients registered at the practice for the full 6 months were included in the analysis). Secondary endpoints were:

• total general practice activity — that is, consultations, referrals to secondary care, completed investigations (for example, blood tests) arranged by the practice, physical examinations carried out at the practice (for example, blood pressure check and digital rectal examination), and any diagnoses; and

• the use of emergency and OOH services in the study and control groups.

As noted above, it could not be established with confidence whether patients had made use of emergency services, so this particular analysis was not possible. It also became clear during the course of the study that it would not be possible to collect data on subsequent diagnoses, as these may have occurred some time after the end of the 6-month follow-up. Several other analyses were carried out: the primary and secondary analyses were repeated, considering only those consultations that had taken place for at least one of the six symptoms identified in the leaflet; the total number of consultations that took place in the same calendar month as randomisation (the period immediately following the mailing of the intervention) were compared between the intervention and control groups.

Statistical analyses

All randomised patients for whom follow-up data were available were included in the ITT analyses. Data were analysed by zero-inflated Poisson regression,²⁴ a technique that takes account of the large number of persons with no consultations at all during the 6-month observation period. Relative risks (RRs) of consultation with 95% confidence intervals (CIs) were calculated. Robust variance estimators were used, allowing for clustering by person. Stata (version 15.1) was used for data analysis.