Scottish Cancer Patient Experience Survey 2018: technical report

3. Sample Design

The sampling was carried out by ISD using IBM SPSS software version 24.0 and was designed to:

• provide results at national, Regional Cancer Network, NHS Board and Cancer Centre levels;
• provide feedback relating to recent experiences;
• include people who had experienced the range of cancer care services; and
• minimise the chances of sending the survey to individuals who did not have cancer or who had died since leaving hospital.

ISD used a national dataset containing records of acute hospital activity (SMR01) to identify people aged 16 or over with an inpatient or day case record with any mention of cancer and a discharge date between 1 January 2017 and 30 September 2017.

Where a person had more than one hospital record meeting the above criteria, the most recent hospital episode was selected. This ensured that each person only appeared in the sample once.

In order to confirm the diagnosis of cancer, and to minimise the risk of sending surveys to patients who did not have cancer, validation was completed using the Scottish Cancer Registry (SMR6). Individuals who were identified from the hospital records but did not have a confirmed Scottish Cancer Registry diagnosis date between 1 July 2016 and 31 March 2017 were excluded from the sample.

The following patients were also excluded from the sample:

• Patients who are known to be deceased.
• Privately funded patients receiving care in NHS or private hospitals.
• Patients treated in a private hospital / hospice.
• Scottish NHS patients treated in hospitals outside of Scotland but whose care was commissioned by an NHS Board.
• Patients who were not resident in Scotland.
• Patients who stayed in hospital for termination of pregnancy and / or conditions relating to pregnancy and childbirth.
• Patients who stayed in a hospital maternity unit.
• Patients who were resident in a long-stay hospital.
• Patients who were being treated for a mental health condition in a mental illness hospital.
• Patients who were being treated in a learning disabilities unit.
• Patients with an ICD10 code of C44 and all other C44 classifications (non-melanoma skin cancers).
• Patients with an ICD10 code of C84 and all other C84 classifications (rare haematological cancers with complex clinical definitions).
• Patients with an ICD10 code of D05 and all other D05 classifications (in situ breast cancers).

Addresses were obtained from the CHI database and checked to ensure that they were complete. Any records where the address was incomplete were removed from the sample.

To ensure a sufficiently large response to the survey, a census approach was taken. This means that surveys were sent to all qualifying patients, rather than just a restricted sample of them. A total of 8,302 people were sampled for inclusion in SCPES 2018.

This sampling approach was proposed following a small review and was agreed with the survey’s Stakeholder Group. It is consistent with the sampling methodology used in the 2015 survey, with one additional exclusion – those with an ICD10 code of D05. This change has been made following feedback from the 2015 survey which showed that, due to the variation in clinical definition of these types of tumours, a high proportion of helpline calls regarding cancer diagnosis concerns were from individuals with this type of tumour. This could lead to distress / anxiety for these individuals and so it was agreed that this tumour type should be excluded from future surveys. Using the same sampling methodology between surveys ensures a consistent approach and allows for robust comparisons over time.

Access to Individuals’ Names & Addresses

Approval was given by the Public Benefit and Privacy Panel (PBPP)^[3] to use NHS data to identify a survey sample and approach individuals. A copy of the approval letter is available at www2.gov.scot/Resource/0053/00537065.pdf.

Data transfers containing individuals’ names and addresses were done securely and were limited to variables required to mail the survey pack to the individual and identify which hospital their sample record related to. All data was accessed, managed and stored in accordance with the data confidentiality protocols described in the privacy notice for the survey^[4].