Patient Experience survey of GP and local nhs services 2011/12 Volume 1: Technical Report

5 Fieldwork

5.1 The Scottish Government contracted Picker Institute Europe⁴ to administer the survey. Picker Institute Europe is an independent, not for profit research and development institute with charitable status, and has provided support for other patient experience survey work elsewhere in the UK. ISD Scotland provided day-to-day support for the administration of the survey along with Scottish Government staff.

Mail-out

5.2 The fieldwork began on 04 November 2011. In total, 605,896 surveys were posted out in four batches between 04 November 2011 and 17 November 2011. Patients were asked to complete the questionnaire by 31 January 2012. Reminder letters were not sent out to those who did not initially respond. For the previous survey two reminders were sent to boost the response rate, but this year's survey was designed to achieve the required number of responses for each practice without reminders. It proved to be cost effective to send more surveys initially than to send fewer surveys and reminders.

Data collection

5.3 Data was collected in the form of hardcopy returns and online returns (including online returns completed by people using JAWS readers⁵). For more information on the format of returns refer to section 8.8 of this report.

5.4 During the fieldwork a freephone helpline answered queries from patients surveyed. In total, 821 telephone enquiry calls were answered by the telephone and language line. This was a noticeable reduction compared to the number received in 2009/10 (3,668). A possible reason for this decrease could be attributed to better supporting documentation and instructions being sent with the questionnaire. Another possible explanation could be that patient experience surveys are now fairly common and people may simply be more familiar with completing them.

Deceased patients

5.5 The patients included in the 2011/12 survey were sampled from an extract taken from the CHI database on 17 October 2011. The questionnaire printing and mail-out process extended from this date through to the final mailing date of 17 November 2011. This meant that some patients would have died between the extract and mail out dates.

5.6 As in previous surveys, the Scottish Government were keen that every possible effort was made to avoid questionnaires being sent to family members of deceased patients. Therefore, similar to previous surveys, a list of patients included in the initial sample was linked to the Community Health Index (CHI) database to identify patients who had recently died. This process also helped to identify patients who had died abroad and was repeated three times during the period between the initial sample being drawn and the surveys being mailed out. We are grateful to Practitioner Services Division (PSD) of NHS National Services Scotland and Atos Origin Alliance (who host the CHI database for NHS Scotland) for their support during this phase of the work.

5.7 This year, an additional process was introduced to further minimise the risk of questionnaires being sent to deceased patients. A list of patients included in the survey was sent to the NHS Central Register (NHSCR) and linked to the National Records of Scotland (NRS) deaths database. This allowed NHSCR to provide daily extracts of deceased patients to ISD. This process also helped to identify patients who had died elsewhere in the UK. Having access to daily death information greatly reduced the number of questionnaires being sent to addresses of deceased patients. We are grateful to NHSCR for their help and support during this stage of the project.

5.8 Between the CHI and NHSCR databases, a total of 659 records included in the initial survey sample were flagged as deceased between the time the 17 October 2011 CHI extract was taken and the final mail out date. These records were analysed by ISD and sent to Ciconi Ltd for removal from the mail-out process prior to dispatch. Of the 659 deceased records, a total of 459 questionnaires (70%) were suppressed and removed from the postal mail out. Analysis of those records not suppressed showed that a quarter of patients had died on the day or the day after the mail out. The analysis also highlighted that majority of the remaining cases died within two weeks of the mail out.

5.9 Any death which occurs in Scotland must be registered within eight days of the date of death. This means that there can be a delay between the actual date of death and the date that it is registered and updated on the CHI and NRS databases.

5.10 This delay, combined with the volume of the mail out process, made it extremely difficult to prevent all questionnaires being sent to addresses of deceased patients. However, as outlined above, efforts were made to avoid this as much as possible.