DATA COLLECTION

GI therapeutic endoscopies

All deaths occurring in hospital between 1 April 2002 and 31 March 2003 were reported
to NCEPOD by designated local reporters for each hospital. Data were requested from
all hospitals in England, Wales, Northern Ireland, Guernsey, the Isle of Man, the Defence Secondary Care Agency and hospitals in the independent sector.

Sample cases were identified from these data by Office of Population Censuses and Surveys (OPCS) codes, which were submitted for the last six procedures before death. Cases were included if death occurred within 30 days of a therapeutic endoscopy, regardless of whether it was the last procedure or not. If more than one endoscopy was recorded in the death data, only the last procedure before death was included.
The following OPCS codes¹ were included in the sample:

The terminology used to describe procedures in this report is based on these OPCS codes.

Procedures coded as one of the above but carried out under radiological guidance, without endoscopy, were excluded. Likewise, procedures carried out at a hospital different to the one where the patient died were also excluded due to the complexity of linking patient data.

Data were collected retrospectively via a questionnaire (Appendix C), which requested information on pre-procedural investigations, the procedure, sedation and monitoring, the clinicians involved in the procedure, and training and audit. A questionnaire with a unique NCEPOD number and covering letter was sent to the clinician responsible for the patient
at the time of their death. In order to reduce the burden on the clinicians, the number of questionnaires to be completed was limited to three per clinician. However, if a clinician returned more than three, the questionnaires were included in the study. A first reminder for return of questionnaires was sent two months after the questionnaire, followed by a second reminder one month later.

Copies of extracts of the casenotes, including the endoscopy report, monitoring charts and autopsy report were also requested. Both these and the questionnaires were anonymised by NCEPOD administrative staff to ensure that individual patients, clinicians and hospitals could not be identified.

NCEPOD wrote to all general practitioners from a list provided by the British Society of Gastroenterologists (BSG), and asked them if they performed therapeutic GI endoscopies. Those who reported that they carried out these procedures were requested to notify NCEPOD in the event of death within 30 days of a procedure.

Upper GI dilation and tubal prosthesis insertion (back to top)

A second dataset was collected prospectively on all patients, of 16 years of age and
over, undergoing an upper GI dilation or tubal prosthesis insertion between 1 January and
31 March 2003, regardless of outcome. NCEPOD local reporters were asked for a list of patients, from their information department, who had had one of the following eight procedures, identified by the OPCS codes¹ :

The terminology used in this report is based on these OPCS codes. The term dilation
and dilatation are used synonymously.

Questionnaires (Appendix C), which consisted of two sides of an A4 sheet, were sent prospectively to NCEPOD local reporters for dissemination to the consultant responsible
for the procedure. No patient identifiers were collected on the questionnaire and pre-paid envelopes were provided so that questionnaires could be returned directly to NCEPOD; this ensured both patient and clinician confidentiality. No casenote extracts were requested.

General practitioners who performed upper GI dilations and tubal prosthesis insertions
were identified, as before, and asked to notify NCEPOD when they performed a procedure included in the study.

Organisational questionnaire (back to top)

An organisational questionnaire (Appendix C) requesting information about the endoscopy suite and organisational aspects of the endoscopy service was sent to the NCEPOD local reporter of each hospital for completion. The questions were based on the guidelines from
the Working Party of the BSG Endoscopy Committee 2001².