Characteristics of Back Pain Attendances at ED

Purpose and Design

Back Pain is a common cause of pain and disability. The National Institute for Clinical Excellent [NICE] (2016) identify back pain as affecting 1 in 10 people and causing more disability than any other condition. In most people back pain resolves within a number of weeks without any healthcare input. For those whose pain does not improve physical and psychological treatments are recommended. These are provided by specialist services and are not available in Emergency Departments or via Ambulance Services (collectively referred to as Emergency Medical Services [EMS]). Thus, arguably only a very small minority of patients with back pain, where a severe medical pathology is indicated (e.g. cauda equina syndrome) should need to access EMS. However, anecdotally many people with back pain self-refer to EMS. The prevalence of people self-referring and their characteristics are currently unknown as these figures are not routinely reported although the information exists within the data collected by many trusts.

This study aims to analyse anonymised data from the Darlington Memorial Hospital (DMH) Emergency Department (ED) to establish the prevalence of those attending with back pain. The characteristics of people presenting over a 12-month period with back pain, along with characteristics of their contact (e.g.

administration of analgesia and the use of diagnostic imaging) will be described and compared to the general characteristics of those attending ED.

These characteristics may help to inform healthcare service design, both EMS and non-EMS, to better meet the needs of this patient group.

Research Methods

This study will be a secondary analysis of an existing NHS database. The study will request anonymised data from DMH ED. The data will cover all adults who have accessed their services in a 12-month period. The information requested will include data related to the attendance, patient demographics/characteristics, geographical location and clinical data.

Recruitment

The data being used for this study is from a currently existing Trust database and therefore there will be no recruitment of participants.

Inclusion/exclusion All individuals over the age of 16 will be included. Those aged under 16 will be excluded as they are not included in clinical guidelines.

Risks, Burdens and Benefits

The data has been collected for clinical care and not for research purposes. Therefore patients will not have explicitly consented for their information to be used for research. The Information Governance Officer at DMH has confirmed that the data will be in non-personally identifiable form when it is released to the research team and is compliant with Section 33 of the Data Protection Act 1998.

The main risk to patients is that their clinical data is being used for research purposes. However identifiable data will only be available to the usual clinical team. Anonymised information will be extracted by a DMH member of staff before it is securely transferred to the team at Teesside University where it will be analysed. Names and addresses will be completely removed from the database before it leaves the NHS.

Some publications require raw data to be supplied. This would be supplied but it will be a non-personally identifiable dataset.

Confidentiality

The Caldicott Principles will be followed:

1. Justify the purpose(s) of obtaining the information
2. Don't use person-identifiable information unless it is absolutely necessary
3. Use the minimum necessary person-identifiable information
4. Access to person-identifiable information should be on a strict need-to-know basis
5. Everyone with access to person-identifiable information should be aware of their responsibilities
6. Understand and comply with the law

Problems arising from the processing of identifiable data and how they will be handled

The database contains socioeconomic data in the form of patients' postcodes. This socioeconomic data is important as it has been linked to emergency medical service use. A postcode is identifiable data. To overcome this DMH will supply only the first four digits which covers a sufficiently large geographical area that this will no longer be identifiable data. The first four digits will be sufficient to identify a Lower Layer Super Output Area code (LSOA code) which is the equivalent index of socioeconomic status.

As the dataset will include all adults attending ED it will inevitably include members of vulnerable groups or others who would usually require additional ethical approval for inclusion in research (for example offenders) however this will not be known to the research team or be apparent in the non-personally identifiable data requested.