A collaborative approach using remote web-based technology to improve major haemorrhage management and transfusion practice at the University Teaching Hospital, Lusaka, Zambia

To use web-based technology to assess current practice and support improvements in key areas by comparison with previously collected data.

The three areas focused on were:

• Recording of haemoglobin results for surgical patients
• Appropriate transfusion requests
• Tracking of blood products

The findings generated from using this approach to data collection were then used by an on-the-ground team at UTH as an adjunct to providing education to staff in Zambia with a view to improving local practice.

The University Teaching Hospital (UTH) in Lusaka, Zambia is the principle teaching hospital in Zambia providing specialist services to the whole population. Health population statistics can be found on the World Health Organisation (WHO)'s health population profile for Zambia. UTH Zambia is the largest hospital in the country and notionally serves the whole country. This project was particularly concerned with post-natal haemorrhage. With over 90,000 admissions a year, a significant proportion are obstetric and maternity in-patient admissions. Perioperative mortality at UTH is very high and usually avoidable. It is commonly due to systems failures. Lack of availability of blood products is the leading systems cause of perioperative mortality at UTH. For example, in 2014, around 7.2 units of whole blood were collected per 1000 people in Zambia compared with 17 units per 1000 population in South Africa¹.

Transfusion practice mainly involves whole blood not blood components in a low resourced environment so any small improvement makes a significant impact. Often, documentation and systems to improve traceability of components are difficult to implement. The overall aim of this project is to improve the use of blood at UTH by providing evidence to support a locally implemented teaching/education programme by working in conjunction with the Tropical Health Education Trust (THET). This work was seen as a collaborative effort to remotely support an overseas project designed to improve practice locally in an environment where resources are stretched.

The project was supported pro bono by the NHSBT International Development Programme (NHSBTIDP), who designed a bespoke Web-based interface for data entry, and provided data and report feedback in a usable format derived from the information submitted. Data governance issues were discussed and addressed.

Communication was maintained via videoconferencing, telephone and email. Data forms were re-designed in the UK for online data submission. Data was entered via the Web-based interface on a case-by-case basis by local collaborators in UTH, using a secure “associate account”. This account could also be used to download data in a spreadsheet format or to generate reports within the software architecture. These reports could be downloaded for local use and presentation to improve practice.

Limitations to this project included the transfer of data from paper forms to an online system potentially resulting in transcription errors. It was not known if the on-the-ground team working at UTH were able to mitigate this. Despite occasional power outages, the system was reported as easy to use. In a low-resource setting, the ability of the system to analyse and report data immediately was a useful adjunct to the project at UTH, to improve small areas of blood transfusion practice in line with accepted good transfusion principles aligned with NICE guideline NG24.

Summary reports including graphics, were generated immediately and used locally. These reports are automatically updated with the addition of new data. A proportionate analysis (n, %) of question responses was provided by the NHSBT Clinical Audit Manager based in the UK. The findings suggested some key areas where fundamental improvements could be made.. These included high levels of un-traceability of components, principally whole blood and documentation of essential parameters associated with blood transfusion practice. Examples of the interface used are included as a supporting document.

This was a successful collaboration between the UK NHSBT and UTH in terms of a data collection approach. A key learning point was that for projects like this to work, the presence of an on-site team to familiarise the local workforce with online data collection approaches and provide a communication conduit is essential. The efforts of the staff at UTH and THET were fundamental to the success of this data collection approach. The clinical aspects of the findings are to be reported by the staff at UTH and THET and will be added once publically available.

This small project demonstrated that:

• It is possible to remotely support service improvement initiatives overseas with potential direct benefits to the user at minimal cost.
• There is potential for further collaboration including mentoring using such systems.
• Offline data caching should be considered to avoid internet downtime.

Reference:

(1) Michelle S. Chevalier, MD; Matthew Kuehnert, MD; Sridhar V. Basavaraju, MD; Adam Bjork, PhD; John P. Pitman, PhD. Progress Toward Strengthening National Blood Transfusion Services — 14 Countries, 2011–2014. Centres for Disease Control and Prevention, Weekly / February 12, 2016 / 65 (5);115–119