Table 1 Trauma registry implementation 12 step plan

Step 1. Select inclusion criteria.

Trauma registries must consider the population of interest and determine the characteristics of eligible cases. It is not practical for a registry to collect all cases, as the burden and the cost of data collection would very quickly outweigh the benefit of the information gained from trivial or minor injuries.

Step 2. Select exclusion criteria.

It is equally important to select appropriate exclusion criteria, as this will assist in identifying cases of interest and increase the likelihood that only cases that meet criteria are registered on the database. Similar to the inclusion criteria, they must be clearly documented and applied rigorously.

Step 3. Select Minimum Dataset (MDS).

Once the population of interest has been considered and the inclusion and exclusion criteria have been identified, the Registry must select data items that will best capture the required information. Data items are otherwise known as variables, data elements or data points. The items must then be documented in a data dictionary that unequivocally defines each element and provides the rules for use.

Step 4. Identify data sources.

Selected data items must be readily available from the medical record or existing hospital systems. To ensure consistent data collection practices and reproducible data, the source of each item must be identified so that the same piece of information can be obtained from the same place for every case. The source should be documented in operational documents for ongoing reference by Registry staff.

Step 5. Determine method of collection.

Data collection methods depend on the existing systems at the facility; the processes that are already in place and whether it is undertaken in real time during the patient admission or retrospectively after discharge. The use of a paper form is the simplest method of data collection and arguably the most practical for a newly established registry. The hard copy form is usually completed by the data collector according to an agreed process, and then subsequently entered into a database at a later time. The advantage of this method is that paper is portable; data entry can be undertaken at any place and any time and does not need sophisticated technical input. Disadvantages include that it is difficult to store appropriately and the information is more vulnerable to human error due to the longer journey from source to database.

Step 6. Determine data collection personnel.

The personnel who collect data must understand the importance of the task, and prioritise it accordingly. Clinical staff will already have some training and understanding of the information they are required to collect, but may not prioritise the task when they have a competing priority to deliver clinical care to the patient. Non-clinical staff may be able to dedicate themselves to the task more completely, but they will require a greater degree of training to enable them to understand the data properly and perform effectively. The decision on who should collect the data is multifactorial and may depend on the outcome of Steps 4 and 5.

Step 7. Decide on reporting requirements.

The Registry will require a suite of reports that will meet its operational and research needs. These reports will enable monitoring of all aspects of trauma service activity at the facility, as well as provide information that drives research. The reporting functionality of a database should be flexible and be able to be generated routinely as well as on an as needs basis.

Step 8. Determine database functionality.

Registries require a system that will accept, store and report the data securely. They have a choice whether to acquire a product externally or build a product internally. If the system is built internally it requires a Systems Requirement Specification (SRS) that documents the business needs of the registry and the functionality required to meet those needs.

Step 9. Construct the database.

If the facility chooses to build a system internally, the primary reference for the software developers will be the data dictionary, previously referred to in Step 3. The system will be configured to the functionality previously determined in Step 8.

Step 10. Create a training program.

When the components of the registry are in place, the pre-determined data collection personnel must understand the purpose of the registry; be committed to the task of data collection and be trained to enter the data accurately and efficiently. A training program should be adjusted to suit the skill level of the participants.

Step 11. Create training materials.

Appropriate training materials should be available to data collectors, not only for use during the training program, but to be retained as a ready reference at all times. Training and reference materials include but are not limited to a data collection manual that provides clear instructions as to when, where and how the information is to be entered onto either the paper form or into the database; medical dictionaries, anatomical charts and other information appropriate to the skill level of the trainees.

Step 12. Deliver the training to data collectors.

The primary aim of a training program is to ensure that consistent data collection practices are in place and all personnel are cognisant of the agreed process. The format of the training should be appropriate to the environment at the facility and the learning style of the participants. Options for delivery of the training may be in a classroom format to a large group; a focus group format to a small group, or one on one. Once the training has been completed satisfactorily, and management is confident that data collection can commence, the implementation of the registry can be considered to be complete and ongoing operations can begin.