Table 4 Methodological best practices and associated considerations for developing a pharmacoeconomic clinical registry

1. Transparency, public availability and dissemination: Enable provision of registry to the public with minimal barriers to access.

• Planning ahead to discuss with funding agency, other stakeholders about making the proposed registry publicly available upon completion

• Should use universally used software to access (e.g. Microsoft Excel)

• Must have clearly defined documentation accompanying the registry to aid in defining terms and usability.

• Promotes greater collaboration, especially among policy-makers in low and middle income countries

• Crowd-sourcing of potential studies for inclusion into the registry

• Identification and correction of potential errors or incorrect information

2. Expert panel: Convene a team of experts in pharmacoeconomics and outcomes research

• Include a team consisting of a trainee along with several PhD-level experts

• Experts should have the time and willingness to aid in the review, data extraction and update meetings

• A team-based approach to curating data for a registry ensures accuracy and allows for provision of continuous feedback

3. Data operationalization: Ensure that the registry is queryable to the extent desired

• Developers of a registry can start by having descriptive variable fields which can then be operationalized as categorical to enhance filtering and querying capabilities

• Variables that are coded as nominal or ordinal categories can aid in data analysis

4. Testing: Provide access to beta version of the registry to various stakeholders at regular intervals to gather additional feedback

• Setting the variables of the registry a priori results in lesser potential to account for future changes and improvements to the registry

• Identify overt transcription errors

• Flexibility and iteration in registry building process allows for customizing registry for the purpose of the individual study