关于临床试验中的 CRFs，您需要知道的一切

Whether you want to expand your knowledge of CRFs in clinical trials, or learn the dos and don’ts of CRF design, you’ve come to the right place! Let’s start with the basics.

What are case report forms?

A case report form (CRF) is a document designed to record all patient information that needs to be collected during a clinical drug trial.

An electronic CRF (eCRF) can be stored in cloud-based systems such as clinical metadata repositories (CMDRs) or Electronic Data Capture (EDC) systems. The terms ‘CRF’ and ‘eCRF’ tend to be used interchangeably, and they’re also referred to simply as ‘forms.’

CRFs play a crucial role in helping to assess the safety and efficacy of clinical products. For a study to be successful, the data collected must be correct and complete. This means that forms must be well planned with meticulous attention to detail. They must comply with the clinical study protocol and record its details. They must also comply with regulatory requirements, such as those defined by the US Food and Drug Administration (FDA).

What are annotated CRFs?

Annotated CRFs, or aCRFs, are a key submission deliverable and a mandatory requirement of the FDA. Each form in a study contains markings or annotations. These annotations map data points on forms to the name of datasets, and variables within those datasets. In other words, “each CRF should provide the variable names and coding for each CRF item included in the data tabulation datasets”, as stated by the FDA guidelines.

Note

The FDA stipulates that a set of blank annotated CRFs be submitted in a PDF document called “blankcrf.pdf”. This document helps the FDA reviewer find the origin of variables in the Study Data Tabulation Model (SDTM) datasets.

Why should CRFs and aCRFs be standardized?

First, what do we mean by standardized CRFs?

Standardization is the process of developing content to meet specific regulatory and organizational standards. Standardized forms will conform with regulatory requirements i.e. be laid out and annotated as required by CDISC’s Clinical Data Acquisition Standards Harmonization (CDASH) standard. They will also conform with the requirements of different internal stakeholders within the study team, as they are given the chance to review and approve the forms for use.

There are many benefits to standardizing your CRFs:

• Standardization means creating forms that cover everything your stakeholders need; investigator, data manager, biostatisticians and data entry personnel can all review and approve the standardized form.
• This means standardized forms can be used repeatedly. This saves time on both design and approval in future studies.
• Standardized forms are compliant with regulatory requirements, so elements such as annotations don’t need to be manually done on each form in the study, saving time and effort.

We’ve analyzed three common misconceptions below:

Myth 1: It’s more cost-effective to design CRFs in text documents.
Fact 1: Designing and storing eCRFs in cloud-based systems can be much more cost effective than other methods.

It actually costs more to continue using text documents or spreadsheets, rather than switching to a CMDR/EDC set up. The most up to date versions of forms and annotations need to be located, edited, approved, or re-created from scratch each time. There’s also a higher likelihood of data errors with this method because formats can’t be enforced so easily. Getting CRFs ready for use can be a time-consuming process, and therefore costly.

In comparison, standardized, approved eCRFs stored in a cloud-based system are easily located and ready for reuse, which creates efficiencies; they’ve already been reviewed and approved by stakeholders, and you know that they capture clear and valid data.

Myth 2: Data stored in text documents or spreadsheets is safer.
Fact 2: Creating and maintaining CRFs using text documents and spreadsheets can be tedious and error-prone. Content in text documents can become corrupted or unintentionally altered by users. Locating the latest version of a CRF that has been approved by all stakeholders can be tricky.

On the other hand, EDCs use encryption to secure data both in transit and when stored, in addition to off-site encrypted backups. With EDCs, access is restricted, and there’s a full audit trail for traceability.

Myth 3: Making the move to using eCRFs and EDCs requires a lot of software training.
Fact 3: There’s less training required than you think. Using a cloud-based system that automates the process of creating eCRFs saves lots of time. Some training is required to get started, but the time saved in the long term makes this method far more efficient.

Using eCRF design software typically gives you access to a technical support team. Plus, EDCs are user-friendly and have built-in help and support.

To recap, studies can be automated and accelerated, which means time and cost efficiencies, data is secure, and there’s training and support for onboarding and beyond.

Whatever system you choose, there are some things you should know about how to design effective CRFs.

Designing CRFs in clinical trials

Good CRF design is essential for a successful regulatory submission, and for getting clinical products to market quickly.

Organizations often struggle with the following design challenges:

• Creating forms that are consistent
• Collecting precise data
• Creating user-friendly form

There are a few things you can do during the design process to help you overcome these challenges:

1. Do proper planning early in the study. This should be done by a team that includes data management, biostatisticians, and clinicians.
2. Define clear objectives and stick to them – what information are you required to gather?
3. Maintain standardized forms – that way, you can reuse the form again.
4. Get user feedback. It’s best to build this into the design and maintenance process.
5. Provide form completion guidelines with specific instructions to reduce data entry issues and ambiguities. This helps investigators fill in forms correctly and ensure that data is accurately captured for analysis.

How do I design quality CRFs?

So, what about the form itself? Below, we’ve listed some dos and don’ts to help you create a well-designed form that can be reused again and again.

Well-designed case report forms must:

• Gather data that is complete, accurate, and high quality
• Avoid duplication
• Be well structured, simple, uncluttered and user-friendly
• Be unambiguous and allow for accurate data entry; use coded lists to limit answers to approved responses only and avoid open-ended questions
• Minimize the use of free-text responses or ‘check all that apply’
• Avoid gathering more data than what’s needed to protect patient privacy
• Include clear guidance using prompts and instructions
• Use consistent formats, fonts, and font sizes across all forms
• Include consistent headers
• Specify units of measurement
• Use visual clues to show how questions are to be answered (for example, the date format, and the number of decimal places)
• Keep related questions together in sections

Example of a poorly designed form

Where it went wrong:

X There’s no guidance given for responses.
X There are no controlled terms to guide the investigator. That means responses are open to interpretation.
X No measurement units are stipulated (is height measured in inches or meters?). This can also make mappings more complex.
X Many questions on a CRF are mapped to variables in SDTM datasets down the line. Allowing free text responses can create a considerable challenge when it comes to mapping and converting data from its raw state to SDTM datasets.

Example of a well-designed form

What are the benefits of storing standardized CRFs in cloud-based systems?

Technology improvements over the last 20 years have led towards creating and maintaining eCRFs using cloud-based systems like CMDRs and EDCs.

Despite this shift, some organizations continue to store their CRF templates in text documents or other similar methods.

In our risk-averse industry, organizations feel more comfortable with a familiar process. They’re under pressure to get studies live and meet the first patient in (FPI). Moving to a cloud-based system involves significant change, and organizations see change as a cause for uncertainty and risk. On top of that, misconceptions around switching to a cloud-based system deter organizations from making the change.