Last Updated on

The ability to collect, manage, and analyze data is critical for clinical research, but poor quality data does more harm than good. Only good clinical data supports decision-making and research conclusions, making careful adherence to clinical data management best practices a must to ensure that clinical data is accurate, valid, and reliable. Sound clinical data management practices are so crucial, in fact, that there’s an organization dedicated exclusively to the cause. The Society for Clinical Data Management (SCDM), a non-profit organization which focuses on the development of industry standards and best practices, developed Good Clinical Data Management Practices (GCDMP©), a comprehensive guidance document that outlines all industry-accepted best practices not covered by current regulations.

Here’s a look at the challenges of clinical data management and the best practices you need to know to ensure the integrity of your data.

Challenges of Clinical Data Management

Challenges of Clinical Data Management

Researchers have been managing data sets for decades, but managing clinical data poses a few unique challenges:

  • The sheer volume of clinical and non-clinical data collected in the healthcare industry makes it challenging to identify relevant data.
  • Data in the healthcare industry is often collected in disparate systems.
  • Clinical and non-clinical data are increasingly studied together, yet it often exists in silos. Departments or individuals may even be unknowingly replicating data that already exists (which, when discovered, represents a cost-savings opportunity).

What Happens if You Have Poor Clinical Data Management Practices?

The consequences of poor clinical data management practices are far-reaching. Small errors – even those that don’t have an impact on the scientific conclusion – can reduce a researcher’s or clinician’s credibility if discovered.

What’s more, seemingly insignificant errors in clinical data can be the catalyst for serious errors in patient care with the potential to cause serious, irreversible harm. Something as simple as a missing space between the drug name and dosage, for example, can be misinterpreted, resulting in a patient being given the wrong dose, as in the following example from CRC.gov: “A handwritten order for ‘cisplatinol (sic) 75 mg/m2’ was subsequently typed as ‘cisplatinol75 mg/m2.’ The last letter (l) was misread as part of the dose. The patient received 175 mg/m2 and suffered hearing loss and acute renal failure.” Misplaced decimal points can result in similar issues, with the potential for equally devastating effects.

Mistakes are both time-consuming and costly to fix. Tracking down and rectifying clinical data errors wastes valuable staff time and puts organizations, studies, and clinical trials at risk.

Some organizations turn to DIY methods for managing large volumes of clinical data, but these methods pose their own challenges, including opening up the door to potential security vulnerabilities, which puts regulatory compliance at risk. Spreadsheets, for instance, one of the most commonly used DIY data management solutions, may contain outdated data or data that conflicts with other data, leading teams to focus on the wrong priorities or miss important trends or correlations. DIY clinical data management solutions are also inefficient and wasteful, leading researchers and clinicians to spend most of their time (about 80%, according to HealthCatalyst) hunting and gathering relevant data rather than understanding and interpreting their data. (Imagine the conclusions you could reach if you could free up 80% of your time to focus on analyzing your data!)

Clinical Data Management Best Practices

Following clinical data management best practices is the best way to mitigate these risks. Here are a few of the essential best practices you should know.

Identify Your Resources: People, Technology & Data Clinical Data Management Best Practices

As a first step, Liaison.com recommends locating all data across disparate systems, and identifying those responsible for managing it:

  1. Identify the type of data being collected in each system or location, or by each department.
  2. Identify and list all technology used to collect, store, manage, analyze, and share data.
  3. Appoint a clinical data management team that consists of representatives from all departments and functions.

Your clinical data management team is tasked with identifying how data is used. One of the key benefits of this process is that it often results in identifying areas where data is being duplicated, which in turn allows the organization to develop improved processes and implement integration to reduce wasted resources.

Develop a Data Management Plan

Who owns or is responsible for data governance? Who is responsible for approving access to shared data? You’ll answer these and other pertinent questions with a comprehensive data management plan. In addition to the initial plan, you should implement ongoing reviews to ensure that your access controls are up-to-date and valid – in other words, that only those with a legitimate need to access certain data are able to do so. Your data management plan should also address:

  • Data storage: What data will be stored, and for how long? When will data be archived?
  • Procedures for verifying data integrity and quality.
  • Procedures for addressing discrepancies to ensure cleaner data. This should include an audit trail, and ideally, a discrepancy database where discrepancies are stored.

Develop Standard Practices for Fit-for-Purpose Data

The quality of data is improved when the data collected is relevant to the purpose (i.e., the goals of the study). Forte recommends eliminating non-critical data points to reduce risk during endpoint analysis. (As a side benefit, you won’t be wasting resources to verify non-critical data.) Here’s how to do it:

  1. Define critical data points. What’s the study’s objective? Or, what’s the question or problem you’re trying to answer? What data do you need to measure this?
  2. Standardize your data collection and monitoring processes to reduce errors in data collection and reporting. This is also a resource-saving step, as you won’t be spending time and resources to track down the source of errors later.

Developing standard practices for fit-for-purpose data improves data integrity and also reduces data variability between teams or departments.

Invest in Staff Education

It should go without saying that all team members should be up to speed on current industry best practices. You should also ensure that all team members are on the same page with – and are actively following – company procedures for data collection, management, and analysis.

Invest in Systems That Support Best Practices

Software solutions and other tools designed to improve productivity can actually discourage the use of best practices if they’re not designed to support them. All tools and solutions used should be compatible with company standard operating procedures, and they should be secure and fully compliant with industry regulations such as HIPAA. As a rule, software solutions should reduce the potential for human error when entering data or reporting into the system, and they should allow for efficient data exporting without compromising security. Don’t forget to consider other features that support your operational goals, such as audits, conditional forms, and the like.

When it comes to software, integration is a primary industry pain point. In fact, a continued push for interoperability is one of the biggest health IT trends for 2019, so interoperability should always be top-of-mind when selecting software tools and solutions. API solutions like Emissary, for example, can solve HL7 integration challenges. Emissary allows for the seamless sharing of data across platforms between EHR, clinical and administrative applications and provides real-time access to clinical and administrative data, both structured and unstructured. With Emissary, you’ll eliminate the need for clinicians, researchers, and other staff members to enter data in multiple systems, eliminating data variability between systems to ensure data integrity.

With clinical data becoming increasingly important, not only in research applications, but also in support of clinical decision-making, reliable, quality clinical data is paramount. By following these five best practices, you’ll be well on your way to effective clinical data management that supports your organization’s goals.

Author: Jeremy Pierotti

 

Talk with one of our experts today.