Medical Coding Procedures in Clinical Research: An Amarex Safety and Pharmacovigilance Perspective

During the course of a clinical trial/study, different types of data are generated and collected for the purpose of creating a bigger picture and analyzing possible trends. This data, collected on Case Report Forms (CRFs) for paper-based trials or electronic Case Report Forms (eCRFs) for web-based trials, contains information such as Adverse Events (AEs), Medical History (MH), and Concomitant Medications (CM). When conducting a multicenter clinical trial, there is high likelihood of collecting data which has been recorded in different fashions, especially when considering international study sites. Because of this, it is very important to have a way of organizing all data to be as uniform as possible, for optimal interpretation and analysis.

Amarex Clinical Research, LLC is a global CRO, which specializes in facilitating and managing clinical trials for sponsor companies in the United States, Europe, and Asia. The Pharmacovigilance unit within Amarex Clinical Research is responsible for medical coding. Medical coding is required by regulatory agencies such as the FDA, and Amarex uses the gold standard in medical dictionaries (MedDRA & WHODrug) and coding practices. This article will detail how the Amarex Safety & Pharmacovigilance team performs medical coding using an internally developed safety platform. It will also discuss how we have tackled some of the coding challenges experienced.

Amarex Coding Procedures 

Coding is a standardized and categorized way of pooling data for analysis.¹ Standardization facilitates data exchange using the same terminology. At Amarex Safety, coding is typically done for three types of data:

· Adverse events

· Medical and Surgical History

· Concomitant Drugs

WhoDrug and MedDRA are the dictionaries used for coding adverse events, medical histories, and concomitant medications at Amarex Clinical Research. The use of a MedDRA term is a requirement for the electronic submission of individual case safety reports to regulatory agencies in the European Union, The United States, and Japan.¹

Coding of medical history and concomitant medications are necessary because it helps determine if subjects who participate in a clinical trial are similar or different from the real world population who will be taking the drug.¹ Furthermore, it helps us know the baseline characteristics for each treatment group and helps determine drug disease interactions.¹ The synopses of both dictionaries are below:

MedDRA: This is a clinically validated international medical terminology tool used by regulatory authorities and the regulated biopharmaceutical industry. It facilitates exchange of clinical information through standardization. The terminology is used through the entire regulatory process from pre-marketing to post-marketing. It supports product evaluation, monitoring, communication, electronic records exchange, and oversight. It supports coding, retrieval, and analysis of clinical information about human medical products including pharmaceuticals, biologics vaccines, and drug device combinations. It is updated twice a year.²

WhoDrug: This is an international classification of medicine created by the WHO program for International Drug Monitoring and managed by the UPPSALA monitoring center. It was created in 1968 and is regularly updated for all coding projects. It is used to code medications, classify these into ATC categories and identify the active ingredients associated with each medication.³

The most updated dictionaries are used at the onset of the study. The most recent dictionary version used is also guided by the rate of patient enrollment. If the enrollment is slow, we wait till we have an appreciable number of subjects and data available for coding.

Interdisciplinary Approach to Coding 

Medical coding at Amarex follows an interdisciplinary approach involving the biometrics, clinical, information technology, data management, and Pharmacovigilance units. The pharmacovigilance unit comprising of physicians and pharmacists leads this collaborative effort to ensure the following:

· Utilization of updated MedDRA /WHODrug dictionaries. Ensuring these updated dictionaries are downloaded and imputed into the coding platform. 

· Determination of dictionary versions to be used. 

· Ensuring dictionary formats are in compliance with the CDISC standards for electronic submission of study data in NDA submission. For example, WHODrug dictionary formats (B3 and C3), to which we have recently migrated. 

· Collaboration with the biometrics and information technology units to ensure that there is a seamless transfer of data (concomitant medications, medical history, and adverse events) into the in-house developed coding software platform.

· Collaboration with the site clinical coordinators and clinical research associates via our real-time EDC with the objective of issuing queries and getting the most accurate verbatim terms to facilitate accurate coding.

· Performing both auto coding and manual coding of terms and employing rigorous medical review of the coding listings prior to database lock.

· Application of scientific and medical rigor by utilizing safety MedDRA Queries (SMQs) in exploring safety signals.

For our premarketing clinical trials, our systems are developed in-house. This unique ability facilitates flexibility, cost efficiency, time efficiency, and excellent output in meeting the needs of our global sponsors especially as clinical trials now exist within an ever increasing complex environment. 

The diagram below gives a summary illustration of our modus operandias it relates to medical coding within Amarex Clinical Research.

Challenges and Solutions 

As previously mentioned above, Amarex built its own proprietary coding platform that incorporates the latest versions of MedDRA and WHODrug dictionaries. This proprietary platform allows us to manage any challenges and provide innovative solutions that cater to our particular needs. The following paragraphs highlight these challenges and the brilliant solutions that arose on account of them. As is the practice here at Amarex, each solution is a product of collaborative teamwork among different departments; biometrics, data management, information technology, data operations, and safety and pharmacovigilance.

· Auto SAE uploads

Problem: In processing SAEs, the safety department would begin the process, and then half way through, require the assistance of data operations to upload SAE cases into the tracker.

Solution: To make the process more efficient, the safety department in collaboration with IT and data operations, devised a means of uploading SAEs directly thereby eliminating the time spent waiting for data operations upload SAEs.   

· EDC linking

Problem: During the coding of terms entered from different sites across the world, it is inevitable to come across misspellings, unrecognizable or ambiguous terms. To clarify the information, a query will need to be placed for the site. In the past, if a subject had 30 AEs you had to open each eCRF in OpenClinica to find the particular AE term to query. For very large studies this presented a very laborious and time consuming challenge. 

Solution: In collaboration with IT, we devised a means of linking the exact eCRF to the coding platform for each term. With our latest innovation, EDC Link, this information is available at the click of a mouse, saving a whole lot of time and effort.

· Free text comment box

Problem: Once a term is queried, it remains un-coded, until the site clarifies or corrects the previous entry. With the safety department comprising of several members, it became important to find a way of notifying everyone on the status of an un-coded term awaiting query response.

Solution: A comment box was added on the coding platform, explaining the reason for the query, safety member initials, and date of query. This prevents duplicity of queries and coding of terms with insufficient/ incorrect information.

· Search feature

Problem: With the large quantity of data received in a trial, a single AE, CM or MH term could be reported several times. As you code, you want to check to ensure there is consistency in coding every single term, regardless of how many times it occurs.

Solution: We added a search feature to our coding platform, such that you can check to see if all the same terms are coded for consistency. For example, for AEs you can search to check that all levels of the MedDRA hierarchy, (LLT, PT, HLT, HLGT, and SOC) is the same for a particular term. For CM, you can use the search tool to verify all same medications are coded to the correct ATC codes and drug codes.

· Placeholder feature

Problem: When coding a single term out of a long list of terms, returning to the list brings the user back to the very beginning and you’ll have to scroll up and down to find your previous location on that list in order to continue coding. This was time consuming to sort through, especially for large studies of several thousand terms.  

Solution: With this new feature, after coding a term, the user is brought back to the original spot on the list where they were before coding the last word. This saves a lot of time and effort in the coding process. 

· MedDRA compare tool

Problem: If the MedDRA version is updated in the middle of a study, which is very common, we basically had no way to see which already coded terms may need to be re-coded based on the updated dictionary. 

Solution: With this feature, we can run a report that compares how all levels of previously coded terms may have changed from the previous dictionary version to the newest one. This allows us to selectively update those terms through all levels of MedDRA coding.

Conclusion

Coding is an important aspect of a clinical trial and organizations working with in house developed coding platforms need to be adaptable to efficiently and accurately meet the needs of their clients while working within the confines of the guidelines and regulations of the FDA. This document set out to describe our approach to coding. It described our interdisciplinary approach to coding, schematically elucidated the steps we follow when we receive a coding project, and finally discussed the challenges and solutions that have characterized our coding procedures.

References

1. Klepper M, Cobert B. (2011). Drug Safety Data: How to Analyze, Summarize, and Interpret to Determine Risk. Jones and Bartlett Learning

2. www.who.int/medical_devices/.../MedDRAintroguide_version14_0_March2011.pdf

3. Lindquist M. Vigibase, the WHO Global ICSR Database System: Basic Facts. Drug Information Journal, 2008, 42: 409-419

Ugo Benson-Nwoji MBBS MPH, Nkem Olanrewaju MBBS, Kunal Naik PharmD, Indu Menon B.Pharm, Shide Badri MD, are safety designees at Amarex Clinical Research LLC, located in Germantown, MD.