home > ict > autumn 2015 > site specific
International Clinical Trials

Site Specific

ICT:Risk-based monitoring (RBM) has improved the detection of errors and risks when validating trial data, but do you think it still falls short in some ways?

Clara Heering: Clinical sites have become increasingly complex in terms of the greater number of staff involved, differing devices, and more numerous variations on procedures. This has introduced threats to data quality that cannot be resolved through typical actions such as retraining site staff. Clinical research associates (CRAs) need new ways to enable accurate assessments of risks within the study data.

While some of the new RBM methodologies include systematic approaches to efficiently identify risks, they do not take into account errors that emerge from cultural, technological or other sources that are impossible to rectify through the typical retraining of investigational site staff. A new method to detect the new kinds of risks in modern trials and then mitigate them is warranted.

You've developed an approach using human factors analysis (HFA) to reveal underlying behaviours and get down to the root cause of study errors. Tell us more about the concept.

HFA is a risk management approach used to great effect by many industries outside of clinical research. It is a uniquely structured analysis that identifi es the root cause of errors made when people interact with people, process and/or technology, and therefore the root cause for site and study errors. HFA has entered clinical research in the last 18 months. Currently, we have seven sponsors using it to mitigate risk and errors within their sites.

HFA methodologies have been used successfully in other industries. Give us an example.

The airline industry has implemented it very effectively to reduce mid-air collisions and fl ight delays. Since 1980, HFA – in combination with other factors – has cut the number of crashes annually by two-thirds.

One example of HFA’s value would be when an airline experiences a series of near-miss situations. An HFA investigation would classify behaviours across a broad category of factors; doing so could reveal that the crashes were caused by specific miscommunications among pilots and air-traffic control that are exacerbated by discordance in descent and approach protocols. Through the identifi cation of the root cause, airlines can confi dently mitigate future risk by deploying multiple targeted technologies, trainings and protocol adaptations to prevent those same errors in communication from arising again.

Similar to the airline industry, clinical trials involve specialists of different backgrounds and performance goals within a complicated system. The addition of HFA into traditional RBM has already helped several sponsors to mitigate risk.

How exactly does it work within the clinical trial process?

Using HFA, the CRA conducts source document review to detect the errors that matter, and then analyses these errors in a systematic HFA framework to assess the error sources at that site. This may reveal that a primary endpoint such as weight is not measured with the protocol-required precision because, for example, a site uses multiple uncalibrated scales. Since a training issue was not the root cause of the lack of precision, retraining site staff would not have solved the issue.

So what critical quality issues can HFA solve that source data verification (SDV) cannot?

While SDV efficiently identifies the presence of discrepancies in data, it is possible that this method misses deeper threats to data quality. In addition, SDV typically adopts a similar approach to resolve every error identified. By contrast, HFA enables CRAs to mitigate risk at the source – which can be a diversity of factors in the more complex modern trial – and thus deploy bespoke mitigation strategies to prevent the same error from reoccurring.

For example, imagine that a CRA identifies that a site has under-reported concomitant medications. Hand-written logs contained the data, but they were not entered into the electronic data capture (EDC) system. In an RBM-based monitoring environment, the CRA would deploy training on proper use of the EDC system. However, a CRA employing HFA would instead fi nd the main human factors behind the problem. This may show that the lack of a consistent oversight process was the root cause; each new study coordinator would make the same mistake. This would lead to the CRA asking the principal study investigator to implement a policy. Retraining would not have solved the issue. Furthermore, during the HFA investigation process, the CRA may identify previously undetected issues about deficiencies in the details collected about adverse events.

Real-time site data can be presented as visual analyses. What practical answers can this give?

Visualisations of site data help to reveal trends quickly. Take a trial in which blood pressure measurements must be recorded precisely, rather than being rounded to the nearest 5mmHg. A visualisation of the number of unique blood pressure measurements recorded at a site would immediately reveal if the unique values clump to rounded numbers, indicating a violation of the trial’s protocol. When HFA is applied at a site found to have usual clustering of blood pressure values ending in 5’s and 0’s, the investigation may reveal that, instead of retraining, the site needs technical assistance to adjust the settings of the blood pressure machines to not report rounded values. The combination of real-time data, visualisation for rapid identifi cation of errors and HFA to address the true root causes of errors affords sponsors a powerful way to safeguard study quality, and resolve issues much earlier in trial execution.

Given its benefits, why do you think the clinical trial sector has not been drawn to HFA before now?

As with many industries, pharma had adopted standards to enable quality processes: the chosen standard for monitoring has been 100% SDV. This method was appropriate when the investigational site had a stable environment and protocols were less complex: the same individual implements all protocol procedures with study subjects, using the same devices and processes throughout the trial.

Today, in a rapidly changing environment, sites deploy a multitude of staff, and regularly introduce new processes and devices. With the increase in protocol and site complexity, we needed to look into new ways of monitoring consistency in data accuracy and precision, helping the CRAs to focus on what matters.

In recent years, the industry has focused on innovations, especially within trial monitoring approaches. Significant momentum has developed behind improving monitoring tools such as RBM to address the challenges of modern trials.

Can you give us an idea of the cost savings in monitoring that can be made?

Initial operational successes in terms of identifying and resolving data quality such as protocol non-compliance have been observed. As sponsors have only recently adopted HFA, specific cost savings metrics are forthcoming as these trials conclude.

Do you think you'll set a trend – will HFA use within the industry now grow?

It is a trend we expect will continue to spread through the industry. From aeronautics to the automobile industry, HFA has had notable impacts on risk management. So far, seven pharma sponsors have set out to incorporate HFA. There is no indication yet that the benefits for the clinical development space will fall short of historical successes in numerous other industries. We are encouraged that HFA can continue facilitating improvements in the quality of clinical trials.

Read full article from PDF >>

Rate this article You must be a member of the site to make a vote.  
Average rating:

There are no comments in regards to this article.

Clara Heering is Vice President, Clinical Risk Management at ICON. She has 23 years of clinical development experience, starting as a CRA and growing into senior operations and innovations roles in large pharma and CROs. Clara has an MSc degree in Risk Analysis from King’s College London.
Print this page
Send to a friend
Privacy statement
News and Press Releases

£10.4m Full-Service Contract Signed With Existing Big Pharma Client

hVIVO to develop new influenza challenge strain to test oral antiviral candidate
More info >>

White Papers

Characterisation of Biopharmaceutical Proteins

Reading Scientific Services Ltd (RSSL)

Over the next five years it is anticipated that there is going to be an explosion in the numbers of biosimilar products coming to market as patents expire. Consequently, in line with regulatory guidance, there will be a commensurate need to provide full characterisation of such biopharmaceuticals. The purpose of this article is to describe the array of the more common techniques used in biopharmaceutical characterisation (typically of protein or polypeptide). For full characterisation of a protein, the protein�s primary, secondary and tertiary structure as well as its physiochemical properties should be assessed.
More info >>




©2000-2011 Samedan Ltd.
Add to favourites

Print this page

Send to a friend
Privacy statement