Ricky Lakhani, Director of Product Management, PHARMASEAL11.20.19
The ever-growing size, risk and complexity of running clinical trials continues to put great pressure on the clinical research industry to evolve practices and challenge not just traditional processes but technologies too. Furthermore, the cost pressures associated with not just clinical trials but the total cost of drug development as a whole adds weight to the argument that companies in this industry cannot simply continue to operate in the same way if they are to introduce the next generation of medicines that will be required for an ageing population. A recent study in 2018 found that “out of 138 pivotal trials assessing 59 new therapies that received FDA approval between 2015 and 2016, there was a more than 100-fold difference in the costs of clinical trials.”1 That rate of growth in costs is just not sustainable in the longer term.
As a scientific industry, we have always employed processes and procedures that are largely scientific in nature. Processes that have been systematic, logical and using the technologies that have been available to us at the time they were actually developed. However, technology in particular has changed dramatically over the years and now changes at such a rate that some of those processes are not just outdated, but revisions of such processes have become outdated even before they have been implemented. Maintaining such processes and technologies just adds to the risk, complexity and cost of running clinical trials.
Monitoring is one of the prime examples of this. Site Monitoring in clinical trials has followed a very prescriptive process with Clinical Research Associates conducting the same activities on the same schedule as if every site is equal. For example, there remains a high focus on 100% Source Data Verification. An “analysis of 9 sample studies demonstrating that only 2.4% of the queries in critical data were driven by SDV”2 highlights that traditional processes are not adding value in the modern world. When it comes to technology, systems have evolved over time to provide some relief to the intensive demands of the CRAs role but have not truly innovated in a way that has forced the industry into changing its practices and processes to benefit from the efficiencies that could be gained. In my mind, there are three important areas that technology needs to focus on in order to drive innovation in this industry which are discussed further below.
Data
Software providers in this industry have always had a focus on the collection of data and providing systems that support data collection to also reduce the amount of errors and overhead of capturing data. EDC systems, for example, introduced edit checks that allowed for data to be queried on entry. This meant that the feedback cycle was reduced, and errors could be handled earlier in the process.
However, what software systems and providers did not put as heavy a focus on, at least in the early incarnations of clinical software systems, was to reduce duplication of data entry when the same piece of data was required by multiple systems. And in fact, this continues to be a problem today even with the expanding wealth of systems and technologies available to integrate data across different systems.
Some larger providers, who have the platform approach and are essentially advocating to be ‘everything to everyone’ have struggled to share data between their own products; let alone with other software providers. Smaller niche providers with targeted solutions have built their products introspectively focused; building for the perfect use case of customers that will use every aspect of their product, and not those that want to utilize aspects of the product as part of a clinical product portfolio. Figure 1 from 2018 shows the spread of software providers within the industry all of whom are collecting at least some data that would be duplicated if being used by the same customer.
Fig.1 Software-Enabled Clinical Trials3
Many software systems have been marketed as having APIs or integration interfaces but that is only a part of the solution. Who actually manages the integration? Provider A, Provider B or the customer? This is a common issue with maintaining agreements to share data amongst systems that ultimately ends up being the customers’ problem. To make a real impact, to really change the industry and to really innovate, software providers should be focused on the outcomes their customers need to achieve with the data that they are collecting and how their solutions fit into the ecosystem. This means providing levels of service such that they can support the those that use every aspect of the product or suite of products, as well as those that require integrating data so that the solution can fit more seamlessly into their systems portfolio.
User Experience
For far too long now, technology solutions in this industry have focused on continually providing functionality and enhancements that simply add to the way things were previously done. For example, even today many clinical systems are essentially spreadsheets wrapped in a ‘system’ user interface that allows users to continue to capture data without implementing workflows or capturing the right data at the right time.
In fact, if we look at the technologies, we use in our day to day life, they behave quite differently from those we use in our professional life. For example, a little simplified, but scheduling monitoring visits is not that different from managing a calendar with recurring meeting invites. Or completing a monitoring visit report isn’t that different from filing in a form or a survey. However, in the B2B environment, we have focused development efforts on adding more columns, more fields, more buttons and more ways to navigate through pages to reach the same place rather than stepping back to understand what the users are truly trying to achieve.
This focus on functionality over usability is a key factor for unhappiness with software systems and why users find other ways to work, avoiding using the systems they have been provided until it is absolutely necessary. Anecdotally, I have personally experienced users who have opted to create their own spreadsheets as it was quicker, easier and simpler than using the validated system that they were supposed to use. They would then, in their own time, copy and paste the data into the system when time was not a limiting factor. The next generation of clinical technologies (some of which are already out there today), must have a laser focus on the people that are using them and the outcomes that they are trying to achieve. Where the user interface and pages have a clear focus, and the information is presented to the user in a clean and uncluttered manner.
Configuration NOT Customization
Any end-user of a process, whether technical or manual, always has an opinion on what they should be doing during the process, the data that they should be collecting and how the systems should be designed. Customization is a common practice in software especially for homegrown solutions that are built for a specific need, but also by software providers that have a strategy of providing the customer exactly what they want—not so much what they actually need.
From a purely commercial standpoint, customization offers software providers the ability to generate revenue by building custom requirements into a product and to then continue to generate revenue through updates and upgrades. It makes a lot of sense from a pure numbers’ perspective. But from a software development perspective and providing value to the customer, it’s a bad idea.
The downsides of focusing on customization are that, as a software provider, you become less nimble—your ability to introduce new features and adapt quickly becomes much more difficult. Furthermore, the overhead on the customer with upgrades becomes less smooth and much more expensive—something that should take seconds or minutes takes days, weeks or even months. This is why I am a huge advocate for configurability over customization as a way of supporting customers’ ability to achieve their goals—albeit bound to an overarching process that ensures that the product has a fundamental core that all customers benefit from.
In clinical software, this is none more present than in areas such as Site Monitoring and Issue Management. With ICH E6 R2 now requiring a “systematic, prioritized, risk-based approach to monitoring clinical trials”, site monitoring solutions in this space need to offer configurability to allow customers to build in monitoring plans that are flexible and can be rules-based as opposed to a one-size-fits-all approach.
The regulations are also indicating a more structured approach for issue management; a key function in the process of site monitoring. This can also be an easy target for customization but does not need to be and could be tailored through simple configuration to support not just what data is captured, but when it is captured and how different escalation paths can work. Issue Management is not new—it is something we interact within our everyday lives when we call a support line—therefore there is no reason to customize what is a very common process.
Customers are bound to have specific requirements for their workflows but supporting them through customization is not the way forward. Configurable systems that are built on a set of core foundations with the ability to tailor rules are what we should be aiming for. This will mean that systems could support a myriad of different permutations without the need for custom development.
The Call to Action
In a recent blog4 about how technology has the potential to be truly innovative and positively disruptive within this industry; I discussed how we have experienced where it has fallen short in the past, e.g. the mini-disc not making a dent on the portable music market. Software in the life sciences industry has to evolve from being like the mini-disc to actually being akin to how Spotify has disrupted the portable music market and provided a positive change in the way we meet the outcome of having access to music on the go.
Software and technologies in the life sciences industry are beginning to—and must continue to be—positively disruptive and not just maintain the status quo in order to achieve quick commercial gains. The industry is now realizing the benefits that greater standardization and re-use of data can provide. If software providers can focus their attention on facilitating their customer’s ability to capture and work with data focused on how users will use that data in their workflows—then this will ultimately benefit the provider as well as the customer themselves. Net result will be more efficient, more accurate and higher quality clinical trials which in turn will seek to reduce cost, provide more innovative treatments and ultimately save more lives.
References:
Ricky Lakhani is an ambitious product management professional with more than a decade of global experience specializing in understanding customer requirements and developing products that are valuable, innovative and successful in supporting pharmacovigilance and clinical operations for life science organizations. He is responsible for managing the entire product lifecycle from strategy, product planning and definition through to development and delivery. Ricky has a Computing BSc Hons from Sheffield Hallam University and has undertaken a number of post-grad courses to continue his professional development.
As a scientific industry, we have always employed processes and procedures that are largely scientific in nature. Processes that have been systematic, logical and using the technologies that have been available to us at the time they were actually developed. However, technology in particular has changed dramatically over the years and now changes at such a rate that some of those processes are not just outdated, but revisions of such processes have become outdated even before they have been implemented. Maintaining such processes and technologies just adds to the risk, complexity and cost of running clinical trials.
Monitoring is one of the prime examples of this. Site Monitoring in clinical trials has followed a very prescriptive process with Clinical Research Associates conducting the same activities on the same schedule as if every site is equal. For example, there remains a high focus on 100% Source Data Verification. An “analysis of 9 sample studies demonstrating that only 2.4% of the queries in critical data were driven by SDV”2 highlights that traditional processes are not adding value in the modern world. When it comes to technology, systems have evolved over time to provide some relief to the intensive demands of the CRAs role but have not truly innovated in a way that has forced the industry into changing its practices and processes to benefit from the efficiencies that could be gained. In my mind, there are three important areas that technology needs to focus on in order to drive innovation in this industry which are discussed further below.
Data
Software providers in this industry have always had a focus on the collection of data and providing systems that support data collection to also reduce the amount of errors and overhead of capturing data. EDC systems, for example, introduced edit checks that allowed for data to be queried on entry. This meant that the feedback cycle was reduced, and errors could be handled earlier in the process.
However, what software systems and providers did not put as heavy a focus on, at least in the early incarnations of clinical software systems, was to reduce duplication of data entry when the same piece of data was required by multiple systems. And in fact, this continues to be a problem today even with the expanding wealth of systems and technologies available to integrate data across different systems.
Some larger providers, who have the platform approach and are essentially advocating to be ‘everything to everyone’ have struggled to share data between their own products; let alone with other software providers. Smaller niche providers with targeted solutions have built their products introspectively focused; building for the perfect use case of customers that will use every aspect of their product, and not those that want to utilize aspects of the product as part of a clinical product portfolio. Figure 1 from 2018 shows the spread of software providers within the industry all of whom are collecting at least some data that would be duplicated if being used by the same customer.
Fig.1 Software-Enabled Clinical Trials3
Many software systems have been marketed as having APIs or integration interfaces but that is only a part of the solution. Who actually manages the integration? Provider A, Provider B or the customer? This is a common issue with maintaining agreements to share data amongst systems that ultimately ends up being the customers’ problem. To make a real impact, to really change the industry and to really innovate, software providers should be focused on the outcomes their customers need to achieve with the data that they are collecting and how their solutions fit into the ecosystem. This means providing levels of service such that they can support the those that use every aspect of the product or suite of products, as well as those that require integrating data so that the solution can fit more seamlessly into their systems portfolio.
User Experience
For far too long now, technology solutions in this industry have focused on continually providing functionality and enhancements that simply add to the way things were previously done. For example, even today many clinical systems are essentially spreadsheets wrapped in a ‘system’ user interface that allows users to continue to capture data without implementing workflows or capturing the right data at the right time.
In fact, if we look at the technologies, we use in our day to day life, they behave quite differently from those we use in our professional life. For example, a little simplified, but scheduling monitoring visits is not that different from managing a calendar with recurring meeting invites. Or completing a monitoring visit report isn’t that different from filing in a form or a survey. However, in the B2B environment, we have focused development efforts on adding more columns, more fields, more buttons and more ways to navigate through pages to reach the same place rather than stepping back to understand what the users are truly trying to achieve.
This focus on functionality over usability is a key factor for unhappiness with software systems and why users find other ways to work, avoiding using the systems they have been provided until it is absolutely necessary. Anecdotally, I have personally experienced users who have opted to create their own spreadsheets as it was quicker, easier and simpler than using the validated system that they were supposed to use. They would then, in their own time, copy and paste the data into the system when time was not a limiting factor. The next generation of clinical technologies (some of which are already out there today), must have a laser focus on the people that are using them and the outcomes that they are trying to achieve. Where the user interface and pages have a clear focus, and the information is presented to the user in a clean and uncluttered manner.
Configuration NOT Customization
Any end-user of a process, whether technical or manual, always has an opinion on what they should be doing during the process, the data that they should be collecting and how the systems should be designed. Customization is a common practice in software especially for homegrown solutions that are built for a specific need, but also by software providers that have a strategy of providing the customer exactly what they want—not so much what they actually need.
From a purely commercial standpoint, customization offers software providers the ability to generate revenue by building custom requirements into a product and to then continue to generate revenue through updates and upgrades. It makes a lot of sense from a pure numbers’ perspective. But from a software development perspective and providing value to the customer, it’s a bad idea.
The downsides of focusing on customization are that, as a software provider, you become less nimble—your ability to introduce new features and adapt quickly becomes much more difficult. Furthermore, the overhead on the customer with upgrades becomes less smooth and much more expensive—something that should take seconds or minutes takes days, weeks or even months. This is why I am a huge advocate for configurability over customization as a way of supporting customers’ ability to achieve their goals—albeit bound to an overarching process that ensures that the product has a fundamental core that all customers benefit from.
In clinical software, this is none more present than in areas such as Site Monitoring and Issue Management. With ICH E6 R2 now requiring a “systematic, prioritized, risk-based approach to monitoring clinical trials”, site monitoring solutions in this space need to offer configurability to allow customers to build in monitoring plans that are flexible and can be rules-based as opposed to a one-size-fits-all approach.
The regulations are also indicating a more structured approach for issue management; a key function in the process of site monitoring. This can also be an easy target for customization but does not need to be and could be tailored through simple configuration to support not just what data is captured, but when it is captured and how different escalation paths can work. Issue Management is not new—it is something we interact within our everyday lives when we call a support line—therefore there is no reason to customize what is a very common process.
Customers are bound to have specific requirements for their workflows but supporting them through customization is not the way forward. Configurable systems that are built on a set of core foundations with the ability to tailor rules are what we should be aiming for. This will mean that systems could support a myriad of different permutations without the need for custom development.
The Call to Action
In a recent blog4 about how technology has the potential to be truly innovative and positively disruptive within this industry; I discussed how we have experienced where it has fallen short in the past, e.g. the mini-disc not making a dent on the portable music market. Software in the life sciences industry has to evolve from being like the mini-disc to actually being akin to how Spotify has disrupted the portable music market and provided a positive change in the way we meet the outcome of having access to music on the go.
Software and technologies in the life sciences industry are beginning to—and must continue to be—positively disruptive and not just maintain the status quo in order to achieve quick commercial gains. The industry is now realizing the benefits that greater standardization and re-use of data can provide. If software providers can focus their attention on facilitating their customer’s ability to capture and work with data focused on how users will use that data in their workflows—then this will ultimately benefit the provider as well as the customer themselves. Net result will be more efficient, more accurate and higher quality clinical trials which in turn will seek to reduce cost, provide more innovative treatments and ultimately save more lives.
References:
- Getting A Handle On Clinical Trial Costs (2019). https://www.clinicalleader.com/doc/getting-a-handle-on-clinical-trial-costs-0001
- Evaluating Source Data Verification as a Quality Control Measure in Clinical Trials (2014). https://journals.sagepub.com/doi/pdf/10.1177/2168479014554400
- Decentralized clinical trials (2018). https://medium.com/elektra-labs/decentralized-clinical-trials-f12fffb72610
- The evolution of technology; being usable as well as functional (2019). https://www.pharmaseal.co/news/the-evolution-of-technology-being-usable-as-well-as-functional/
Ricky Lakhani is an ambitious product management professional with more than a decade of global experience specializing in understanding customer requirements and developing products that are valuable, innovative and successful in supporting pharmacovigilance and clinical operations for life science organizations. He is responsible for managing the entire product lifecycle from strategy, product planning and definition through to development and delivery. Ricky has a Computing BSc Hons from Sheffield Hallam University and has undertaken a number of post-grad courses to continue his professional development.
