How optimization and simultaneous electronic support of actual laboratory processes can boost efficiency

Laboratories working in the pharmaceutical industry in the areas of R&D and quality control find themselves increasingly having to cope with conflicting demands — tougher regulatory requirements and harsher economic realities. In order to meet these demands, new ways of dealing with process, data and system management are necessary. This article shows how the paperless lab (as an integrated process, and not as a system implementation) can meet these challenges and boost efficiency.

Why new concepts for laboratory data management?
On the one hand, the authorities require more data with more quality, such as with the new EU GMP Annex 11 and the more intensive FDA inspections regarding 21 CFR Part 11. And, on the other hand, there are the commercial pressures requiring more data in less time, typically with the same level of staff or less. On top of that, advances in technology and laboratory equipment mean that even more data is being generated faster, which creates almost insurmountable obstacles for conventional documentation and data-management processes in regulated laboratories. These obstacles, in turn, causes bottlenecks for the development and approval processes.

The root of all evil: hybrid systems
When you look at data management in the lab, what you usually see is a mixture of various independent, non-integrated data-processing systems on paper and in electronic form. Often, paper is the preferred documentation medium when you have a mix of countless computerized systems, such as analyzers, office applications and higher-level systems like laboratory information management or enterprise resource planning (ERP) systems. Such a scenario results in producing a hybrid system with numerous media gaps — this is the real root of all evil, leading to inefficiencies, quality and compliance risks and unnecessarily long throughput times, which prevent businesses from hitting their targets.

Table 1 shows the typical key performance indicators of such a laboratory. The high level of quality risk (due to the high number of manual data transcription steps) is countered with extensive control steps — but that results in less efficiency and longer throughput times. The use of isolated systems prevents the timely transfer of information, which leads to further unforeseeable delays and additional costs.

 

Table 1: Typical key performance indicators for quality control and development laboratories
KPI	Value*
Effort documentation and control	45%
Number of quality-relevant transfers (per approved batch)	100
Number of redundant data transfers	250
Process time	up to 20 days
Systems in operation (paper and electronic)	10
*With the exception of the documentation and control effort, all numbers refer to the approval or analysis of one batch. Quality-relevant transfers have a direct impact on the result; redundant data transfers provide references and cross-references.

And last, but not least, it can result in a considerable amount of the enterprise’s intellectual capital being wasted. The cost of using various isolated systems to collect data for modern knowledge-management-systems (statistics, data mining, reporting, exception handling, etcetera) is simply too high. So, much knowledge that could be extracted from the data collected remains untapped.

 

Figure 1: Distribution of tasks for a specific laboratory process as a percentage of work hours. The colors display the categorization in four categories; the documentation and control categories could be substantially reduced by the introduction of a paperless laboratory.

In the industry, there are three different approaches to addressing this root problem. The first — and actually no real solution — is the optimization of the existing hybrid system by adapting the existing processes and systems. It goes without saying that such an approach only brings selective and slight improvements. The second is the introduction of electronic documentation systems (often described as electronic lab notebooks) that show the paper data in electronic format. While the introduction of such a system brings certain benefits for quality and compliance purposes, the real problem is simply transferred from paper to an electronic format (“paper on glass”) and the hoped-for gains in efficiency are only very small. And the practicality of introducing electronic documentation into the laboratory (using tablet-PCs or other mobile devices) is highly questionable. The third approach, which we discuss in more detail below, is the optimization and simultaneous electronic support of the actual laboratory processes — the introduction of the paperless lab.

The way to a paperless laboratory
Since introducing a paperless laboratory involves more than simply implementing another IT application, the procedure model we use here takes into account other perspectives in order to build a sustainable overall concept. These perspectives fall into three main areas — the business needs, the user’s needs and the technical perspective.

 

Figure 2: Distribution of non-process activities per department as a percentage of work hours.

The business point-of-view defines the project’s goals and vision and ensures that the introduction of the paperless laboratory complements the company’s overall business objectives. Typical goals for producers are increasing net cash flow, reducing development time and, thereby, generating additional sales or reducing warehousing and stock costs to free up more capital. Naturally, various other goals or combinations of targets are possible — but what is decisive is that a quantitative connection can be made between laboratory activities and business objectives. This is achieved by defining a business-specific cause-effect model and implementing it into a corresponding financial model.¹That is the only way that you can ensure that the implementation project meets the company’s targets and that the cost-benefit analysis accounts for all relevant factors.

The user’s point-of-view is the central element to developing the paperless laboratory concept. Based on a thorough process and system analysis, the user’s perspective will be summarized in the form of process descriptions and data streams. As well as the actual laboratory procedures — such as sample flow and processing, supporting processes will also be mapped (such as apparatus maintenance, reference substance management and reagents) in actual state. For every process step, information will be collected from any systems used, any responsibility changes and any data inputting and outputting. Key performance indicators will be collected from the process analysis for later integration into the financial model. Vital interfaces with higher-level processes will be analyzed and documented. The analysis process is also the foundation for the subsequently conducted multi-moment analysis — a methodology that allows quantitative information to be derived from and for processes.

Multi-moment analysis^2,3 provides statistically sound and accurate information about the use of resources per process step or sub-step — and for all processes. To collect data, every laboratory worker is equipped with a mobile device or PDA (configured for their particular tasks) that requires them periodically, but randomly (on average every 20 minutes) to select the task they are currently performing from a list. The impact of multi-moment analysis on the workflow is minimal, because just a simple click suffices — no recording of times or other parameters is required. Over a typical period of two weeks, sufficient data is collected to enable highly detailed statistical analyses to be made on costs, time and clustering for processes, process steps and task categories.

 

Figure 3: Aggregate values for all process and non-process activities for five different categories per department and as a total amount. The documentation and control categories show the potential of the introduction of a paperless laboratory.

Figures 1–3 show examples of various evaluations from a multi-moment analysis. The multi-moment analysis not only allows processes (independent of whether a new system is introduced or not) to be optimized where the greatest need or benefit lies, but also permits the work stages to be categorized — thereby enabling a quantitative assessment of the potential benefits that the implementation of a paperless lab would bring (Figure 3). That, in turn, provides the foundation for a fact-based business case and for the comparison of various implementation scenarios and their economic benefits. The qualitative process description (together with the key performance indicators and quantitative statements from the multi-moment analysis) thus provides the crucial information for the paperless laboratory concept development.

Taking the technological (i.e. IT and equipment) point-of-view — and looking at the existing infrastructure, company standards and long-term strategy — enables a comprehensive concept to be developed for the automation of the laboratory data-flow process. Only then will such a concept be in line with the company’s commercial targets.

To develop the paperless lab concept — as well as considering the three perspectives outlined above — you also need a set of principles that allow for a definition of process targets based on the process description of the actual-state processes. The key principle and the vision of the paperless lab is the self-documenting process — a process that produces GxP-compliant documentation and eliminates unnecessary tasks from the workflow. That naturally means that manual data collection and transfer are eliminated wherever possible by interfacing to and from devices and systems — and where that is not feasible, by using barcodes for fast, error-free data collection. That also means that redundant or fragmented data is eliminated, that the “single-source-of-truth” principle is implemented and that data is available to every authorized user in real time. That, in turn, leads to improved processes, faster decision making and better teamwork.

The application of these principles to the actual-state processes enables target-processes to be defined and functional requirements to be established. As part of the paperless lab concept, the delta between the current functionality and the required functionality will be generated in the gap analysis. This is the only way the filling of such gaps may be conceptually approached — naturally always considering the underlying business targets.

Possible scenarios also include the modification of existing IT systems by adapting existing applications to close functional gaps and multiple scenarios to close the remaining gaps with other applications. Once, the only kind of application specifically aimed at laboratory use was a laboratory information management system (LIMS). However, today there are many types of applications providing overlapping functionality. For example, electronic laboratory notebooks (ELN), sometimes for quality control referred to as laboratory execution systems or LES (derived from manufacturing execution systems or MES); archiving or raw-data management systems (scientific data management system or SDMS); specialized applications for device and system integration, increasingly also software that was not originally laboratory-specific, such as product-lifecycle-management (PLM) systems or ERP systems. Some software firms also offer combinations of the above-named applications, and there is generally a trend toward an extension and overlapping of functions or convergence of applications.

There is no quick answer as to which application combination is the best for a company. But, by using SWOT analysis (strengths, weaknesses, opportunities and threats) to compare the various possible combinations — naturally accounting for the overall business targets and not just the needs of the laboratory — the selection can be reduced to one or two scenarios.
The next step is to select suitable products to fit the desired application combination for the developed scenario. With a sound concept, defined target processes and their functional requirements, that is a relatively easy undertaking. As well as the functional considerations, it is also necessary to consider the complexity of the IT landscape and soft factors, such as the readiness of the software suppliers to cooperate.

Before making any final selection — and to clear up any technical issues and to get prospective users involved at an early stage — it is recommended piloting the whole solution in small, well-defined stages. Such a pilot scheme — which costs little in terms of money and risk — should cover as many areas of the paperless laboratory as possible and helps fine-tune the final implementation planning. The implementation of the final solution is made in accordance with widely accepted standards, such as GAMP.⁴ Because the pharmaceutical industry is so highly regulated, the whole system must be validated to meet strict requirements.^5,6 The work invested in the procedure model (e.g. process analysis, concept development, setting target processes and piloting) pays off here, contributing toward the necessary documentation. It should not be forgotten at this point that the paperless laboratory is not just the introduction of a system, but is also — above all else — process re-engineering. It is, therefore, necessary to support the new optimized process landscape with all necessary standard operating procedures (SOPs) and guidelines to maximize the benefits of the paperless laboratory.

Faster, more precise and more economic
Because of its integrated and process-oriented approach, the benefits of the paperless laboratory compared to using a specific application (such as an ELN) are several factors higher. The automation of the data flow and the continued elimination of documenting and related control activities significantly boost efficiency for laboratory staff and management. Already through these effects alone, efficiency gains of 20 to 30 percent are feasible (depending on a company’s situation).

Additionally, the virtual elimination of manual tasks substantially reduces processing times and quality risk. System-specific automatic checks ensure compliance and data consistency and reduce control activities to atypical results (“review by exception”), thereby accelerating processes and improving resource allocation. Since automatic documentation also automatically generates numerous process-relevant parameters — such as throughput times for certain tasks, equipment utilization or sample logistics — there is an excellent source of data for laboratory management. The paperless laboratory delivers key performance indicators for its own continual improvement, free-of-charge — and enables the comparison of organizational units using high-quality data. And the availability of real-time data benefits other areas outside the laboratory — such as simplifying inter-departmental cooperation, supporting knowledge management and improving follow-up processes. From the users’ point of view, the paperless lab substantially simplifies the workflow and reduces the number of systems implemented. It also supports user-specific portals and sharpens focus on what is important.

It is clear that the sum of the paperless lab’s benefits is enormous. As with all investment projects of this size, after implementation, evidence should be presented to show that the estimated business case and the reality correspond. This can be done at any time with the help of multi-moment analysis, which can also be used to quantify the impact of any process change on the workflow.

References
1. Ritter J. Reducing Cost by Automating Laboratory Workflow. G.I.T. Laboratory Journal. 2009;13(7 – 8):32 – 34.
2. Haller-Wedel E. Das Multimomentverfahrenin Theorie und Praxis. Munich (Germany): Carl Hanser Verlag; 1969.
3. Simons B. Das Multimomentzeitverfahren, Grundlagen und Anwendung. Kologne (Germany): Verlag TÜV Rheinland GmbH; 1987.
4. ISPE: GAMP 5, a risk-based approach for GxP compliant computer systems; 2008.
5. Food and Drug Administration: 21 CFR Part 11.
6. EU-GMP Annex 11.

Ulf Fuchslueger is the owner and  Andreas Schild is a senior IT consultant at Vialis AG, Liestal, Switzerland. They may be reached at editor@ScientificComputing.com.

Acronyms
ELN Electronic Laboratory Notebook ? ERP Enterprise resource Planning ? GMP Good Manufacturing Practice ? LIMS laboratory information management system ? MES Manufacturing Execution System ? PLM Product Lifecycle Management ? SDMS Scientific Data Management System ? SOP Standard Operating Procedures ? SWOT Strengths, Weaknesses, Opportunities and Threats