OR WAIT null SECS
Carmen Medina, MPH, PhD, is vice-president, Technical Services, PAREXEL Consulting.
By improving data systems and quality metrics, businesses and company cultures can be strengthened.
Regulatory authorities worldwide are making it clear that determination of the industry’s compliance and quality status will increasingly be informed by the type of quality data companies collect, the mechanisms used to collect data, and how the data are interpreted.
The European Medicines Agency (EMA), in implementing the Identification of Medicinal Products (IDMP) protocol, has ensured that by 2018 all regulatory submissions will be fully electronic, requiring companies seeking EMA approval to harmonize data across all business units and data systems (1). In the United States, FDA’s draft quality metrics guidance indicates the agency will expect companies to elevate and expand systems to proactively monitor production, quality systems, equipment, and facilities to generate more meaningful and actionable quality data and metrics (2). Other FDA initiatives, such as the risk-based Pharmaceutical cGMPs for the 21st Century (3), and the FDA Safety and Innovation Act (4), reflect universally rising expectations for quality data.
Companies have little choice but to revamp the ways in which they collect, manage, and report quality data. Inadequate or non-compliant data systems may lead to significant costs and consequences that could begin with inspectional observations and warning letters and escalate to product recalls. Even worse, substandard data systems-and the flawed quality data that result-may lead to unsafe products, patient harm, and reputational damage.
Bio/pharma companies have choices in how they respond to these expectations, and whether they can seize the opportunity to turn compliance into an enterprise-wide initiative to reduce costs and gain durable competitive advantages. Clearly, if sustainable commercialization is the goal, the choice is easy.
Whatever costs companies incur to improve data systems and quality metrics will be dwarfed by the potential savings they can reap by having actionable data that can prevent a drug shortage or a non-compliance event. To make sense of data and produce the quality metrics regulators want, companies first need to know which data matters. The relative weight of each data input must be understood and ranked in accordance with its impact on outcomes (that is, its criticality). To do this, gathering data in a uniform format is the first step. Then, the quality metrics derived from the data need to be able to answer questions such as: What happened? How often? What is the root cause? What actions are required for remediation and prevention? What are risks of recurrence? What is the best possible outcome? And unless the answers to these questions reach the appropriate people at the right time, the data cannot be termed actionable.
Based on figures from 2016 annual reports of the top 10 pharmaceutical companies, combined research budgets amount to $70.5 billion (5). Moreover, reported global figures are $149.8 billion for pharmaceutical research in 2015 (6).
However, with all the reported spending on research, a great deal of the data that are accumulated remains siloed and does not get adequately transferred for future use by quality assurance-much of it remains within R&D units providing minimal insights for post-commercialization trouble-shooting and risk prevention. Unfortunately, most firms are submerged in data they are not optimally using to make timely decisions about problems, risks, and opportunities. Based on the research expenditures reported above, and given what is known about costs associated with specific millstones such as development, manufacturing, and commercialization, the pharmaceutical industry probably collects more data than any other industry, but much of it remains unmined-neither entered nor stored uniformly and spread out among various business units within the company-making it difficult or impossible to analyze, query, or share. Often, data are removed both physically and organizationally from the people who need them to make timely decisions.
For example, a cold and flu medicine manufacturer faced a significant drug shortage at the peak of the flu season because its plant operators were not receiving real-time information; they were physically and functionally removed from the technical team analyzing production outputs. Consequently, the on-line operators did not receive the production metrics that showed a decline in throughput efficiency in sufficient time to prevent the shortage. The analytics were reengineered to report data trends at regular and shorter intervals. Reports for the production, development, and quality assurance teams were modified to show the same data, in the same format, simultaneously.
The real value of data is retrieved when it can feed advanced analytics, and thereby drive continuous improvement at all stages of the product lifecycle (see Table I). With analytics and actionable quality metrics, developers can build predictive models or algorithms that show what produces desired outputs versus what might be deleterious; shorten development times by ensuring that product quality is defined statistically during process design and maintained throughout its lifecycle; reduce the chance of regulatory delays for new products nearing commercialization by ensuring pre-approval inspection goes smoothly; prevent shortages for marketed products; and avoid regulatory sanctions that can deny needed medicines to patients and damage a company’s brand.
Table I. Benefits of lifecycle quality metrics. MDRs are metadata repositories. SAEs are serious adverse events.
Many companies collect a majority of the metrics needed to improve systems and processes. What they are missing, however, are enterprise-wide analytical tools to enable them to track trends, predict problems, and take timely action.
For example, the cold and flu medicine manufacturer was advised to look to their existing quality metrics for answers, not to build a new system or generate more data. An external review showed the manufacturer was not using the right analytical tools to evaluate process performance. Using their own data (which was robust) and systems (which were adequate), their analytics were customized to show how throughput efficiency could be improved as a result of better utilization of their equipment’s capacity, coupled with enhanced continuous verification of process outputs.
Subsequently, the company increased its throughput by 40% without investing in new production methods or data systems; this action ended the drug shortage that threatened the company’s revenue and reputation, as well as leaving patients vulnerable to the flu.
Most activities in the drug development business can be viewed as a problem, a risk, or an opportunity. An enterprise-wide feedback loop of quality metrics, informed by compliant-and therefore-harmonized data, can provide a platform for industry-leading decision making at all levels of a company.
Competing priorities within a pharmaceutical company can be better understood and managed using data that sorts and categorizes risks and opportunities. A 360-degree view of operations, gained through the effective analysis of quality metrics, increases operational predictability by ensuring the rapid detection of process variations. It can also ensure accountability throughout an organization and lay the foundation for a culture of continuous learning and improvement.
By promulgating new rules and guidelines, regulators today are demonstrating how critical actionable quality metrics will be in the pharmaceutical industry of the future. Smart companies realize that improving their data systems and quality metrics is not simply a matter of compliance; it is about transforming their businesses and strengthening their company culture.
1. EMA, Substance, product, organisation and referential (SPOR) master data.
2. FDA, Guidance for Industry: Part 11, Electronic Records (CDER, 2003).
3. FDA, Pharmaceutical cGMPs for the 21st Century-A Risk-based Approach (Final Report- September 2004; Department of Health and Human Services U.S. Food and Drug Administration).
4. FDA, Food and Drug Administration Safety and Innovation Act, July 9, 2012.
5. Fierce Biotech, The Top 10 Pharma R&D Budgets in 2016, April 26, 2017.
6. Evaluate Pharma (2016) World Preview 2016, Outlook to 2022. London: Evaluate Ltd., p 27.