Increasing the effectiveness of the quality system by promoting the quality culture

Continuous medical progress and demographic change in many industrialized countries are driving a steadily increasing demand for medicines. In addition, value chains in the pharmaceutical industry are becoming increasingly fragmented due to outsourcing of production to contract manufacturers. Both trends are leading to an increasing number of production sites for pharmaceutical products worldwide. The human resources of regulatory authorities such as the U.S. Food and Drug Administration (FDA) cannot keep pace with this growth. To ensure the continued high quality of FDA-approved drugs, the agency is currently working on a risk-based approach to inspection planning called the Quality Metrics Program. Higher-risk manufacturing sites will continue to be inspected regularly, while lower-risk sites will be inspected less frequently. Selected quality metrics will play a role in assessing this risk. As part of a research project funded by the US FDA, work is being carried out on the scientific basis for risk assessment. Based on the results of the first year, the project was extended for a further year in summer 2017.

Data basis
Since 2004, the Institute of Technology Management at the University of St.Gallen has been benchmarking the operational excellence of pharmaceutical companies. Today, the database comprises more than 350 sites, making it the second largest database for pharmaceutical production worldwide. Since 2016, additional key figures have been collected on the performance of quality control laboratories at pharmaceutical production sites. This database will comprise over 50 laboratories by the end of 2017. Both databases form the basis for analysing pharmaceutical quality systems.

The Pharmaceutical Production System Model (PPSM)
The basis for the analyses is a model developed during the research project, which reproduces the basic elements of a pharmaceutical production system. The model allows the research team to assign the available key data from the databases to the individual elements of the pharmaceutical production system in a structured manner. The risk assessment indicators proposed by the FDA (published in 2016 in the Quality Metrics Draft Guidance) Lot Acceptance Rate (1), Customer Complaint Rate (2) and Invalidated OOS Rate (3) can be analyzed in the overall context of the pharmaceutical quality system using the developed model.

Scientifically, the model was essentially inspired by two models of excellence. Like the Sand Cone Model by Ferdows and DeMeyer (1990), the developed model follows a hierarchical sequential approach of the competitive factors quality, flexibility, speed and cost effectiveness. Moreover, the developed model combines two key perspectives of any improvement program in the context of excellence. Similar to the European Foundation for Quality Management (EFQM) Model, the model developed within the research project includes a capability dimension (so-called enablers) and a result dimension (performance indicators).

The research team has already used numerous statistical methods to examine the system and its components in detail and to derive generally valid statements. The foundation of the model is the quality culture of the management and the employees. In order to evaluate excellence, effectiveness, i.e. the provision of the right medication in the right quality, in the right quantity and at the right time, was operationalized on the one hand, and efficiency, i.e. the use of resources for this purpose, on the other. Further elements of the model are the stability of the production, the reliability of the suppliers as well as the robustness of the laboratories, which test the finished products before the market release. Figure 1 shows the model developed during the research project.

Quality culture as a basis
The key role of quality culture is widely discussed both in the industry and in the literature. There is consensus that a high quality culture is a critical foundation for business success (Barney, 1986; Digalwar & Sang-wan, 2011; Jochimsen & Napier, 2013; Yu & Kopcha, 2017). Within the research project, one of the main objectives was to analyse the role of quality culture within the pharmaceutical quality system and to complement the qualitative prior understanding with quantitative analyses.

From the main objective derived the specific intention of the research team to analyse the relationship between quality culture and the effectiveness of the quality system of pharmaceutical production sites (see figure). In order to identify whether there is a significant relationship between the two dimensions of PPSM, the sites in the St.Gallen OPEX benchmarking database were divided into two comparison groups and a statistical t-test was performed. For the allocation of the sites into the comparison groups, the key figure Service Level Delivery (OTIF) was used as a proxy for effectiveness (4). The first group includes all sites that are among the top 10 % in terms of this metric. The second group includes all sites that are among the worst 10 %.

For the analysis, quality culture was considered as an aggregated variable consisting of quality behavior and maturity level as well as "engagement indicators". Both quality behaviour and quality maturity represent a grouping of a large number of attributes. The sub-category "Engagement Indicators" represents a group of indicators that allow a statement about the extent to which employees are involved at their location (e.g. number of suggestions for improvement). The quality behaviour describes characteristics of the individual employee within an organisation that can be observed by outsiders. Among other things, aspects such as commitment and active support by superiors in problem solving are summarized in this category. The quality maturity category refers to approaches and methods as well as system characteristics that can be implemented, e.g. the introduction of a standardized process for problem cause analysis (e.g. DMAIC circle).

Based on the statistical analysis, the research team was able to quantitatively demonstrate that the sites with a high effectiveness of the quality system have a significantly higher degree of quality culture than the sites with a low effectiveness of the quality system. This result was additionally demonstrated for each individual sub-category of quality culture. This correlation exists in the areas of quality behaviour and maturity as well as for the "engagement indicators". Also taking into account other factors that influence the effectiveness of the pharmaceutical quality system (e.g. stability of processes or reliability of suppliers), the research result shows a significant correlation between quality culture and effectiveness of the quality system. This frequently discussed correlation could thus be empirically proven on the basis of data from the St.Gallen Operational Excellence Benchmarking Database.

These and other analyses will help the research team, the industry and the FDA to gain a better understanding of the interrelationships of the quality system. In 2018, the FDA plans a first data collection in the industry, the University of St.Gallen will accompany the further development of the risk-based approach for another year. Further information on the numerous analyses of the research team are summarized in a report on the 1st year of research. This was presented at the ISPE Annual Meeting in San Diego in October 2017 and has since been made freely available on the Institute's website.