Flawless: From Measuring Failure to Building Quality Robustness

Many pharma companies today continue to work on resolving issues and process deviations rather than emphasizing what it takes to build products and processes that are robust

By Andrew Gonce, Paul Rutten and Vanya Telpis, McKinsey & Company

1 of 2 < 1 | 2 View on one page

Having zero errors — being flawless — should be the one overriding objective for quality in the pharmaceutical industry.

Over the past seven decades the pharmaceutical industry has grown tremendously and developed a complex system to ensure that patients receive high-quality products. Yet as the levels of complexity have risen, so have the numbers of quality incidents — far faster than the rate of growth. Something more is needed.

PROBING FOR ROOT CAUSES
The hard truth is that there have been big increases in the number of incidents reported by the U.S. Food & Drug Administration (FDA) and in the severity of those incidents. Many of the most acute drug shortages can be traced to quality problems. And most regrettably, there are still instances where lives are lost or health is damaged as a consequence of bad product quality, as happened during the heparin scandal of 2008, or during the drug supply shortage for patients with Gaucher and Fabry diseases in 2009 to 2010.[1]

Pharmaceutical companies are attuned to the challenges, of course. They have been working overtime — and spending plenty — to ensure that quality issues are detected and reported in order to comply with regulators’ quality standards. They have been all too aware of the risk of site closures for noncompliance — and of the enormous damage to reputation that several industry leaders have already incurred. At the same time, the public’s expectations are rising. Now, the expectation for the industry is straightforward: zero lives lost, zero incidents, zero recalls and zero defects.

So why is quality still such an issue, despite decades of vigilance and in spite of so many advances and innovations in production methods and in pharma science and technology overall? In part, the rise in the number of incidents can be explained by steadily increasing product volumes — up by 6 percent per year in the past decade alone. The risk of errors has also risen with the increasing complexity of the pharma supply chain (for example, the number of SKUs on an average packaging line is increasing by 8 to 10 percent every year) and with the fragmentation of the market — from more industry players to more nodes on the supply chain. Moreover, new product introductions are more complicated, featuring everything from advanced coating materials to drug-device combinations to far more diverse patterns of usage by patients. Those factors alone add up to a fundamental challenge: simply to keep quality incidents at the same levels as today, the industry needs to improve control of its processes tremendously.

A more fundamental challenge is that the pharma industry does not readily share learnings, particularly when those learnings stem from quality failures. Information that is shared tends to be focused on the failure incidents themselves, such as observations and warning letters. The industry bias toward punitive measures for noncompliance motivates pharmaceutical manufacturers to focus on tracking failures when they should be encouraged to investigate how to prevent failures in the first place. Interestingly, pharma does much less to learn from its failures than is typical of the nuclear, petroleum and aviation industries, for instance. The nuclear incidents at Chernobyl and at Fukushima, as well as BP’s Deepwater Horizon oil rig disaster, led to publicly accessible and easy-to-grasp explanations that delivered new ways of designing facilities, procedures and processes. The nuclear industry has a history of producing well-written reports that are easy for the general public to understand and are based on clear root-cause analyses that highlight technical, management and cultural failures. The nuclear industry also uses easily searchable databases that contain a wealth of quantitative information on issues such as frequency of failure modes. But there are no comparable levels of reporting and no similar analyses or forums for the biggest pharma incidents.

Additionally, McKinsey’s longtime experience in the industry has shown that many pharmacos are struggling to increase their process capabilities in line with rising expectations and ever-present cost pressures. Too often, these upgrades of the underlying processes and facility investments drop down the list of priorities.

There’s also the challenge of shifting mindsets across an industry that has focused predominantly on compliance rather than on truly knowing the root causes and effects of quality issues. Too frequently, we see failures attributed to individuals rather than being traced back to process or systems issues, fixes focused on retraining instead of permanent corrective actions, inspection responses focused on the warning letter rather than the spirit or intent of the observation. Many letters reference inadequate investigations and preventive actions. If it is tough to tackle the underlying processes and systems, it is markedly more difficult to shift the mind-sets of the quality group, not to mention those of the operations department.

There are other root causes. The industry is saddled with a set of products whose process design has been geared for speed to market, not for quality in mass production; there are few incentives to reformulate and retest products that were proven effective decades ago. Very few pharmaceutical manufacturers have found ways to make low-cost updates to existing processes and face expensive change controls or regulatory filings, which means that known quality issues or underperforming processes can linger for years.

1 of 2 < 1 | 2 View on one page
Show Comments
Hide Comments

Join the discussion

We welcome your thoughtful comments.
All comments will display your user name.

Want to participate in the discussion?

Register for free

Log in for complete access.

Comments

No one has commented on this page yet.

RSS feed for comments on this page | RSS feed for all comments