Interested in linking to "CAPA and Root Cause Analysis"?
You may use the Headline, Deck, Byline and URL of this article on your Web site. To link to this article, select and copy the HTML code below and paste it on your own Web site.
By Doug Bartholomew, Contributing Editor
Investigating the cause of a quality failure or other production problem is something that all pharmaceutical companies must do — some more frequently than others. The more comprehensive and structured the investigation process is, the more effective it will be.
That’s the consensus of consultants and other experts in the corrective and preventive action (CAPA) field. While CAPA is handled differently at many pharmaceutical manufacturers, best practices for handling complaints and investigations revolve around certain core activities, a basic process and, more often than not, some enabling technology.
The CAPA complaint “root cause” investigation process is paramount. “Having a successful CAPA is more a cultural thing for a company than merely having the technology for CAPA,” says Simon Jacobson, research analyst for manufacturing operations at AMR Research in Boston.
In Jacobson’s view, most large pharmaceutical companies have more trouble “breaking down the silos” — in which they might have several different CAPA processes and technologies being used in different departments and plants — to achieve a single managed view of the process. Many pharmaceutical companies have installed several CAPA solutions, “but there is no common way for them to do trending and put preventive action in place across the company,” he says.
Surprisingly, smaller pharmaceutical firms often take a more systematic approach to handling CAPA complaints than larger companies. “At single-site operations, people are more inclined to work together, so that, for instance, the R&D chief knows the manufacturing guy,” Jacobson observes. “But with the bigger companies, there are more silos and there is a greater need for a common, systematic approach to CAPA.”
Often, CAPA problems can be traced to the lack of consistent process requirements and written procedures. Because of these inconsistencies across sites, companies may have to chase the same complaint at multiple facilities. “This is preventing organizations from halting what could be systemic inefficiencies,” Jacobson says. In addition, he says, “The lack of an organization-wide systemic management process prevents a company from successfully auditing their processes and taking effective preventive actions, such as trending across multiple sites.”
In most cases, a CAPA investigation is set in motion by an event — either a laboratory investigation, a manufacturing inconsistency, a regulatory or internal audit, or a customer complaint. “Any of these investigation records are entered into a CAPA system, which can spawn a related CAPA investigation,” explains Robert Fetterman, president of Technical Business Solutions (Royers Ford, Pa.).
“The majority of companies initiate their CAPA processes once a complaint or field failure is discovered,” says AMR’s Jacobson. “This is not only too late because the patient is already at risk, but companies are hard pressed to comply with the FDA’s CAPA requirements because they lack insight into the root cause of the device malfunction.”
The CAPA process, of course, is broader than just the investigation of complaints. But investigation of the root cause of problems lies at its core. As Jacobson says, “At the heart of a successful CAPA is a definition of the problem and an assessment of who, what and where is at risk, as well as the potential impact on the organization should the problem grow.”
In general, CAPA experts recommend that root cause investigations follow a four-step process:
Fetterman is nearly completion of a project to install a CAPA solution globally at a major pharmaceutical firm. “We’re developing the system and rolling it out to all their 30 sites worldwide,” he says. The firm is using Sparta Systems’ (Holmdel, N.J.) TrackWise quality management software package because it is easily customized, Fetterman says. “It lends itself to quick configuration of workflows and modifications to meet the business needs.” It also runs on the Oracle systems that are in use at many pharmaceutical companies, he says.
With any CAPA investigation, having a prescribed workflow or set of activities with designated people assigned to carry them out is essential. “You can track a CAPA event and the multiple activities required throughout the workflow process,” Fetterman says. Each CAPA record is linked to the original “parent” quality event, so that a supervisor or manager can check the system and quickly see that the event has been tracked through to closure, or if not, exactly where it is stalled or held up.
“This is important to CAPA complaint investigation, that there be a system to make sure that these problems are tracked all the way through to completion,” Fetterman says. “This way, the plant manager knows that appropriate actions were taken . . . and that the corrective actions taken are, in fact, solving the problem.”
The software’s workflow capability helps to ensure that each step is taken in a timely fashion. The system assigns investigation tasks to team members, tracking each task through to completion. Each investigator documents his or her findings in the system, with results and attachments electronically included. The lead investigator is notified as each step is completed or, conversely, unduly delayed.
Unfortunately, many companies have yet to computerize their CAPA complaints and investigations. The danger of manually documenting CAPA problem investigations is that it has proven to be unreliable, Jacobson points out. “The FDA has addressed CAPA as the number one good manufacturing practice (GMP) deficiency,” he says. “And the FDA is more insistent than ever that manufacturers maintain consistent files and procedures across their organizations and embed FDA regulations into company standard operating procedures regarding the CAPA process.”
Some companies have implemented electronic solutions for pieces of the CAPA process. “But very few pharmaceutical firms have implemented a comprehensive process” like the one Fetterman is installing for his current client, he says. “They had a totally manual process before,” he adds.
In Fetterman’s experience, it is extremely difficult to get people across all sites to follow the CAPA complaint and investigation process in the same way. “The biggest challenge we had with this global rollout was coming up with a consistent and reliable approach to handling CAPA complaints and investigations worldwide,” he says.
To establish consistency, companies must have a strong directive from senior management for the project, and a commitment by the organization that “this is the way we are going to handle it,” he says. “You just don’t want to have separate workflows for different sites.”
Of course, even the best technologies and processes are only part of the solution. The cultural issue of getting people in the organization to perform their CAPA responsibilities is the key to any successful program for dealing with CAPA complaints and investigations. “How to incent employees to work on the same page is important,” says AMR’s Jacobson. “The technology is the enabling component, but someone has to own responsibility for the process. A lot of companies have put together cross-functional teams.”
FDA-certified consultants who specialize in CAPA can facilitate its acceptance, Jacobson says. “Having a credible outside voice to come in is helpful,” he points out.
This is especially true for pharmaceutical companies that plan to install a specific software package for CAPA. “For a system like TrackWise, you do need specific expertise on how to configure it,” Fetterman says.
But CAPA technology and consultants will take a company only so far, says Wade Speir, senior consultant in the life sciences practice at Kepner-Tregoe, Inc., a Princeton, N.J. consulting firm. “The software is only a part of the solution,” he says. “There are the process issues, which often require changing people’s behaviors.”
The most difficult aspect of CAPA problem solving is getting a sharp fix on exactly what constitutes the problem at issue, Speir says. “By far the hardest part of problem solving is getting a clear consensus of what the problem is,” he says. “You need to identify what’s actually happening as well as what is expected to happen.”
For example, pills often show silicone contamination from the production process. The manufacturer needs to know exactly how much, if any, silicone was expected to appear in the pill under optimal circumstances. Though drug companies may not admit it, “There’s always some contamination from all the valves and machinery,” Speir says. Knowing the proper baseline — whether there is zero or the most minute amount of contamination — is essential to determining how much actual deviation exists.
To identify the root cause of a CAPA problem, the manufacturer must first specify what the problem is, where it occurs, how often, and how big the problem is. It must also test possible causes against the facts of the occurrence. The idea is to see which cause explains the problem. At the same time, this process also should point out the possible problems that could occur and why they are not.
“The idea is to be able to explain why this batch was bad, as well as why the preceding lot and the one after the bad one were okay,” Speir says. Once the company has a hypothesis for what caused the problem, the next step is to test it — either by experimentation, or by fixing it and seeing if this solution works.
Companies need to be sure that their CAPA root cause analysis investigates all possible causes and tests the suspected one to make sure it is, in fact, the real cause of the problem. It’s important to distinguish between the observed symptoms of a problem and the fundamental or root cause. “You want to be sure people are fixing the right issue and addressing the right cause,” says John McLaughlin, principal consultant at Taratec Development (Bridgewater, N.J.). “You don’t want an engineer putting a fix on a problem without addressing the underlying issues.”
The root cause analysis involves answering a battery of questions that ask “why” with the goal of digging deep until the reason for the problem is identified. Sometimes the cause of the problem may only be a symptom of a more pervasive issue. For example, if a poorly trained operator was the immediate reason for the production of a bad batch, the root cause analysis should delve deeper to determine if the plant’s training program itself is at fault. The question should be asked, could other operators be poorly trained in the future?
“Pharmaceutical firms should be looking across the organization to see if a larger problem exists,” adds McLaughlin. “The good news is always great to hear, but the bad news is what you really want to hear. Companies should remember that, at the end of the day, the CAPA process is supposed to help them.”
All this activity must be documented to prepare a full investigation report for the FDA. Much of this information can be drawn from the batch record, which in many instances, can be found in binders of information. “For many companies this is quite a challenging effort,” Speir adds. “And usually the people doing it are not the highest paid in the organization.”
Identifying the root cause of a CAPA event is only part of the overall CAPA action to be taken. The other half of the process is determining what actions can be taken both to correct and prevent such occurrences.
PharmaManufacturing.com is the site for knowledge, news and analysis for manufacturing and other professionals working in the pharmaceutical, biopharmaceutical and biotech industries.