MD+DI Online is part of the Informa Markets Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Developing Dynamic Device History Records

  Medical Device & Diagnostic Industry Magazine MDDI Article Index

Medical Device & Diagnostic Industry Magazine
MDDI Article Index

Originally Published MDDI October 2005

Electronic Records

Deciding between paper and electronic device records may seem obvious. Less obvious are the considerations that go into creating a DHR that is not only functional, but also makes the manufacturing process more efficient.

By Joseph Vinhais

FDA regulations help ensure medical device safety by mandating the documentation and enforcement of all specifications for production processing, components, quality assurance, packaging, labeling, installation, maintenance, and servicing. The device master record (DMR) contains this specification information, which medical device manufacturers tightly control.

The quality system regulation (QSR) specifically calls for the documentation of each step of the manufacturing process.1 Required items include manufacturing dates, manufactured quantities, and distributed quantity. The QSR also asks for device identification and control numbers, the device primary identification label, and labeling used for each production unit. Acceptance data, such as inspection, test, and quality assurance results, are also required. And manufacturers must provide authorized acceptance signatures to demonstrate that the device is manufactured in accordance with the DMR. FDA also requires that the manufacturer capture all relevant information to reflect strict adherence to the DMR specifications during the production of an individual device or lot of devices.

All of this information is documented in-house using a device history record (DHR). Manufacturers can use various methods for device history recording to improve record-keeping. Ultimately, however, they may want to develop a dynamic electronic device history record (eDHR) that can free storage space, improve product quality, and promote greater manufacturing flexibility (see Figure 1).

Figure 1. Dynamic DHRs can help lower risk and increase return.

Paper or Electronic?

The market research company Y2G Associates recently surveyed 300 manufacturers in the medical device and other industries regarding compliance status and productivity improvements relative to eDHR deployments. Annual sales of the firms surveyed varied from more than $250 million (30% of the surveyed) to less than $100 million (60%). Nearly 62% of the respondents had no electronic DHR system.

Paper device history records produce significant storage needs and raise quality issues. Quality analysis is complicated and time-consuming, with low throughput. Although most companies employ off-site storage, capacities vary, and usually companies store several months of DHRs on-site. Each paper DHR folder measures 2–10 in. thick because of the volume of data that manufacturers must collect. Characteristically, a mid-sized manufacturer using paper-based DHRs generates an average of 50,000 physical DHR folders a year.

Besides the phenomenal amount of storage these papers require, they are difficult to use. Files are often incomplete; important papers get lost or are overlooked. And once they are in storage, it is even more difficult to reference the files.

A seemingly obvious solution is for manufacturers to convert paper DHRs to electronic ones, and many manufacturers are beginning to make that switch. But it is not enough only to convert from a paper system to an electronic one. Simply converting to electronically stored DHRs does not make data easier to locate. Nor is it easier to trace revisions or validate process changes. A better option is to adopt an electronically enforced system. Electronically enforced means that the system pulls data segments and assembles them so users do not have to sift through endless information. It also incorporates checks, balances, and self-auditing mechanisms in a stepwise fashion to prevent operators from performing inappropriate functions. These checks also ensure that the correct people are notified of process deviations. In addition, a full-cycle audit trail defines exactly what has happened to the device at every point in its life cycle. A manufacturing compliance platform using a manufacturing execution system (MES) with eDHRs in its framework can help manufacturers maintain regulatory and business goals under the same system to increase efficiency.

Management System Options

While the information contained in a DHR can vary from manufacturer to manufacturer, fully automating DHRs can help manufacturers partially meet the challenges posed by both compliance and business growth needs. Manufacturers should focus on implementing automation in the following key areas:

• Expanding automated DHR implementation to include in-production unit and batch test records, rework, and field service.

• Changing track-and-trace performance metrics for defective components or processes within hours or minutes to reduce costs and mitigate risks.

• Reducing the number of records requiring correction to reduce shipment delays resulting from inadequate documentation.

Most manufacturers with some electronic systems in place have enterprise resource planning (ERP) and product life cycle management (PLM) systems. If they are especially savvy, they may also have an MES.

ERP focuses on what to build and serves as an order processing and financial management system. What ERP systems do well is provide a single, logical view of the enterprise. It is useful for helping standardize business processes and managing supplied material purchases and production orders. In addition, enterprise systems are the keepers of rolled-up financials. These benefits, however, do not encompass production reality and real-time feedback. ERP systems typically do not provide granular detail below the order level.

Additionally, most ERP systems have a static bill of materials and higher-level, financially driven process routings. They tend to focus more on accounting transactions than data distribution and collection transactions.

PLM systems are engineering systems that focus on how to build devices. PLM systems manage the engineering process and provide a centralized repository of documents. They also serve as the system of record for revision control and document distribution. Unfortunately, PLM systems typically cannot handle complex specifications such as formulations and recipes. In almost all cases, PLM is a static representation of the data—a snapshot in time. Although perceived as a great solution for DHR storage, PLM's real benefit is in the design history file because of the link to engineering and collaboration. PLM systems can store DHRs electronically, but they may not be able to achieve the complexity of configuration and modeling required to gain the detail within operations and production.

MES focuses on products and processes as they are built, and it often works in conjunction with ERP and PLM. In other words, an MES evaluates what it needs from PLM and ERP systems. Typically the systems work on a specific order. First, the PLM system defines the steps necessary to build a product. Next the ERP system evaluates the materials and timing of production. During production, MES dynamically documents the materials consumed, the individuals involved in production, any tests conducted, additional resources, and other details. The real function of the MES is to provide these granular details of the complete product history. It records real-time work-in-progress status and enforces proper routing, documentation, and resource accountability. For example, as a production process consumes materials and resources, users can find out exactly what the process consumed (e.g., percentage of material used) and determine the true cost of quality and manufacturing compliance based on these real figures. Simply stated, MES is the workhorse. It has an enforceable, traceable, electronic traveler that focuses on capturing the millions of transactions that occur in order to make a product compliant, on time, and within budget.

Electronic Device History Records

Static Systems. Although converting DHR systems from paper to electronic is a step in the right direction, not all electronic systems are the same. An eDHR can be either static or dynamic. There are several key differences between static and dynamic eDHRs. Companies using a static eDHR assemble the information at the completion of production and then scan the appropriate documents into the PLM system for electronic storage and retrieval. In most cases it is too costly to scan the entire contents of a DHR. Often, companies only scan the vital and summary information of the DHR. They do not take into account items such as test data noncompliance records (NCRs), supplier corrective action requests (SCARs), corrective and preventive actions (CAPAs), etc.

Each application in the manufacturing process has a unique purpose, and together they operate in a stepwise fashion in manufacturing. First, there are planning functions, which can be 100% computerized via an ERP system. Items such as the product type, its time frame, quantity, and plant information all belong in an ERP system. Second, manufacturers engineer and design the products. These functions, including the units, device history file or specification, and device master record or master batch record, can be electronically tracked by a PLM system. Third, operators receive instructions on procedures, training, and certification. This information is given via a combination of systems. Most manufacturers produce these electronically and then print them out. Fourth, validation, verification, and control of operations are usually manual in a static system, such as a paper traveler.

In other words, for each task, operators receive instructions telling them what to do and then they fill out a form documenting what they did, including values, checks, and events. Next, the manufacturer must have the required regulatory information for review by FDA (e.g., exceptions, device and batch history records). Finally, manufacturers analyze and release the product. The analysis includes quality information, open NCRs, open CAPAs, SCARs, complaints, and so on. Release documentation includes information about samples, product pass/fail, and the provision of certificates of analysis. Much of this information is computerized, but manufacturers print reports. The reports are then filed in the physical DHR folder. Additionally, the manufacturing or product traveler, if not used as the DHR itself, contains much of the verification that processes were completed and that data were collected. But dedicated filing like this does not always happen in real manufacturing plants.

Handwritten entries onto the traveler are a frequent source of errors. Furthermore, continually performing accuracy checks (e.g., to see whether test readings meet specifications, whether the right parts or labels are selected, and so forth) can dramatically slow production. In addition, manually matching labels and components requires verification, which is both costly and time-consuming.

Tracing a problem for a particular product, product line, or product family back to its source can be a paper nightmare; it typically takes hours, days, or sometimes weeks to identify all of the suspected units or lots. Most of this process is not typically incorporated into the overall pricing structure (as overhead, for example), so these extra steps erode margin and profit.

Dynamic Systems. Several important features define a dynamic eDHR. A dynamic eDHR has identification and traceability both forward and backward. It uses and verifies control numbers, unit tracking, lot-and-batch tracking, serialization, and lot and date codes. It is important that these functions synchronize with either the ERP or PLM system to ensure total control. For example, lot splits and combines are improved through synchronization, because if a process error affects only 10% of a product, the manufacturer can send the other 90% through and rework only the problematic 10% before putting it back into the original lot. Alternatively, if a customer places an order for 100 units but needs 10 units quickly, the manufacturer could expedite those 10 units. An ERP system has problems with lot splits and combines because it has to edit the original order or create a subset order. But when used in conjunction with an MES, the splits are much easier to manage. The MES can adopt business rules to model alternate process workflows and additional routings with control and verification requirements.

It is equally important for a DHR to have full tracking and traceability. Ideally, a DHR contains all data associated with a given device. It includes inspection and test data, NCRs, SCARs, and internal corrective action requests. It also encompasses CAPAs, complaints, and return merchandise authorizations. An integrated CAPA solution is one that ties production nonconformances as well as SCARs with complaint management. Finally, a solid DHR leaves a straightforward audit trail with detailed lists of all resources used at all times, documenting individuals accessing the system, as well as all manufacturing steps, equipment, and materials used.

A DHR should document every transaction that occurred in the manufacturing process. Every single step has at least one task and a record, and most have several tasks. Manufacturers need this complete audit trail with records within the DHR. The data are very important—they show which test equipment, which documentation, and which values a given test provided. They detail decisions users made and list all personnel who worked on or made a decision in regard to the device. A truly complete DHR is very difficult to attain, because it requires that all appropriate data for the product are in one place. The benefits of an eDHR is that the data from other systems are tied, synchronized, linked, and integrated together. And even though that often means recording millions of transactions, today's technology makes it possible to maintain files and link to others when necessary. For example, a laboratory information management system (LIMS) contains a record of everything that happened in the lab, such as test data. The MES system is smart enough to know that it needs those test data, but it also knows those data are owned by the LIMS. Therefore the MES links into the LIMS with a specific lot number or batch number for verification of release and completeness.

With a dynamic eDHR, automation is fully integrated. A manufacturing compliance platform using a dynamic eDHR would result in yield improvements and rework reduction, and it would eliminate final DHR checks by quality control. An MES within a compliance platform serves as infrastructure within the factory to bring together disparate data sources, while creating enterprise-wide control and visibility. In addition, real-time analysis of manufacturing operations and tracking of all activities make processes transparent to both regulators and manufacturers.

Specifically, the system captures and records every process step, calibration, and measurement automatically. Bar codes and wireless tags are used to monitor and enforce the correct process and ensure that the right work instructions are available and accessible. The system prints the correct labels as needed, and each station can enforce proper operator certifications. Finally, every item that ships is instantly traceable back to every lot, operation, component, production worker, and supplier. As a result, manufacturing takes fewer steps, and manufacturers achieve faster production. These labor savings alone can drop unit costs by more than 10%.

Conclusion

Although there are many options for setting up DHRs, the ultimate goal is to ensure that they are accurate, current, and easy to maintain. A manufacturing compliance platform that uses dynamic DHR features comprehensive traceability by serialized instrument and serialized component and provides quality information, including incoming inspection data from distributors. Nonconformance and rework tracking result in yield improvements and rework reduction, and manufacturers get the real-time quality information they need for effective analyses. This information helps them resolve issues before they affect other devices, customers, or patients. The fact that final DHR checks by quality control personnel are not needed further reduces analysis time from days to minutes. Online traceability means that manufacturers have necessary data readily available—anytime, anywhere—and that means a better bottom line (see Table I).

Annual Growth Rage 20%

Savings Year 1

Yield

Units per Year
Value per Unit
Est. Yield Increase
 
Stent, 1 line
156,000
$2000
1%
$3,120,000
Yield
Units per Year
Value per Unit
Est. Yield Increase
 
Resuscitator, 2 lines
6000
$800
13%
$624,000
DHR Review
Nonproductive Hours/DHR
Labor Rate
DHRs per Year
Est. Time Reduction
 
Stent, 1 line
30
$50
260
50%
$340,200
DHR Data Collection and Review
Nonproductive Hours/DHR
Avg. Labor Rate
DHRs per Year
Est. Time Reduction
 
Resuscitator, 2 lines
3.15
$30
6000
60%
$340,200
Nonconformance Records (NCRs)
NCRs per Year
Cost per NCR
Est. NCR Reduction
 
 
2000
$800
70%
$1,120,000

Other relevant documents include 21 CFR 210, 21 CFR 211, and 21 CFR 11. For more information on MES, visit www.mesa.org.

Reference

1. Code of Federal Regulations, 21 CFR 820.

Copyright ©2005 Medical Device & Diagnostic Industry

Hide comments
account-default-image

Comments

  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Publish