Harm, Hazard and Risk Part 1

Harm is an adverse outcome or impact, hazard is a potential source of harm, and risk is the likelihood of harm occurring under certain circumstances. Risk is a combination of the chance that a hazard will cause harm and how serious that harm could be. Risk is usually described as being "high", "medium", or "low".

For example, when crossing a road, the cars and other traffic are the main hazards, and the risk is a combination of how likely it is that you'll be hit by a vehicle, along with how seriously you might be injured. 

Ref

ISO 14971, the International Standard for Risk Management

Flying is a routine activity and there are thousands of commercial flights every day. Yet, despite a strong track record of safety, commercial aviation is a hazardous activity. Last year, we were reminded of the risk by two nearly back-to-back fatal crashes of the Boeing 737 Max 8 aircraft that led to a combined 346 fatalities and the worldwide grounding of the entire fleet. In both of these crashes, a sequence of events unfolded due to the interaction between pilot actions and a software control called the Maneuvering Characteristics Augmentation System (MCAS). The trigger event was faulty sensor data from the Angle of Attack (AOA) sensor which caused the MCAS to activate in an effort to stabilize the plane by lowering the nose. When the pilots tried to counter by trying to manually get the plane to nose-up, the MCAS continued to issue the nose-down commands. This sequence of events is shown in the top right of the graphic above. The hazardous situation in this case, as shown in the bottom right of the graphic above, is the failure to achieve altitude after take off due to these events. The result is a disastrous fatal crash with no survivors.

Harm, Hazard and Risk Part 2

ISO 14971 Basic Concepts – Hazard, Hazardous Situation and Harm

1) Hazard

ISO 14971 defines a hazard as a potential source of harm. What does that mean? Is it a “thing”, an “action”, or an “activity”? Well, it could be anything that could result in any harm. It doesn’t have to, but it can. 

Driving a car is a hazard even though we do it every day. 

Surgery is a hazard because it could result in all kinds of complications. 

At the most basic level, use of a medical device itself is a hazard. 

Remember, we are not yet talking about the type of harm or if it is just a small injury or something more serious. We will consider that when we get to the definition of harm later in this blog. 

One problem we have seen in practice is when people look at failure modes associated with a medical device as part of their risk analysis. This is usually done during engineering risk analysis using tools such as an FMEA, or Failure Mode Effects Analysis. A failure mode is a way in which a device can fail to meet specifications or its intended function. Engineers try to think ahead of all possible failure modes so they can control them and improve the reliability of the device. 

But a failure mode, in itself, is not a hazard. It could act as a trigger event that could lead to harm by activating exposure to one or more hazards. This difference is subtle, but important. Controlling failure modes is important for reliability, but it is not sufficient for analyzing safety risks within the context of ISO 14971. That is why using an FMEA for risk analysis is useful, but not sufficient for safety risk assessments. Yet, we have frequently seen FMEAs used as the only tool for this purpose. 

Instead, it is better to look at different types of hazards within the context of the device and its use scenarios. ISO 14971 provides guidance on different classes of hazards such as energy hazards, biological or chemical hazards, information hazards and functional hazards. It is helpful to make a master hazards list under different categories so you can evaluate them holistically within the scope of a given medical device. A failure mode may be associated with one or more hazards, and this standard list can help mapping of failure modes to applicable hazards. In this way, the FMEA can be used exclusively for failure analysis and implementing controls for reliability, and the mapping of failure modes to hazards can facilitate a connection with safety risk analysis. 

2) Hazardous Situation

A hazardous situation is a circumstance that exposes people, property or environment to one or more hazards. 

A related concept to understand in this context is “foreseeable sequence of events”. There is usually a sequence of events that lead to a hazardous situation where people are exposed to hazards. Exposure to hazards through these situations is needed for harm to occur. Again, it may not always happen, but it could. 

As an example, if a medical device is supposed to be sterile, any breach of sterility due to packaging issues, or how it is handled in the use environment, may expose the patient to infection causing microorganisms. The hazard in this case is biological – for example, bacteria, viruses or other infection agents. The foreseeable sequence of events is defective packaging leading to breach of sterility and presence of infection agents. The hazardous situation is this device being used by or on the patient, thereby exposing the patient to these infection agents. 

One problem we have seen is that a clear statement of the sequence of events and resulting hazardous situation is generally missing. Sure, there is a lot of discussion during risk analysis about how a hazard may result in harm, but such discussion is not usually captured accurately to provide a clear understanding of sequence of events and hazardous situation. The reality is that engineers usually do a bottoms-up analysis from device failures, while medical experts focus on a top-down analysis starting from harms and working backwards to hazards. Although these are complimentary analyses, often they are done in isolation by different teams of experts. The result is generally incorrect or missing relationships between hazards and harms, which makes it difficult to properly estimate and evaluate the resulting risks. Inaccuracies in risk evaluation is a common reason for low effectiveness of the risk management process. 

It is through an iterative back and forth bottoms-up and top-down analysis that we can come to a clear understanding of hazardous situations.

ISO 14971 provides guidance on events and circumstances that can lead to hazardous situations. It is a good practice to prepare a master table that clearly shows hazards and statements of sequence of events leading to hazardous situations. 

3) Harm

In the context of safety risk management under ISO 14971, harm is defined as injury or damage to the health of people, or damage to property or the environment. 

It is a very broad definition. Harm could be as simple as a minor inconvenience from a health point of view, or as serious as a life-threatening emergency or even death. In this way, harm has two components – the type of harm, and the severity of the harm.

We have seen two problems in defining harms during risk analysis. The first is not using a standard terminology. Harms should be defined in medical terms, and should ideally be done in consultation with a medical professional. Typically, people have used different complaint codes over the years to describe harms that have been reported. These complaint codes are not standardized and often evolve over time. It is not unusual to see “Other” as one of the most frequently used complaint codes because a reported condition may not always fall under any of the exiting categories. One good resource is MedDRA, or Medical Dictionary for Regulatory Activities. It is a recognized source of medical terms which can be used to create a master harms list for your product portfolio.

The second problem we have seen is inconsistent, and often incorrect, assignment of severity levels to harms. In our experience, the underlying problem is the lack of clear statements of sequence of events and hazardous situations. It is possible that the same condition, let us say a bacterial infection, could have different levels of severity. It is the link between a hazard and hazardous situation that can help us identify the appropriate severity level. In practice, this is not done and the result is inconsistent assignment of severity levels for the same harm because different teams come up with a different analysis of sequence of events and hazardous situations.

It is a good practice to build a Master Harms Severity table that has multiple lines of clear statements reflecting different scenarios. This work is best done in a cross functional team environment with engineers, medical professionals and risk management experts. 

In conclusion, these three terms – Hazard, Hazardous Situation and Harm – are inter-related. It is really important to clearly understand each term and how they relate to each other. It requires a lot of work upfront to build a master table of hazards, hazardous situations and harms relevant to a product or product family. But once you do this work, your risk analysis will be more accurate which will help you manage these risks more effectively. 

What is there in Revised ICH Q9 Part 1

1.High levels of subjectivity in risk assessments and in QRM outputs

Subjectivity is commonly present in a QRM process, as it can be introduced in the many risk assessments, especially in how hazards, harms and risks are perceived. With the revision of the ICH Q9, in a risk assessment, the terminology of the first step has been altered from “Risk Identification” to “Hazard Identification”, for improved alignment with the current definition of Risk Assessment in the guideline. 

The hazards must be identified and later proceed with the analysis and evaluation of risk, regarding the exposure. No further alterations in the risk assessment suggested flow were implemented. Nonetheless, subjectivity in a QRM process can never be fully eliminated – it may be restricted by acknowledging partiality, the application of proper QRM tools and a maximization of the use of relevant data and sources of knowledge.

2. Failing adequately management supply and product availability risk

Quality/manufacturing issues, including noncompliance with Good Manufacturing Practice (GMP), have been a frequent cause of product availability issues. Despite the guideline addressing this topic in its definition of harm as a form of “loss of product availability”, an effective pharmaceutical quality system should empower supply chain robustness and sustainable GMP compliance.

 QRM can help design monitoring systems for detection in departures from a state of control regarding product availability of the supplier, robustness of manufacturing facilities in the process, and acceptability of supply chain partners over the lifecycle. The new revision of Q9 addresses different factors that could affect reliability, such as the Manufacturing Process Variation and State of Control, Manufacturing Facilities and Oversight of Outsourced Activities and Suppliers.

Statistical Process Control in Pharmaceutical Quality Assurance

Statistical Process Control in Pharmaceutical Quality Assurance 

1.Control Charts: X-bar, R-chart, S-chart, and p-chart keep process variation in check.
2. Pareto Analysis: Pinpoint significant quality influencers for targeted improvements.
3. Histograms: Illuminate data distribution aiding in process stability assessments.
4. Scatter Diagrams: Visualize variable relationships to enhance quality mechanisms.
5. Cause-and-Effect Diagrams: Uncover quality issue roots with Fishbone Diagrams.
6. Process Capability Analysis: Evaluate if processes meet quality benchmarks.
7. Regression Analysis: Probe variable relationships crucial for quality optimization.
8. FMEA: Prevent quality issues by identifying potential failure modes.
9. ANOVA: Compare means across groups to ensure quality consistency.
10. Trend Analysis: Proactively manage quality by spotting data patterns.
11. Control Plan: Outline steps to uphold quality standards throughout production.

FDA Requirements for Process Validation

Synopsis of the US FDA 483on Process Validation 

In the Warning Letter, the FDA criticised the fact that no microbiological tests were named in the validation protocol and that the active ingredient specification did not correspond to the current process. Process validation studies were also missing for other medicinal products with regard to different active ingredients and dosage forms.

In the company's response letter, the FDA lacked a detailed description of how the company will ensure in future that the manufacturing process consistently produces medicinal products of suitable quality. The FDA refers to the validation life cycle and specifically mentions the importance of developing the manufacturing process and monitoring after the actual validation. Each significant stage of a manufacturing process must be adequately developed and ensure the quality of the starting materials used, the in-process materials and the finished drug product. Process qualification studies include intensive monitoring and testing of all significant process steps to characterise within-batch variation and evaluate batches to determine whether an initial control state has been achieved.

Successful process qualification studies are required prior to commercial distribution, according to the FDA. Referring to its process validation guidance, the FDA writes that without adequate process validation that considers all production factors and parameters that may affect product quality, the company lacks the basic assurance that it can reproducibly deliver products that meet specifications.

Specifically, the FDA requires:

A detailed summary of the validation programme to ensure that a "state of control" is maintained throughout the product life cycle, together with the associated procedures.

A description of the Process Performance Qualification (PPQ) programme

A description of the monitoring activities to assess intra-batch and inter-batch variability to ensure state of control

A schedule for the implementation of PPQ for each of the marketed medicinal products.

A detailed programme for the development, validation, maintenance, control and monitoring of each manufacturing process, with regard to intra- and inter-batch variability, with the aim of achieving a state of control

A programme for the qualification of facilities and equipment.

A comprehensive, independent assessment of the change management system. This assessment should include procedures to ensure that changes are justified, reviewed and approved by your quality unit. Your change management programme should also include requirements for determining the effectiveness of changes.

US FDA 483 2024 Alkem Laboratories

The authority has issued a new Form 483 following an inspection of Alkem Laboratories Limited, a drug manufacturer located in Baddi, Himachal Pradesh, India. The document was published on 09 April 2024 and goes back to an inspection from 19 to 27 March 2024.

The FDA observed several deficiencies during the inspection, particularly related to the quality system. The 17-pages report lists a total of 10 observations.

One significant observation (Observation 1) relates to the failure to thoroughly review unexplained discrepancies and failures of batches to meet specifications. Specifically, there was a failure in the preventive maintenance of a Perkin Elmer UV Spectrophotometer, resulting in its retirement without conducting an investigation or impact assessment on previously generated test results. Batches tested with this malfunctioning instrument were released into the US market, raising concerns about the accuracy and reliability of the test results.

The other observations include the following aspects:

No shipping studies of finished products shipped to US markets were performed.

Change Controls are not managed and closed within the specified timeframe.

The company failed to adequately perform and assess the GxP impact for computerized system/software.

Appropriate controls governing computer acquired data have not been established.

The firm failed to handle and store drug product containers at all times in a manner to prevent contamination.

Analytical results were not documented according to the SOP.

No annual visual examination of reserve samples was performed.

Understand ICH Q7 S11

What is expected in terms of impurities for APIs extracted from herbal or animal tissue origin [ICH Q7, 11.2]?

In cases where the API itself is the extract from an herbal or animal tissue preparation, all constituents of this extract (concomitant constituents) might be considered to be part of the API. Therefore, a production process-related impurity profile (except, for example, solvents used in the process), would generally not be expected. However, for all APIs derived from herbal or animal sources, tests and limits for possible contaminants originating from these sources (e.g., pesticides, mycotoxins, viruses, herbicides, elemental impurities and wrong species) should be established, based on a risk assessment. In cases where herbal or animal sources provide material that is further processed to yield a chemically-defined API, all constituents other than the API are considered impurities. In this situation, the API manufacturer would be expected to establish an impurity profile as well as an API release specification that would include impurity limits. In any case, it is the API manufacturer’s responsibility to establish batch release specifications for APIs to ensure that they are safe and of high quality, consistent with appropriate regulatory requirements, applicable compendial specifications and regional expectations [ICH Q7, 11.21;  ICH Q9; ICH Q11

In cases where an API test method is changed, which method should be used for stability studies already in progress?

The company should decide and justify the decision of which method to use. All test methods for stability studies [ICH Q1A] should be validated and demonstrated to be stability indicating prior to use [ICH Q7, 11.51]. Any changes to stability test methods should be documented. Applicability of the changes to the existing stability studies should be assessed and may require filing in accordance with regional requirements for post-approval changes [ICH Q7, 13.11]. 

Understand ICH Q7 S10

What is meant by ‘appropriate specifications (of each batch) prior to blending’ [ICH Q7, 8.41]? 

As a principle, no batches with Out of Specification (OOS) results should be blended [ICH Q7, 8.41]. Blending is defined in [ICH Q7, 8.40]. Individual intermediate and/or API batches should demonstrate conformance with the filed specifications prior to blending. In regions or circumstances where there are intermediates and/or APIs that do not require filing, conformance with the release specification should be demonstrated. 

What is meant by ‘APIs and intermediates can be transferred under quarantine to another unit under the company’s control when...’ and is this applicable to contract manufacturers?

ICH Q7, 10.20] states ‘APIs and intermediates should only be released for distribution to third parties after they have been released by the quality unit(s). APIs and intermediates can be transferred under quarantine to another unit under the company’s control when authorized by the quality unit(s) and if appropriate controls and documentation are in place’. The second sentence in [ICH Q7, 10.20] describes transport situations that are not considered distribution. It provides for physical movement (transfer but not release) of quarantined material to another unit. This unit can be on the same site, different site (within the same company), or a contract manufacturer (see final paragraph below). The goal of transfer under quarantine is to allow transportation and testing in parallel. Material that is transferred under quarantine is not to be used for further processing until all testing and quality review is complete and the material is released by the quality unit as defined in [ICH Q7, 2.22]. This provision for transfer under quarantine is included in ICH Q7 for situations where a company is shipping APIs or intermediates from one unit to another and has both the need to expedite the shipping and the material management system in place to prevent use of the material before full release. Examples of circumstances where transfer under quarantine may be needed include extraordinary supply chain requirement(s) (e.g., short shelf-life), and materials with a lengthy timeframe for required test(s) (e.g., some microbiological tests, etc.). With appropriate oversight as described in [ICH Q10 2.7], including a written agreement as described in [ICH Q7, 16.12], and appropriate ongoing controls, a contract manufacturer may be considered a ‘unit under the company’s control’. There is a joint responsibility by both parties to clearly justify and document the need to transfer the unreleased intermediate or API, and to ensure appropriate control is maintained to prevent use before full 

Understand ICH Q7 S9

Which tests are considered to be identity tests? 

For incoming production materials, identity tests and related methods should be used as described in the relevant sections of a Pharmacopoeia monograph, in an approved regulatory filing or in an in-house specification (including method/analytical procedure) [ICH Q7, 7.30]. When available, a discriminating test should be considered for identification testing. The visual examination of a label or the material is not considered sufficient except in the cases described in [ICH Q7, 7.32]. 

Is it possible to extend the expiry date or retest date of a raw material and what is the acceptable practice to determine how

Manufacturing and labelling of raw materials for use by API manufacturers is outside the scope of ICH Q7. As such, retest and expiry dates, as defined in ICH Q7, do not strictly apply to raw materials and may be used in a different manner by the raw material supplier. Expiry date, as defined in the glossary of [ICH Q7, 20], applies specifically to the API. API manufacturers may re-evaluate [ICH Q7, 7.5] and then use a raw material after the ‘expiry date’ or ‘retest date’, based on an appropriate scientific and risk-based justification (e.g., understanding of material attributes, testing, and stability). Similar justifications may be used to extend the date by which the material should be reevaluated. It is the responsibility of the API manufacturer to ensure the raw materials are appropriate for the intended use at the time of use. 

Can yield ranges defined for the first batch differ from latter batches within a campaign? 

Yes. Differing yield ranges [ICH Q7, 8.14] may be described and justified in the manufacturing procedure/master batch record explaining the ranges [ICH Q7, 6.41]. For example, the first batch in the series of production of batches of the same material (campaign) may leave residual material in the equipment, resulting in a low yield in the first batch and contributing to an increased yield in a subsequent batch of the campaign. 

Understand ICH Q7 S8

What is expected in terms of evaluation of suppliers of materials? 

Different phrases are used to describe the expectation for evaluation of suppliers of materials [ICH Q7, 7.11, 7.12, 7.31], including traders, if any. [ICH Q7, 7.12] states that all materials are purchased against a specification and from suppliers approved by the quality unit [ICH Q7, 7.31]. Prior to approval of any supplier, an evaluation should be conducted using a riskbased approach [ICH Q9, Appendix II.5; ICH Q7, 7.31].  More extensive evaluation is needed for suppliers of those materials classified as ‘critical’ [ICH Q7, 7.11].

What is meant by ‘full analysis’ [ICH Q7, 7.31] on batches of raw materials to qualify a supplier? 

A ‘full analysis’ should include all tests specified by the user of the raw material in the regulatory filing. In cases where no filing is required, the full analysis should include tests in other formal written specifications issued by the user of the raw material [ICH Q7, 7.31]. A raw material supplier’s Certificate of Analysis (CoA) may not necessarily align with the user’s specifications. 

Are on-site audits required in the evaluation of suppliers? 

No. An on-site audit is not required; however, an on-site audit could be a useful tool in the evaluation of a supplier. A risk assessment of the material or the service provided can be used to develop an audit strategy and manage the ongoing evaluation of suppliers [ICH Q7, 7.11, 7.31].