Risk Assessment - a frequently underutilised Tool for evaluating microbiological Findings in the pharmaceutical Industry (Part 1)
Risk assessments (RA) are on everyone's lips, whether in the pharmaceutical, food or cosmetics sectors. A federal authority with almost 1,000 employees, the BfR - Federal Institute for Risk Assessment - even has the term in its name. Definitions change hands when it comes to the mostly theoretical introduction to the topic. Our distorted perception is perhaps particularly well characterised by a bon mot from Prof Dr Dr Andreas Hensel, President of the BfR: he speaks of the 'fear of the Western European chain smoker of dying from a snake bite'.
In practice, people have often failed to take all factors into account. This is the case, for example, with the beleaguered Qualified Person, who in many cases is liable for private assets despite all efforts to limit liability through insurance. QPs - insofar as they are socially conditioned by the pharmaceutical industry - are naturally risk-averse. The result: they would rather write off a batch than put it on the market 'under duress' (keyword: residual risk). This fundamental stance in favour of drug safety and patients is certainly to be welcomed, but it ignores the fact that players in the pharmaceutical world often (have to) ignore other risks, such as those arising from global division of labour and supply chains, not least due to economic constraints. The reports of GMP deficiencies uncovered during inspections by various authorities (e.g. via www.gmp-com-pliance.org) are eloquent testimony to this.
What other reasons can be given for the reluctance of those subject to regulation in the pharmaceutical sector to use risk assessments (RA)? (see Tab 1 on page 10).
Regulatory background to the assessment of the microbiological status of finished medicinal products
It is important that the pharmacopoeia actually offers the possibility of using alternatives beyond the methods and limits specified therein. Chapter 5.1.4 (Pharm. Eur.) is generally used to assess the microbiological quality of non-sterile pharmaceutical preparations and substances.
Recommendation
Vienna, Austria3-6 November 2026
Contamination Control & Workshop Risk Assessment in Contamination Control
Tab. 1: Risk assessments (RA): Reasons for reluctance on the part of the pharmaceutical industry (personal compilation and evaluation by the author)
It states:
"In addition to the microorganisms listed in Table 5.1.4.-1 (acceptance criteria), the significance attached to the presence of other detected micro-organisms is assessed according to the following criteria:
- Use of the product: The risk varies depending on the site of application (eye, nose, respiratory tract).
-Type of product: its ability to promote microbial growth; appropriate antimicrobial properties
-Method of administration
-Intended recipient group: The risk for newborns, infants and debilitated individuals may vary.
-Use of immunosuppressants, corticosteroids
-Presence of diseases, wounds, organ damage - If justified, an assessment of the relevant factors is carried out, taking the risk into account. This assessment must be carried out by personnel trained in microbiological analysis and the evaluation of microbiological data.
- For starting materials, the assessment shall take into account the treatment to which the product is subjected, current control techniques and the availability of materials of the desired quality." **
See also SUTTON (2006) and BECKMANN (2015).
However, according to our own research, the instrument is rarely used in the Federal Republic of Germany to rationally justify any deviations that do not compromise drug safety. The authors have provided evidence in one case that, for example, the pharmacopoeia method for counting so-called bile-tolerant Gram-negative bacteria using the probable number method (PN method) does not provide valid results, so that a food microbiological method should be given priority here (BECKMANN et al. 2014). In addition, it had already been proven that the microflora subsumed under the aforementioned term belongs to the plant-typical, so-called autochthonous colonisation (BECKMANN et al. 2011), an extensive risk assessment of all detected taxa was initiated and, ultimately, a limit increase was justified (internal data unpublished). This change notification was accepted by the national regulatory authority.
Critical evaluation of the pharmacopoeia text
1. The wording cited under a.-c. (see below) can rarely be 'refuted' and must therefore always be taken into account, since the argument is valid in principle that the target group for drug applications often or always consists of the groups of people or constellations mentioned under a.-c.
a. Intended recipient group: The risk for newborns, infants and debilitated persons may vary.
b. Use of immunosuppressants, corticosteroids
c. Presence of diseases, wounds, organ damage.
2. Many risk assessment systems, e.g. in toxicology, factor in a significant safety factor. Depending on the system and the risk ap petite of the risk assessor, this factor ranges from 10 to 100.
3. The pharmacopoeia requires that the evaluation be carried out by personnel 'who have training in microbiological analysis and the evaluation of microbiological data.'** To the author's knowledge, despite approximately 12,000 degree programmes in Germany, neither microbiological analysis nor the supreme discipline of evaluating microbiological data is taught. At best, specialists can be recruited who have a solid theoretical and practical laboratory training in bacteriology and mycology, diagnostic experience (!) and practical experience with evaluation using (semi-) automatic conventional biochemical, molecular biological and/or MALDI-TOF methods. It is imperative to familiarise oneself with the databases and algorithms used in the evaluation tools, the general significance of individual reactions and taxonomy. Furthermore, for quality assurance reasons, the diagnosis requires a final plausibility check, as laboratory practice shows that 'paper' or 'electronic data' are patient.
Tab. 2: Key topics/questions that, in the author's view, must be answered in writing as part of an RA beyond the application of a specific methodology
Thinking outside the box
In other areas of production, lively research is being conducted into how microbiological risks (e.g. from Salmonella, Listeria monocytogenes and Campylobacter jejuni/coli) can be reduced through targeted recipe and process management. This growing field is known as 'predictive microbiology' and finds its practical regulatory counterpart in the fact that, for example, food manufacturers who cannot conclusively demonstrate that their products will not allow further proliferation of Listeria, for example, must submit methodically and economically complex challenge tests that comply with standards.
Recommendation
Vienna, Austria3-5 November 2026
Contamination Control - Requirements, Measures and Strategies
When working in predictive microbiology, the question of the ALOP (Appropriate Level of Protection) of an FSO (Food Safety Objective = parameter under consideration) must be answered. This means that a certain level of damage is accepted in the food sector and a strict zero-risk strategy is not pursued. In the case of salmonellosis in the Netherlands, for example, a specific ALOP study calculates 128 cases of illness and one death per 100 million individuals (STRATAKOU et al. 2016). No empirical values are available for the pharmaceutical sector.
The ICH guideline Q9 on quality risk management
The initial version of the European Medicines Agency (EMA) from 2006 has been replaced by the 2023 version and provides a good overview of various accepted risk identification and control systems (ANONYMOUS 2023). Annex 1 briefly discusses the following in detail:
FMEA - Failure Mode Effects Analysis
FTA - Failure Tree Analysis
FMECA - Failure Mode, Effects and Criticality Analysis
HACCP - Hazard Analysis and Critical Control Points
HAZOP - Hazard Operability Analysis
PHA - Preliminary Hazard Analysis
In fact, the above-mentioned guideline only explicitly mentions the terms 'contamination' and 'microbial' in the HACCP and HAZOP systems. HACCP, in turn, is mandatory for food production and originally stems from the safety considerations of the US space agency NASA. The focus here is on identifying and controlling precisely those processes from the detailed processes that ultimately influence food safety, e.g. temperature control within a pasteuriser.
The ICH guideline also recommends in Annex I; I.1. that certain basic techniques be used to structure the RA:
Flowcharts
Check Sheets
Process Mapping
Cause and Effective Diagrams (ISHIKAWA- or herringbone diagram, ANONYMUS 2023)
The methods described above attempt to identify, assess and manage real and implied risks as objectively as possible. This often involves generating numerical values that are supposed to corre-spond to the risk. Here, it can be shown that there is a direct correlation between the respective result and the personal risk perception of the evaluator (see also the groundbreaking findings of psychologists led by Prof. Dr. Gerd GIGERENZER, long-time director at the Max Planck Institute for Human Development, Berlin, e.g. 2007).
When it comes to certain microorganisms in pharmaceutical products, an approach based on the work of Scott SUTTON (e.g. 2006) has proven to be a good first approximation.
In addition, changes in consumer behaviour must be taken into account. Keywords here are: shitstorm, public outrage, life in the virtual bubble (PÖRKSEN 2018).
Unfortunately, there is no such thing as 100% safety. Appropriately, the physician and communications manager Prof. Dr. Klaus HEILMANN (2002) prefaces one of his books with a SHAKESPEARE quote from 'Macbeth': 'Know that safety has always been man's worst enemy.'
| Part 2 of this publication, which will be published in one of the next issues, |
About the Author
Dr Gero Beckmann is a veterinary specialist in microbiology and, after many years in leading positions at contract laboratories, now works as an independent consultant in the fields of microbiology and hygiene.
Notes:
* Not to be confused here with RMPs - risk minimisation plans in the context of pharmacovigilance.
** Translation by editorial team
Literature
- Anonymous (2023): ICH guideline Q9 on quality risk management Step 5 vom 3.2.2023; EMA/CHMP/ICH/24235/2006 Committee for Medicinal Products for Human Use; zul. abgerufen 18.6.25
- Beckmann G et al. (2003): An Investigation into the Microflora of Medicinal Plants during Growth and Harvesting - the Plant-associated Microflora of Melissa, Valerian and Parsley include Enterobacteria.
- Pharmeuropa 15 (2), 291-298
- Beckmann G; Berns M; Goos K; Bradtmöller B; Beermann C (2014): Experimentelle Untersuchungen zur Validität des sog. PN-Verfahrens nach Ph. Eur. zur semiquantitativen Bestimmung galletoleranter, gramnegativer Keime. Pharm. Ind. 76, Nr. 5, 780-786
- Beckmann G (2015): Risikobewertungen von Mikroorganismen im pharmazeutischen Betrieb - Chance und Herausforderung. In: Concept Heidelberg (Hrsg.): Pharma-Technologie-Journal "Analytische Qualitätskontrolle und pharmazeutische Mikrobiologie".175 - 187, ECV Editio Cantor Verlag, Aulendorf
- Gigerenzer G (2007): Bauchentscheidungen: Die Intelligenz des Unbewussten und die Macht der Intuition. G. Bertelsmann, Gütersloh
- Guardiola J (2015): Valsartan Skandal: Rechtliche Betrachtung - Teil 2, GMP-Journal v. 3.3.25, zul. abgerufen 18.6.25
- Heilmann K (2002): das Risiko der Sicherheit. Hirzel Verlag, Stuttgart
- Pörksen B (2018): Die große Gereiztheit. Wege aus der kollektiven Erregung. Hanser Verlag, Berlin
- Stratakou I, den Besten H, Zwietering M (2016): Predictive microbiology for spices and herbs. Berlin. Proc. Konferenz SPICED, 1.-2.6.16.
- Sutton S (2006): How to determine if an organism is "objectionable".
- Pharmaceutical Microbiology Forum Newsletter 12, 2-9.
