Quality Risk Management (QRM) has played a mandatory role in shaping an overall and continuing process of minimizing risks in product quality throughout its lifecycle as part of the regulatory requirements in pharmaceutical industry.

Therefore it is important for us to have some basic intelligence on the chances of failure of the factors affecting the pharmaceutical manufacturing systems to leverage its performance from limited resources and to balance the risk on the most critical systems.

Typical systematic quality risk management involves no other than assessment, control, communications and finally review of risks on the overall pharmaceutical product quality across its product life-cycle. Every step of this systematic process in managing risks will revolve around lots of sound, credible decision makings.

Now the challenge is how to use decision making and apply it to each step in Quality Risk Management as outlined in the guidance for industry, ICH Harmonised Guidelines Q9, which commonly acknowledged by the US FDA as well as by the EU.

Chicken or the Egg

To date, experts in the industry have been arguing about the chicken or the egg question that ‘is effective QRM implementation facilitating better and well versed decisions’ or ‘are logical and evidence based decisions leading to a successful QRM process’.

No doubt both can be correct as long as they are used at the right time and the right applications. Therefore decision making processes can adapt in their reliance on decision maker’s judgement and analytical reasoning method, or known as formal analysis, to meet their goals of good decision making.

The goals of good decision making includes scope, complexity, precision, neutrality, evaluability and transparency.

Scope

A decision making process should address all the issues relevant to the decision maker. Experts can provide judgement that can gather on their broad knowledge to consider risk set before them, however formal analysis can only address risks translated into its standard statement.

As the resultant, expert judgment most possibly has better comprehensiveness on the issue than formal analysis can furnish depending on how much the reach is needed and how well the experts can provide it.

On the contrary, experts often overestimate their ability to think broadly in regards to consider many issues systematically, leading to experts often make decisions by just do calculations to make the expert judgment call for the key issues involving risk identification.

In this case, it will be more appropriate for them to do better by addressing the item in a standard set of risks to be tackled, rather than picking out the knowledge from thinking broadly.

Complexity

Some risks arise, involving the complexity needed to interpret each elements of the relevant evidence thoroughly. Formal methods are usually used to compute statistical assessments of accumulated observations so that it can help to avoid imprecision and biases from human intuitive thoughts.

Expert judgment of statistician comes into play when choosing suitable methods, evaluating usable data quality and addressing difficult situational analysis eg. outliers, missing data, imperfect randomization. Decision makers will also have to take in the considerations of the intensity and the priority of the risks before a planning for solution regarding an issue.

Precision

In the risk assessment process, formal analysis is able to define calculated risks precisely only when there is a well communications between the experts who provide the computational inputs and the decision makers who use the results from data quality.

However, precision is easily overestimated if clear and simple terminologies, for example ‘impact’, ‘issue’, ‘progress’ can mean different things to different people or vice versa. Such ambiguity is likely to cause people to guess wrongly as well as to misunderstand the intentions of the presenter.

Such miscommunication can happened without easily noticed unless the degree of shared understanding is instantly accountabled for.

Neutrality

The choice made in risk quality management by decision makers should reflect their organization values. With the help of experts to summarize the relevant evidence, the decision makers can apply those values.

If the experts’ personal values are potentially affecting their work, the decisions chosen could be subverted, be it with formal analysis or with expert judgment.

The best way to balance expert judgment and formal analysis in achieving neutrality is to enable experts conveying the values integrated in their work, along with the comparability of using alternative values, so that decision makers can choose the standpoint which is closely aligned to their mandate.

Evaluability

Formal methods can address how definitive a risk analysis can be by varying the inputs through the range of plausible values used for sensitive analysis in a risk assessment process, helping decision makers to differentiate if their choices change.

Standard deviations and confidence intervals are good examples of statistical measures of variability that can guide the choice of values for sensitive analyses from what has been observed e.g. in statistical process control.

The use of expert judgement to identify plausible values is common as it can be conveniently evaluated in terms of consistency and accuracy.

Consistency is evaluated by having different ways in asking the same question and examining the compatibility of the answer, while accuracy is evaluated by eliciting judgments sufficiently precise to make comparison to subsequent events.

Transparency

Transparency relates to the importance of good communications between experts and decision makers in the decision making process, so that relevant risk analyses are performed in a continuous human to human interactive manner, minimizing communication breakdown of any sort.

Both formal analysis and expert judgment can differentiate the information needs in risk assessment, while behavioral research can have the information conveyed effectively and evaluating recipients’ understanding.

Decision making processes should simplify communication, by emphasizing on the most critical risk issues and formulating the principle for regulators’ choices.

References

Fischhoff B, Kadvany J, 2011 Risk: a very short introduction. Oxford: Oxford University Press, 2011

ICH Harmonised Tripartite Guidelines: Quality Risk Management Q9, International Conference on Harmonisation of Technical Requirements For Registration of Pharmaceuticals for Human Use, 2005

Marceau G, 2012, “Risk management and Organizational Communication: Two Cases in the Pharmaceutical Industry,” International Journal of Business,17(4), 2012: 325-341