In a previous post, I presented the different types of clinical supplies planning systems. Today I would like to dive into another critical factor in defining your strategy towards excellence in clinical supply planning:
In-house versus outsourcing, and the different types of vendors
In-house versus outsourcing
Like for any outsourcing decisions, many factors may influence the choice. It may be the organisation's culture to make/buy and keep control of all decisions, or to outsource anything that is not core. Then the size, structure and skills of the team, and whether bringing in supply chain specialists is an option. Of course, cost of purchasing or outsourcing is not negligible, but beware that the cost of building (and maintaining) your own tool can easily be wildly underestimated! The size and complexity of your clinical trials portfolio will also guide the decision, a small biotech with only 1-5 new trials per year might be better off completely outsourcing the supply planning, and avoid the struggle of creating and maintaining skills and knowledge of a complex in-house system.
In-house make or buy
The most common in-house-made "system" is in fact spreadsheets. As explained in my previous post, there are significant issues with such an approach, from the lack of standardisation to the relatively high risk of growing large intertwined spreadsheets and creating calculation errors. IT-managed spreadsheets, partially locked for edition, with a clear data versus formulas separation, is a safer option.
If you are looking into the more advanced planning systems, I would highly discourage trying to build your own. Unless you have an awesome developer team, supply chain specialists, and mathematicians/statisticians who can spare a large portion of their time for the next 2 years, plus maintenance and improvements afterwards.
I would rather recommend buying / renting professional software, so you would not have to worry about hardware (some offer cloud-based options), software development or maintenance, and may benefit from an acquired specialist experience, influenced not only by your company but by a number of industry peers.
If you do want to create your own system, it is possible to do so and there are in fact a variety of programming libraries available in open source or commercially. To cite only one, OscaR is an open source (LGPL) simulation and optimisation library written in Scala (and hence Java-compatible), it is actively developed in Belgium in a nearby university. I dream of a full-fledged, open source clinical supplies planning system. Anyone in?
Whichever your preferred solution, there will be a significant need for change management and team restructuring. The most frequent approaches I have encountered are:
- "Everyone on board". This can work for smaller bio/pharma companies with a small clinical supplies department, but can be very challenging for larger groups (difficulty in imposing processes, maintaining sufficient knowledge, making the most out of the system...).
- A small team of experts (1-3 FTEs). The most productive option, but there is a higher risk in losing knowledge with departing employees, and it may be challenging to propose an attractive career path to them.
- Key users, as knowledgeable and involved as the above experts, but coaching the other users instead of performing all the analysis themselves. Perhaps the better choice for larger organisations.
Outsourcing has the advantage of drastically flattening the learning curve to benefit from advanced systems, and let your teams focus on the operations. On the other hand, it bears an additional level of complexity in creating a fruitful partnership with a particular provider.
There are many options out there, and they are not born equal. Here is a little bit of guidance.
- IRT / IWRS have been gradually integrating some forms of forecasting and/or simulation into their system. Strong advantages are their obvious fine integration with the IRT, alignment with the resupply algorithms (which is key, and a challenge in every other option below), and the ability to re-plan in the course of the trial based on actuals (randomised patient, current inventory...). A weakness is that the industry tends to think about IRT too late for production planning.
- Some CROs now propose planning or simulation services to their clients, based on either their own solution or licensed software. They are in a good position to perform, since they are involved in other key aspects of the clinical supply chain management. However, one can wonder whether they are not shooting themselves in the foot by helping their clients reduce shipments or material needs, thereby reducing their own bottomline. The opposite stance would be that this is the right path for them to increase their competitiveness and create value for their clients.
- 4PL or Fourth Party Logistics is a good place to look, since in their single-point-of-contact position, they have all the connections they need to know and follow up closely on the study.
- Some of the forecasting/simulation software vendors offer services as well. An obvious advantage is that they are in the best position to master their own software, and can possibly tailor it to your specific needs. On the other hand, software development and supply planning are two very different jobs, so make sure they have qualified resources. The coordination effort may be more intense as well, with one more actor to involve. Also, they may be tempted to promote the use of their system for trials that would be better handled differently.
- Finally, quite a few specialist consultants out there are knowledgeable in clinical supplies, one or more planning systems, and IRT. Their advantage is to be more neutral than the other options outlined here (no conflict of interest) and can probably better align with your organisation's objectives. They may also be in a better position to make recommendations on which system(s) to use for your trials. However, same remark as above about involving one more party.
Are you in favour of in-house or outsourcing? Which option would best suit your organisation? How would you get the most out of a selected solution, with minimal hassle?