A standardised approach

A request for information (RFI) procedure should be designed to obtain information on the capabilities of potential vendors, in a format that enables comparisons between vendors. Even after responding to dozens of RFIs for medical writing services in its 20 years of working for the global pharma industry, Trilogy Writing & Consulting is still surprised at how inappropriate many of the RFIs it is invited to participate in are for the services it provides.

Leaving aside those RFIs that were not designed for medical writing services – for example, it receives many RFIs that were designed for full service clinical research organisations – many of the questions Trilogy is confronted with either cannot be answered, cannot provide useful insight, or are challenging to answer due to a lack of specificity.

This article examines some examples of the types of questions the company has received, illustrating why they often elicit little useful information that can be almost impossible to compare objectively between vendors.

Questions that cannot be answered

RFIs frequently include requests such as ‘provide a redacted version of a clinical study report’. There are two problems with this request. First, for confidentiality reasons, Trilogy cannot share any document written for another client, even when redacted. With enough effort, it would almost always be possible for the recipient to deduce whom the document had been written for. Secondly, this type of question aims to assess the standard of writing that can be expected from a vendor. This overlooks the fact that regulatory documents are almost always a compromise among many stakeholders and are rarely written exactly as a vendor would prefer them to be written. So, in the world of technical documents, there is rarely a gold standard document that one can refer to.

A further example of questions in this category is ‘list the top three pharma clients by revenue that you are currently providing similar services to and what percentage of your total business does each one represent’. Apart from the fact that Trilogy cannot divulge information on the clients it works for, its annual revenue is publicly available, therefore, the recipient could then easily deduce the volume and value of work it does for each of the companies cited.

Questions that provide little or no insight

Almost all RFIs Trilogy receives include questions where either the relevance to medical writing services is not apparent, or the questions provide no insight. For example, questions such as ‘provide details of your expertise with logistics’, or ‘has your organisation entered into a financial risk-sharing agreement with one or more sponsors’, have no relevance to medical writing. Even if Trilogy had a risk-sharing agreement, this would be confidential and proprietary information. Financial risk sharing would be highly unusual for a medical writing provider, and the question provides no insight into how the company provides medical writing services. Another example is ‘how will you keep our company informed of industry direction and trends’. The company provides bespoke medical writing services and the purpose of the RFI should be to enable the client to assess its ability to deliver these services, which delivering industry intelligence would not normally be a part of.

The following question was also unusual, as it was not qualified in any way: ‘Is system data available in real time? What is the ability to create ad hoc reports from all data fields?’. As a medical writing provider, Trilogy writes documents and does not collect data or maintain data systems for its clients. Maybe the question alluded to metrics relating to medical writing services, but then this should be specified. While understanding that clients may want to monitor the service they are paying for in some way – see below regarding service level agreements (SLAs) – this is challenging to qualify or quantify. Where specified in sufficient detail, such requests have led to monitoring and performance reporting being potentially almost as much work as the medical writing services themselves.

A frequent type of question in this category is along the lines of ‘list five unique attributes about your approach to a corporate partnership that distinguishes your company’s capabilities from your competitors’ or ‘please describe how you differentiate your company with regard to main competitors’. The basic problem with this type of question is that the company rarely has inside knowledge of how its competitors operate or what their capabilities are beyond high-level information about their service offerings, such as writing regulatory documents. Operating successfully as a medical writing provider involves many facets that go beyond writing, such as working conditions and support infrastructure to mitigate the almost guaranteed shifts in timelines and project specifications, and many of these measures will not be visible from the outside. So, while Trilogy can answer such questions in a general manner, with what it believes may be distinguishing characteristics, Trilogy’s responses may in fact only be what it knows are important for its organisation, and it cannot position these characteristics in the context of its competitors. This means that clients are then almost predestined to receive little or no insight into those characteristics that are relevant for the type of service they are seeking.

In short, clients can only gain the insight they need if they phrase the question in terms of those characteristics that, in their view, are important distinguishing characteristics, and then have the respondents provide the information on how they match up to these expectations.

Questions where the client should provide specification

The questions that are probably the most inadequate for fulfilling the fundamental purpose of an RFI, which is to provide useful information that can be compared between vendors, are those where an inadequate or no specification is provided by the client. This is the most frustrating type of issue encountered in medical writing RFIs. It is ultimately the result of the client either not having a clear concept of what they are looking for or investing insufficient or inappropriate resources in designing the RFIs around their known needs – especially in the context of input from medical writers.

The most extreme case Trilogy has received was an RFI in which it was asked to provide estimates of times required and costs for preparing low, medium, and high-complexity versions of a range of regulatory documents, including clinical study protocols and reports, and CTD summaries. The request for CTD summaries, at each level of complexity, was for a ‘module 2.5’ and ‘module 2.7’. In this case, no information was provided for defining complexity. Anyone familiar with regulatory documents in clinical development will know that there is no standard or effective definition of complexity and that module 2.7 consists of four separate types of documents that can each vary in complexity. When Trilogy requested further details, none were forthcoming and it was asked to provide a ‘realistic estimate’. Without knowing exactly what the client would require – either in terms of documents, their complexity, or the process for working with the client – a realistic estimate can, at best, only be based on Trilogy’s interpretation of the client’s requirements. The problem is that other respondents will each have their own interpretation of the client’s requirements, preventing the client from making objective comparisons and, instead, leave them comparing apples with oranges.

Unfortunately, the majority of RFIs include inadequate information on what the client actually requires. A friend of Trilogy’s senior partner, Douglas Fiebig, who works in the construction industry, spends much of their time responding to RFIs. In their industry these are highly detailed, from the architect’s plans through to the exact specifications for the materials to be used and binding timelines from the issuer of the RFI. While acknowledging the obvious differences between clinical research and the construction industry, many of the basic principles of process, such as careful and binding planning, could be applied to clinical research. In this context, RFIs are part of the planning process for medical writing needs, akin to the architect’s plans. After all, no one would ask for the cost of building a ‘high-complexity’ house without providing any kind of plan.

Of course, in the same way Trilogy responds to these inadequately designed RFIs by providing some kind of standardised metrics and ranges of cost based on its (subjective) experience, a construction company can also provide a subjective response. But this never happens, and for a good reason. If someone wants a house built, they will want it done exactly to their specifications. This is the way it should be when seeking medical writing services.

Other requests in this category go beyond time and cost, such as ‘please provide a listing of your proposed SLAs with their definitions’, ‘please include any SLAs that set you above your competitors’ and ‘provide proposed SLAs for document types’. Again, here the client should be specifying the SLAs that they require and the purpose of the RFI should be to assess the extent to which respondents can fulfil the terms of these SLAs. Also, the company cannot know what SLAs its competitors may be working with. Another example is ‘please describe your approach to cost-savings for our company’s programme (both pre/post implementation)’. The issue here is that the client provided no information on their programme, so it was impossible to provide specific recommendations based on their experience to date. If an envisaged work package had been provided in detail, with insight on the process followed in their company, Trilogy could then have assessed potential synergies that could lead to cost-savings.

The meaning of ‘both pre/post implementation’ was also not clear, so Trilogy could not address this. In general, any potential cost-savings are best assessed and implemented when it has had previous experience of working with the client.

Finally, while the overall intention is clear with questions such as ‘Can you support peaks in the volume of our company’s work as necessary without sacrificing the quality of the end product’, the lack of suitable definitions render the question impossible to answer objectively without further specification. In the absence of appropriate definitions, there is too much scope for subjective interpretations of what the client means by ‘peak’, ‘volume of work’, ‘as necessary’, ‘sacrificing’, ‘quality’, and even ‘end product’. In this case, it would be more effective if the client had described a specific scenario and then have respondents provide a specific solution.

Fit for purpose

Many of the RFIs Trilogy receives for medical writing services are not fit for purpose. They often lack appropriate questions that elicit useful information on the vendor’s services and allow comparisons between vendors. A major issue is that RFIs rarely provide an appropriately detailed specification for the services required, leaving the respondent second-guessing the client’s intentions and providing information that may not match the client’s expectations.

Rather than attempting to describe work packages where the client themselves may be uncertain of the exact specification or requirement, a good approach for medical writing RFIs is to have a two-part process: firstly, a fact-finding questionnaire aimed at characterising vendors in as standardised a manner as possible and, secondly, have selected vendors conduct smaller pilot projects so that the client can assess the actual performance and base any further choice of preferred vendor based on this real-life experience.

In the longer term, it would be advisable for the medical writing community in pharma and vendor companies to devise standardised questions and definitions for use in medical writing RFIs, such as the building blocks for a common RFI standard used across the industry. This approach would enable a focus on questions that elicit useful responses and enable objective comparisons between vendors, simplifying the process and making it more effective for all stakeholders.