Key opinion leaders in psychiatry: A conflicted pathway to career advancement

The Lancet review by Hickie and Rogers (2011) of the antidepressant agomelatine attracted critical correspondence from around the world about its misrepresentation of the drug’s efficacy, clinical significance, and comparative side-effect profile, as well as the relationships between the authors and the drug’s manufacturer (Barbui and Cipriani, 2012; Carroll, 2012; Howland, 2012; Jureidini and Raven, 2012; Lloret-Linares et al., 2012; Serfaty and Raven, 2012).

Unfortunately this is not an isolated case. The academic literature is undermined by a combination of publication bias, selective reporting of outcomes and adverse events, misleading conclusions unsupported by the data, and citation distortion (Greenberg, 2009; Ross et al., 2012). Many of these misrepresentations can be traced to the pharmaceutical industry (pharma), but ultimate responsibility lies squarely with the named authors and journal editors.

Conflict of interest is of course universal; we all bring biases and prejudices to any discussion. Personal financial gain is perhaps the lesser influence (at best, financial entanglements provide a surrogate marker of the capacity of authors to separate themselves from conflicting interests. We might need to know about authors’ ethical behaviour in other spheres to more accurately judge their capacity to maintain independence in data analysis). Of more importance are the ongoing partnerships that pharma has with influential doctors, referred to as key opinion leaders (KOLs) or thought leaders. A KOL is created whenever a psychiatrist who has influence on the practice of other doctors engages with pharma as an advisor, researcher or educator. Whatever the intent of the psychiatrist, the company has an eye to the KOL’s impact on sales. The most important KOLs have a dramatic effect on national and international practice, but also valued are those KOLs who influence the prescribing habits of local networks of doctors. It is interesting to speculate why this phenomenon has attracted little attention from academics, with the most authoritative paper on the subject being written by a medical journalist (Moynihan, 2008).

It has been argued that provided the relationship between KOLs and the drug industry is transparent, ‘there is much benefit to be gained on both sides’ (Buckwell, 2008). Both do gain; typically the psychiatrist’s career development will be advantaged and their work with pharma will enhance the marketing of medications. Rather it is patients and the health system that stand to lose through resultant inappropriate prescribing. The ideal of a company and an expert coming together to develop and promulgate an exciting new advance is rare. KOLs tend to be recruited after the most important phase of drug development, in the 2 or 3 years before a product launch (Cutting Edge Information, 2009). And, as is the case with agomelatine, the focus for KOL promotion is usually a drug that represents no significant advance on those already available.

It might be argued that a KOL can maintain independence by avoiding promotional activities. But pharma documents clearly indicate the promotional role of KOL teaching and research (Forrest Laboratories, Inc., 2003), as does a dispassionate reading of trade literature. In response to more stringent regulation, three-quarters of companies no longer manage their KOLs through marketing divisions, but through ‘medical affairs’ or dedicated thought leader development departments (with up to 45 staff). However they still classify around two-thirds of KOLs as ‘promotional’ rather than clinical, and the majority of pharma’s KOL expenses go towards promotional activities, consistent with pharma observing the letter rather than the spirit of regulation (Cutting Edge Information, 2009). Drug companies are advised how to manage ‘productive relationships’ with KOLs in the face of ‘increasingly restrictive guidelines and demands for transparency’ and how to ‘recruit thought leaders’ and ‘match the right thought leader with the right activity and the right price’. KOLs are ‘segmented’ into tiers according to their value to the company and assessed for their ‘return on investment’. Most prized is the ‘expert-in-field’, where the ‘opinions of a small number of prominent thought leaders proves the dominant weight on … the adoption and usage pattern of a product’, but pharma also invests in those whose influence is more localised (Cutting Edge Information, 2009).

KOL management is subtle, but there is little doubt that well-resourced pharma calls the tune. It is in the company’s best interests to ‘maintain the illusion of the key opinion leaders’ independence and integrity’, so that some KOLs appear to remain ‘genuinely unaware of the marketing message they are disseminating’, particularly if there is a synergy between their pre-existing views and the needs of the company. Marketers are aware that they may get value from a KOL who ‘does not always jump at a new drug’ and who therefore ‘may look more objective to the public’ (Cutting Edge Information, 2009). Doctors and pharmacists called medical science liaisons (MSLs) who can ‘talk science’ with KOLs help to establish ‘KOL ownership’, undertake KOL ‘identification and relationship building’, brief KOLs and question them about their activity with competitors (Cutting Edge Information, 2010). Surveys suggest that each month, a KOL can expect to have at least four face-to-face and 15 phone or email contacts from a company (Cutting Edge Information, 2009).

It is claimed that individuals become KOLs because of their excellent research and clinical work, and then enter into partnerships with pharma to promote approaches that benefit patients (Buckwell, 2008). This narrative ignores the role played by drug company patronage in establishing and enhancing an expert’s profile. Industry identifies early career academic physicians for potential status within the profession (BioMedical Insights Inc., 2012), positive impact on drug sales (KOL, 2012; Moynihan, 2011) and opinions that ‘align with marketing messages’ (Fugh-Berman, 2008). Pharma recognises that these doctors are more often motivated by career advancement than personal financial gain (Cutting Edge Information, 2009). Research support is therefore very attractive, offering the KOL career-enhancing access to funding and publications in an environment where attracting public grants is very difficult, especially for early career academics. Simultaneously, investigator-initiated research yields pharma ‘publication in a medical journal to advance a drug’s – and the company’s – profile and credibility’ (Cutting Edge Information, 2011).

Similarly, funding to kick-start new and innovative professional organisations benefits both pharma and the KOL (International Early Psychosis Association, 2002). Pharma also helps to build networks around KOLs, facilitating access to editorial appointments, scientific committees, and government advisory roles, so that KOLs become influential over policy, service and research funding, and the dissemination of data and opinion.

What we know about KOLs has important implications for young psychiatrists. Engagement with industry will offer them career advancement, status and a voice. But doctors who partner with pharma are likely to have less control over their relationships than they perceive; while made to feel important and valued, they will be intensively scrutinised by the companies retaining them. KOLs may not believe that they are biased by their affiliations with industry; doctors often regard themselves as immune from drug company influence even as it affects their behaviour (Dana and Loewenstein, 2003). What may be a mutually beneficial relationship for doctors and industry can harm patients.

And in assessing what is said and written by KOLs, even when there is no apparent conflict in relation to that specific presentation, doctors should be wary of the background influence of the relationships, past and present, between the author and the promotional departments of various pharmaceutical companies.