closes on https://www.gov.uk/government/consultations/the-disclosure-of-industry-payments-to-the-healthcare-sector/disclosure-of-industry-payments-to-the-healthcare-sector#how-to-respond – CLOSES 16th OCT 23

-We agree that information about payments and benefits from “healthcare provider organisations and organisations connected to the provision of healthcare” should be publicly available. 

-However the document concentrates on pharmaceutical companies. It is important to consider the potential for conflicts to arise in numerous other domains. For example, conflicts may arise through involvement with technology companies, apps, as well in sponsorship deals with healthcare professionals who are eg social media influencers. Additionally, doctors who benefit from driving more patients to their private practice through increasing diagnosis within their speciality may also have a conflict in the recommendations they make. Therefore we agree, but the definitions of what benefits are declared requires to be clearer and wider. We think it should reference all sources of income in health related and commercial activity. 

-We agree that publishing payment information on websites in a prominent position is desirable. However 6-7% of people are not online. An estimated 7 million people have a reading level of or below a 9 year old. Further, a simple declaration is unlikely to give the information needed to understand the impact of a it, far less the options to respond to it. We will expand on this below. Therefore a declaration should not be treated as all that is necessary to ensure transparency.  

-The government should therefore consult particularly with disadvantaged groups with regard to online access and literacy in order to ensure that these proposals lead to useful information for all groups. There is a balance between a burden on declarers and a need for timely transparency. We suggest that large transfers of value should be prioritised for earlier reporting than one year. We suggest that a central register could be maintained but where it is possible to update information at any time. This could be used by other organisations to extract information they require as a declaration at any point, time-stamped. It could be used therefore in multiple venues eg academic journals, NHS and private  practice, etc.

-We agree that this information is likely to be useful. However it is uncertain as to whether this will be helpful to patients unless it is designed to enable useful transparency. We would like to draw your attention to two recent pieces of research examining UK systems for professionals declaring conflicts. These are: 

a) McCartney M, Bergeron Hartman R, Feldman H, et alHow are declarations of interest working? A cross-sectional study in declarations of interest in healthcare practice in Scotland and England in 2020/2021BMJ Open 2022;12:e065365. doi: 10.1136/bmjopen-2022-065365 https://bmjopen.bmj.com/content/12/11/e065365

This is an analysis of whether the current declarations being made by healthcare professionals in the UK are fit for purpose. The proformas created by the Grant review were not always in use in England, and 76% were incomplete by design, difficult to find, and often did not state what actions would be taken in respect of a declared conflict. In Scotland, only 14% of boards published staff registers of interest. 

b) McCartney M, Metsis K, MacDonald R, et al‘You feel like you’ve been duped’: is the current system for health professionals declaring potential conflicts of interest in the UK fit for purpose? A mixed methods studyBMJ Open 2023;13:e072996. doi: 10.1136/bmjopen-2023-072996

https://bmjopen.bmj.com/content/13/7/e072996.full

This study asked students, professionals and lay people about doctors’ conflicts of interest: a large majority thought they were important and wanted to know about them. We then asked them to find ‘famous’ working UK doctors’ declarations of interests. Only 3% of participants found these full declarations, and 36% found some relevant information. Therefore, even if doctors are making declarations of interest, they are not transparent because they are difficult to find.

Our linked research also found that declarations of interest are felt to be important by patients, want to know about them. But, even when doctors are making declarations, they are difficult to find, and often incomplete by design.  

This means that the current system for making declarations is not working well enough to enable effective transparency. Further, even if declarations are transparent, we do not know the best way to describe doctors’ conflicts, to allow patients to judge the validity of that doctors’ recommendations. 

-The types of data that are being proposed to be drawn are not substantively different from those already recommended by the Grant review (NHS England, 2016). Therefore it is difficult to see how this will improve the ability of people to understand the potential impacts of conflicts which are declared. 

-Regarding the Disclosure UK website: The resources present (MM is one of the panelists on the videos supplied by Disclosure UK) are outdated and otherwise the information provided has all been written by Disclosure UK. This is biased. For example, the information available on the information sheet ‘How we work with HCPs’ names multiple advantages but raises none of the issues of biases (for example the potential for conflicts to cause bias). Nor is there any assistance for readers in interpreting what the meaning of a declaration might be. Information should be independently written with fair expression of benefits and harms associated with payments. 

Compliance standards: 

-Declarations should be held indefinitely. This is because there is no evidence that the potential for bias decreases according to time. In fact, long term patterns of industry engagement may be helpful for understanding the context of a declaration. 

-There is evidence that voluntary declarations are incomplete, as per our research cited above. We do not know if statutory declarations will improve the quality of declarations. They may improve the amount of declarations made but not necessarily the quality. Therefore, assessments of compliance need to reflect meaningful compliance – i.e whether the information is useful rather than just a ‘data dump’. 

-Sampling for compliance, with specific investigation of concerns, may be appropriate but we suggest that this needs to be tested in real life to ensure effectiveness. The monitor should be independent. 

-Reports of non compliance should be possible by any person and through multiple routes. This is because a variety of people may have information of relevance, and because not everyone has online access. Further, some people may wish to report information anonymously. 

-In terms of triggers to determine the need to investigate a case, this seems to be a question asking what triggers should be ignored. It is not clear which would be safe to do so. 

There is no evidence as yet generated from the UK whether financial penalties will result in more compliance or not. Other sanctions (such as restrictions on professional practice) may be more effective. Research is needed to understand what the most effective power would be. 

-There is no threshold where a payment can be said to be non-significant. Even small items such as pens, notepads, food items have measurable impacts on behaviours. Therefore we do not agree with a threshold. 

 -In terms of commercial sensitivity, unfortunately, this can be used as a means to obscure essential information. There will already be a delay on reporting of transfers of value to a register. It is undetermined as to whether this concern represents a tangible commercial risk. If this is allowed, we suggest that early assessment and interrogation of cases are completed independently and the decision thereafter reviewed. 

-The justification for not providing names of professionals receiving education funded by industry is lacking. This is a benefit-in-kind. Education sponsored by industry has been associated with poorer quality information and bias. We have seen examples of our reporting in our research where blanket statements have been made about educational sessions run by industry for departments in hospital, for example, but without any information necessary to interpret this. It is our view that professionals should declare sponsored educational sessions they attend. Industry routinely collects information about the individuals who attend. Other examples are of clinical NHS posts which are funded by industry. This information should be clear to other professionals and patients. 

-Voluntary compliance with government issued guidance does not work, as we have shown in our published research cited above. However we do not know why this is. For example, is the burden of paperwork too great, or difficult to follow, or inefficient? Does it depend on factors outwith the professionals’ control? Is the guidance hard to follow? Do professionals have the time and resources to complete the forms? Without research we will not know why completion of these forms is poor  and fail to offer better solutions. Voluntary publication by any agency is likely to result in gaps in information. 

-Current reporting systems are chaotic and compliance is generally poor as we have demonstrated. However professionals can be asked to complete disclosure forms on multiple platforms, eg different NHS organisations, for presentations, academic publications, roles at Colleges, etc. This makes it harder for individuals to locate a declaration (because it may, or may not, be in multiple venues). Rather than having multiple venues for declarations it would make more sense to have a single, complete register at a single location that other organisations could refer to. 
-In terms of impact on professional-patient relationships, o

ur research has found that patients think conflicts are important and want to know about them. Our research found that patients thought they would trust professionals with financial conflicts less than professionals without financial conflicts. 

-However research from the US has suggested that patients do not change their intended actions where financial conflicts are known to be present. We suggest that we should be cautious and consider this essentially unknown in practice. We suggest that research should be done to understand this as trust is an essential part of medical practice.  We need to understand whether a gain, or loss in trust is appropriate, or not, and what actions patients are able to have should they think that a health professionals advice is less trustworthy. 

-In terms of the potential for proposals to increase impartial decision making, this is unknown and is likely continue to be so unless specific research is commissioned to test and analyse proposals for declarations. It is possible that a large amount of work becomes wasted if declarations are poorly completed or not done in a way that allows for their effective management. We urge the government to consider this problem in a wider context, ie to what purpose are declarations to be put to? Transparency alone will not solve the problem that the IMMDS review identified, ie that financial conflicts endangered patients. Rather, this problem requires a series of steps, of which transparency can be only one. We would be happy to meet with you and explain our research and the gaps in knowledge in this area. There is a risk that much effort goes in to improving transparency but that the information made available by transparency cannot be used effectively to reduce harms. We suggest that the government considers commissioning independent academic research capable of filling these gaps. 

Conclusion  

We support the need for transparency in healthcare. However changes need to be made to ensure there is ‘better’ transparency, not just ‘more’ transparency. There is a risk that poor (though compliant) declarations may make the situation worse, not better, particularly if the information given to citizens about the nature of transfers of value is biased or incomplete. There are numerous uncertainties. However, while the GMC has emphasised the need for doctors to be transparent, there is a lack of evidence finding an effective way to do this. Further, we do not know the best way to communicate the potential for bias to patients. Additionally, patients may not be able to make choices which allow them to try and remove the impact of a potential conflict. This again emphasises the GMC instruction to “avoid conflicts whenever possible”. 

In broader terms, we are concerned that doctors (including those who are trying to act honestly and with integrity) may not recognise the potential of conflicts to cause bias, and to identify bias when it occurs. This is a human condition. We doubt that it is possible to remove all potential conflicts and bias to create completely reliable insight. Instead we think it is better to try and build solutions which remove the potential for bias from the structures of healthcare as much as possible. 

We are not aware of standardised teaching in medical schools on conflicts of interest or postgraduate medical education on this, or indeed within the NHS for managers who are expected to understand and manage conflicts of interest. 

We think there is a need for high quality research in this area (we are currently applying for grants, however funding for this type of work appears scarce due to its’ nature). We suggest that commissioned research examining how declarations should be designed in order for them to meet needs is needed. This would help us understand 1)what is necessary to achieve transparency 2) what is necessary to make that transparency effective. This is important as transparency alone should not be used as an end-goal. Rather, transparency should be used as a means to achieve a reduction of the harms to people and healthcare systems caused by conflicts of interest. Transparency is necessary to manage conflicts, but without a clear plan as to how to interpret the information provided by transparency, it is difficult to see how it can be useful. Therefore research should not only focus on what is needed to produce transparency, but also the ways in which this information should be used to reduce its’ harms.