The GP informed the Powells that he would refer Robbie back to hospital immediately that day but did not inform them that Addison's disease had been suspected.

The referral letter was not typed until after Robbie had already died and was backdated to the day following the consultation.

In a statement after Robbie's death this GP stated:

"An Addisonian crisis is precipitated by an intercurrent illness and the stress it induces."

Dyfed-Powys Police investigated Robbie's death between 1994 and 1996 but asserted, supported by the Crown prosecution Service in Wales, that there was no evidence of crimes committed by the GPs who, incidentally, were retained by this police force as police surgeons.

Following a complaint by Will Powell (Robbie's father) in 1998 against the Deputy Chief Constable of Dyfed-Powys Police, regarding the inadequacies of the criminal investigation, a second criminal investigation was agreed, which commenced in January 1999.

As with the first criminal investigation, there was a gross failure to adequately investigate the criminality of the doctors. This resulted in Will Powell making a formal complaint against the Chief Constable of Dyfed-Powys Police in late 1999.

This complaint against the Chief Constable resulted in Dyfed-Powys Police appointing an outside police force to review Robbie's case in 2000. Detective Chief Inspector Robert Poole [DCI Poole] from West Midlands Police was appointed.

DCI Poole’s investigation report, entitled 'Operation Radiance', which was based on the documents provided to Dyfed Powys Police in March 1994, by Will Powell and his solicitor, was submitted to CPS York in March 2002. This report put forward 35 suggested criminal charges against five GPs and their medical secretary. The listed charges were:

• gross negligence manslaughter
• forgery
• attempting to pervert the course of justice
• conspiracy to pervert the course of justice.

DCI Poole's investigation also resulted in a disciplinary inquiry by Avon & Somerset Constabulary into Will Powell's allegations of misconduct against Dyfed-Powys Police officers with regards to their two inept criminal investigations between 1994 and 2000. Dyfed-Powys Police was found to have been 'institutionally incompetent' but no police officer was made accountable.

In April 2003, Will Powell met representatives from the CPS in London, who accepted there was sufficient evidence to prosecute two GPs and their secretary for forgery and perverting the course of justice. However, they would not prosecuted because of (1) the passage of time, which was caused by a decade of cover ups between 1990 and the appointment of DCI Poole in 2000, (2) Dyfed Powys Police had provided the GPs with a letter of immunity, and (3) the available evidence had been initially overlooked by the police and the CPS, between 1994 and 2000, for a variety of reasons. 

Following a 2013 adjournment debate, in the House of Commons, the Director of Public Prosecutions subsequently agreed, in October 2014, that there would be an independent review of the decisions made by Crown Prosecution Service, in 2003, not to prosecute, when there was sufficient evidence to do so. The reviewing Queen's Counsels have been provided with a report, written by myself ( a healthcare IT professional, former head of IT in an NHS trust and clinician) on major anomalies in Robbie's Morriston Hospital computerised records, which were erased during the first criminal investigation between 1994 and 1996. The review has not been concluded six years on.

The letter below (and also attached) from the English and Welsh Ombudsman was sent on 10 November 2020 sets out the case for a Public Inquiry.

Datix as a company has seen many changes since I stood down as chief executive in 2015. The most noticeable is a change of name to RLDatix, reflecting the acquisition in 2018 of Canadian rival RL Solutions. Some things, however, have not changed. Healthcare professionals still complain about the length and complexity of the Datix forms. They still complain about the lack of action from the incident reports they submit. They still complain about getting into trouble as a result of reporting an incident themselves (particularly reports about staffing levels). And they still complain about the threat of someone else including them in an incident report as a means of coercion: “If you don’t do this, I’m going to Datix you.” All of these factors are also common to incident reporting systems from other suppliers, but because Datix has the lion’s share of the UK market, they have contributed to an overwhelmingly negative sentiment about Datix.

The issues

The problem with complicated and contradictory forms is that Datix gives local administrators complete freedom to design the forms themselves. This results in forms that get longer and longer over time, as new people need to collect new information. The best forms I’ve seen are very short and contain the date, the time, the reporter’s details and free text boxes for a description of what happened and what action was taken. The very best forms I’ve seen have an additional free text box: “Your safety ideas”, asking the reporter if they can think of any ways that this type of incident could be avoided or mitigated in the future. It’s a good way to encourage people to think about safety; however, it does rely on someone at the other end of the report actually listening and responding.

The issue with the lack of feedback is that it relies on someone following up, investigating and then reporting back on the incident. Or if the incident isn’t going to be investigated, the reporter should be sent an explanation. If reporters don’t get any feedback and can’t see any changes made as a result of reporting, they’re going to stop reporting. This is not a problem with the incident reporting software, but an issue of the system within which it is used.

The issue of the threat and fear of reporting is more deep-seated and harder to change. It’s partly linked to the other two issues – if incident reporting has no positive outcomes, it’s seen only as a burden and a tool for punishment. It’s also a symptom of a culture of fear, bullying and a lack of resources, where stressed managers want to discourage the reporting of incidents as they don’t have the time or resources to do anything about them. There are constant calls for culture change. But culture change is difficult and it’s hard to know where to start. We can, however, take incremental actions that contribute to a shift in culture.

Culture change

One example is the former Calgary Health Region in Canada, which had a culture where incident reporting was being used for performance management, with managers reprimanding staff who reported incidents. Recognising this was having a bad effect on staff and patients, Calgary Health Region reconfigured Datix so that the managers couldn’t see information that would identify the reporters. This didn’t change the culture overnight, but it gave staff confidence that they could report incidents in an environment free from punishment. Coupled with the setting up of a separate central department responsible for safety and investigations, this set the organisation on the long road to culture change. An excellent write up of the system that Calgary implemented can be found here.

 Would that system work here in the NHS? Yes it would help, but it doesn’t go far enough in a system where incident reporting has got such a bad name. We need something much more radical. What if we were to abolish incident reporting completely?

Automated incident reporting systems

This doesn’t mean we have to remove investigation and learning from the patient safety toolkit. It does mean that we can obtain information about incidents from places other than manually input incident report forms. The technology already exists to do this. We can monitor a hospital’s IT systems in real time to see if an incident had happened or for signs that an incident was about to happen. There would be no need to replace existing incident management systems, just the method of getting the incidents into the systems and a change to the processes around them.

Such an automated incident reporting system already exists – again, in Canada – at The Ottawa Hospital. The hospital devised rules, called e-triggers, that automatically create an incident record based on certain criteria in other hospital IT systems. One such trigger might be a return to the emergency department within three days. The creation of the incident record also sends a notification to a clinician to review the record and answer some simple questions to determine if a follow up or investigation is needed. You can read some of the results from the system in this BMJ Quality & Safety paper. 

Although they haven’t done away with incident forms completely, this is a step in the right direction. I don’t know of anyone who has done anything similar here in the NHS, but I believe this system would go a long way towards the goal of eliminating the threat of “I’m going to Datix you”.

A call to action

• Set up triggers to automatically send potential incidents from other IT systems into existing patient safety reporting systems. Software suppliers should take the lead on this.
• Simplify current incident report forms so they are as quick as possible to complete.
• Give clear guidance on what incident reporting should and should not be used for, with assurances that no one will get into trouble for reporting an incident or being included in an incident report.

Do you have any ideas on how we can improve incident reporting and prevent the threat of “I’m going to Datix you”? Please join the discussion on the hub.

Further reading:

• Patient Safety Spotlight Interview with Jonathan Hazan, Chair of Patient Safety Learning
• Nursing can help end the travesty of ‘Datix abuse’ 
• Why Investigate? Blog series
• Root cause analysis: Why we need to change the focus
• Patient Safety Incident Response Framework

I use Robbie's story in all of my teaching on ethics and clinical governance.

The report is in response to a member of the public raising concerns when visiting a patient at home in the ‘clinically extremely vulnerable’ category. They noted that while they were visited by district nurses in PPE, their care workers did not wear this, advising that this was not required. The patient later died, and their death was confirmed as COVID-19 related.

HSIB found that during April the guidance made available by Public Health England for care staff in this regard was inconsistent. While the primary guidance did not refer to the need to wear PPE when visiting ‘clinically extremely vulnerable’ individuals, other guidance issued in the same month did set out these additional safety provisions. As a result of this, multiple versions of the guidance were available to care workers, who would not be aware of the PPE requirements if they referred to the earlier version of this.

Speed of the response

HSIB state that they brought this to the attention of Public Health England on the 28 April 2020. They subsequently replaced the primary guidance with a link to a version with the additional PPE provisions on the 13 May.

Given the importance of clarity on infection control and PPE, it is very concerning that the conflicting guidance remained live on the gov.uk website for a further two weeks after the issue was identified.

A wider system issue

The report acknowledges the complexity of providing and keeping up to date such a wide range of guidance, particularly in a crisis scenario, noting that this creates “a risk that patient safety issues may be missed”.[2] When considering the learning potential of this case, HSIB suggest that “there is an opportunity to introduce a document management system for guidelines to ensure that the latest information is available”.[3]

While this specific issue is now resolved, it is disappointing that there is no wider recommendation relating to the systems risks above identified by HSIB. Patient Safety Learning believes that there should be an additional recommendation on this that clearly identifies the relevant healthcare bodies responsible for looking into this.

There are also questions about how updated guidance is published and shared. Commenting on this in The Independent, Jane Townson, Chief Executive of the UK Homecare Association, mentioned problems with guidance being updated late at night with little notice.[4] She also stated that “there was a very high risk that care providers were not alerted to the changes unless they belonged to a membership association”.[5]

Who can lead this change?

While we have noted Public Health England's specific role in this case, formulating this type of guidance can involve a number of bodies across the UK, such as:

• Department of Health and Social Care
• NHS England and NHS Improvement
• Public Health England
• Public Health Wales
• Public Health Agency Northern Ireland
• Health Protection Scotland

When system-wide patient safety issues arise all these organisations have a role to play. We know that when it comes to implementing changes the system is “confused and complex, with no clear understanding of how it is organised and who is responsible for what”.[6]

Patient Safety learning believes it is vital that there is a clear approach to addressing such underlying safety issues. We need to ensure that learning and recommendations for change are prioritised and implemented widely across the health and social care system.

References

1. Healthcare Safety Investigation Branch, National Intelligence Report: Personal protective equipment (PPE): care workers delivering homecare during the COVID-19 response, August 2020.
2. Ibid.
3. Ibid.
4. The Independent, Coronavirus: ‘Confusing’ advice from Public Health England put patients at risk, watchdog says, 26 August 2020.
5. Ibid.
6. Care Quality Commission, Opening the door to change: NHS safety culture and the need for transformation, 2018.

Know what you’re writing about. You should be given a clear instruction or question in writing. If you haven’t been given this, ask for it.

Consider if you’re at risk. If your conduct or practice is being questioned by your employer or agency, then – provided you were a member at the time of the incident – use the RCN’s statement checking service accessed via RCN Direct on 0345 772 6100. If you’re being asked to provide a statement purely as a witness, and you don’t believe there is any risk to you, simply follow the RCN guidance (link below).

Be clear. Your statement should explain events from start to finish as clearly and simply as possible. Explain when things happened, who was there, and what you did, saw and heard. Try to avoid offering an opinion not based on facts.

Be relevant. Do your best to answer the question or allegation you have been set. If you can’t remember something, say so. Very few people can perfectly recall every event that’s ever happened to them.

Be compliant. If you’re a registered nurse, follow the National Midwifery Council's Code of Conduct, particularly the ‘Promote professionalism and trust’ section. Ensure you follow your employer’s local policies and confidentiality guidelines too.

List all documents referenced in your statement. If possible, state where to find them.

Format your statement. Add page and paragraph numbers, double space your lines and ensure pages have clear wide margins at each side.

Check it. Review each paragraph carefully, checking that your statement only communicates exactly what was asked for or required. Look at whether you can provide evidence for the facts stated. Check the facts you provide are clearly and objectively explained.

Keep a copy. You may need to refer to it in the future. 

The RCN’s statement writing guidance covers these tips in more detail, has a statement writing template you can use, and provides guidance on what to do if you are asked for a statement in other contexts such as if a coroner or the police ask you for a statement.

Follow the link below for more information.