A review into the quality of NHS complaints investigations where serious or avoidable harm has been alleged

Annex D: The visits – summary


In April and May 2015, we visited six acute trusts4 across the country. These included smaller acute trusts, large trusts, trusts that had recently been in special measures, as well as trusts that had been performing well. We asked the trusts questions about how they investigate allegations of avoidable harm and how their complaints process is set up to investigate and learn from complaints. We spoke to a wide variety of staff, including directors of nursing, complaints managers, complaints staff, divisional leads, and governance leads.  

We used the information from these visits to validate or highlight gaps in the information that we obtained from the survey and the review. We also looked to find examples of good practice.

Below is an overview of the feedback we received from these six trust visits. 

What we found

We were made to feel welcome, and generally, trust staff spoke to us openly about the complaints process and their approach to investigating allegations of avoidable harm. The staff we spoke to were keen to improve the system.  

We have not quantified how many trusts provided certain responses. This is because we only spoke to six trusts and this, therefore, cannot be representative of all trusts. However, themes did emerge. Equally, the information we gathered helped validate the information we had already collected. 

The themes we looked at:

How the complaints teams and process is structured:

Often the complaints teams do not, structurally, sit with the governance teams, but within the nursing directorate. This means the governance and complaints systems run in parallel. The complaints teams tend to liaise with complainants and deal with minor complaints, but do not investigate patient safety incidents. Generally we found that the complaints teams sent complaints about patient safety incidents to the division where the complaint arose to be investigated by clinical staff within that division.  

However, one of the trusts we talked to was in the process of changing its approach, and its complaints team (who are lay people) will be investigating patient safety incidents. This is unless the complaint has already been reported on the relevant patient and risk management software (Datix) and investigated within the division.We did not find any consistency about who would be investigating the complaint, and the level of training of investigators. Some trusts had a list of trained investigators within the divisions. Other trusts did not necessarily use trained investigators, but said that incidents were investigated by ‘the appropriate person’.  

Another variation we found was that in some trusts a trained investigator would investigate a serious incident, but anyone could investigate a patient safety incident that did not meet the criteria of a serious incident. Trusts also told us that investigators did not necessarily have time in their working week to do the investigations, but had to do this in addition to their clinical or managerial workloads. 

The complaints staff we spoke to were all keen to resolve complaints and were persistent in following them through to the end. In some trusts, it appeared to be personalities and persistence that was improving the complaints and investigations process, rather than the investigations process itself. The majority of trusts were open in telling us that they did not feel they had a culture of openness.

Investigation process (patient safety incidents)

In general, we found that complaints staff speak to the complainants and agree the scope of the investigation, and then pass the investigation over to the division where the patient safety incident occurred. However, one trust was starting to use complaints staff to investigate patient safety incident that did not meet the criteria for serious incident.  Complaints teams generally told us that when they received a complaint about a patient safety incident they would cross reference it on the trust’s logging system - most commonly Datix - and if the incident was not already reported they would report it.  Different trusts said there were different levels of reporting of patient safety incidents by clinical staff on Datix before the complaint was raised.

The larger trusts told us that it can be difficult to obtain clinical records, whereas the smaller trusts found this less of a barrier. Trusts that had an electronic records system said they were better able to get access to clinical records. 

Some trusts relied on statements and did not interview staff because they said interviews were too difficult to arrange. Trusts also reported poor quality written statements and having to keep going back to the clinicians to get the information they needed.

Some trusts said that clinicians were unwilling to review their colleagues’ work, which made getting  an independent clinical review difficult. However, the majority of trusts could get clinical reviews from within the division where the incident occurred for patient safety incidents, and some sought reviews from different divisions for serious incidents, but there was no consistent approach to this. Trusts’ complaints staff reported difficulties in challenging clinical opinions.  Generally, external clinical reviews were only sought for serious incidents and larger trusts found it easier to get an independent clinical review. Trusts reported difficulty in obtaining independent clinical reviews where the speciality was rare and the number of clinicians working in that field at that trust was limited. 

It was generally reported that doctors were more unwilling or slower to provide opinions and statements than nurses. Trusts considered that where the complaint response was quality assured by staff not involved in the care, this introduced an element of independence. 

We found variation in whether trusts dealt with serious incidents, and patient safety incidents that did not meet the serious incident criteria, in the same way, or whether they approached them differently. 

Equally, we did not find consistency in how the investigations were approached. Some trusts had a root cause analysis (RCA)5 template that the investigators followed, and others simply said that the investigator would choose how to approach the investigation on a case-by-case basis.  The process and approach also differed between divisions within the same trust. Trusts generally expected the investigator to analyse the information and uncover why things went wrong.

We found that complaints teams tended to have a weekly meeting with the divisions where the complaint arose to discuss progress of outstanding investigations, and this helped the complaints team manage the process.


We found, in general, that divisional leads quality assured the investigation reports, which were then quality assured by various senior managers and the chief executive. We were told that when a lot of people were in the quality assurance chain the process was longer and harder. This is because staff would tailor the write-up of the investigation and/or response to suit an individual’s style, and it would then go to a different individual who would have a different personal preference about writing style. Trusts considered that the quality assurance chain introduced an element of independence. The complaints teams also quality assure responses before they are sent out and will query the complaint response if it does not answer the question, or is not written in plain English.
Trusts told us that complaints and patient safety incident/serious incident investigations were discussed at regular governance and senior management/board meetings. Trusts reported a move towards better identification of trends of where things are going wrong. Trusts reported that senior management gave complaints priority. Trusts also told us that governance and/or auditing of any changes that were implemented is an area that needs improvement.


Trusts reported that since the Duty of Candour requirements came out in November 2014 they inform patients more reliably about patient safety incidents. Trusts all reported that they have improved how they respond to complaints, and are aiming to give complainants clearer explanations of what happened and why. Trusts also reported that they explained, in their responses, what improvements had been put in place as a consequence of the complaint. Some trusts reported that local resolution meetings with complainants helped communication, and others said that written responses worked well.

Implementation and learning

The majority of trusts said that the investigator was responsible for drawing up action plans for learning from a complaint. Usually the heads of division will sign off an action plan once the investigator has drawn it up. Trusts told us that not all investigations (even upheld ones) resulted in an action plan.

Trusts also told us that an area they needed to work on was sharing with staff what had been learned from complaints and investigations. They said that patient safety incidents and investigations were discussed at high level governance meetings, and that learning was cascaded down through matrons to ward staff. However, there was inconsistency in how this translated into changes in delivering clinical care. 

Trusts also said that monitoring and auditing any changes was an area that needed improving, and there did not appear to be any robust processes in place to make sure this happened. Trusts said that the culture around learning from complaints and patient safety incidents needs to improve. Trusts also told us that it is difficult to achieve cross-divisional or trust-wide learning, as currently divisions appear to work as isolated units.

Serious incidents

Trusts did not have a consistent process to identify a serious incident. They told us that, often, these had not been reported before a complaint was raised. They also told us that clinicians in some trusts use their experience to ‘spot’ serious incidents, whereas other trusts had a central risk team that flagged serious incidents.
It is more likely that serious incidents are investigated by a trained RCA investigator who will use an RCA investigation template, but this is not guaranteed. Again there is no set process to investigate these complaints. Some trusts follow the same approach for patient safety incidents and serious incidents, and others do not.


Trusts told us that the barriers they face are:

  • Difficulty getting access to clinical records;
  • Problems contacting staff who have moved;
  • The use of temporary staff, which makes it harder to identify and track people;
  • The challenging pace and scale of work;
  • Poor interpretation of the available evidence;
  • Lack of a system for learning from complaints;
  • Lack of a culture of openness; and
  • A culture where doctors who do not accept it when complaints staff and investigators challenge them about their statements or reviews.

Areas for improvement

Trust staff suggested these areas for improvement:  

  • Create a check list for complaints team to help them identify if a complaint should be reported as a serious incident.
  • Train complaints staff in investigation skills.
  • Standardise processes for investigating patient safety incidents that do not meet the serious incident criteria, and use of an RCA template, irrespective of whether the issue was raised by a health professional or as a complaint.
  • Better collaboration across the divisions when investigating and learning from patient safety incidents and complaints.
  • Better ownership and dissemination of learning and action.
  • More resources, including appropriately trained staff.
  • Better consistency and quality of investigation reports.
  • Better and more consistent monitoring of the effectiveness of action plans/change.
  • More thorough, but not unnecessarily cumbersome, quality assurance processes.
  • Senior acceptance of changing culture in respect of openness.
  • Buddying system with different trusts for clinical reviews.
  • Cross trust learning methods such as the National Patient Safety Agency (NPSA) or Medicines and Healthcare
  • Products Regulatory Agency (MHRA)6  alerts could help share learning across the country.
  • Creation of a pool of national clinical advisers to review cases.
  • More consistent national guidelines (we were told that the new serious incident guidelines are cumbersome).

4 The trusts provided information anonymously. 

5 A methodology in which steps are taken to identify, and tackle, the root causes of any errors or failings identified as the result of an investigation, in seeking to prevent them from recurring.

6 The MHRA regulates medicines, medical devices and bloods for transfusions in the UK.