Dismissal for Misconduct: To what extent can past conduct be taken into account?

When carrying out an investigation into alleged misconduct by an employee, the temptation may be to find as much evidence as possible. A recent case in the Employment Appeal Tribunal (EAT) has addressed the question of whether an investigation can actually be too thorough.

How “fairness” is tested

“The Burchell Test”, as set down in British Home Stores Ltd, v Burchell [1978] IRLR 379, is used as the starting point in unfair dismissal cases to determine if the dismissal of an employee for misconduct was fair or not. This three-limbed test states that a dismissal for misconduct will be unfair under s.98 Employment Rights Act 1996 (“ERA”) unless:

  • The employer believed the employee was guilty of such misconduct;
  • The employer had reasonable grounds for believing the employee was guilty; and
  • At the time the belief was held, the employer had carried out a reasonable investigation.

Additionally, the later case of Iceland Frozen Foods Ltd v Jones [1982] IRLR 439 stated that the tribunal will also have to consider whether the decision to dismiss passed what has now become known as the “range of reasonable responses” test i.e. the responses that a reasonable employer may have adopted in those particular circumstances. This test applies to both the decision to dismiss and the process followed.

Factors that will be relevant

When considering the determining factors to either dismiss or retain an employee, Thomson v Diosynth Ltd [2006] CSIH 5 stated that a “spent” (i.e. expired) disciplinary warning should not be considered in the decision making process. However, the Court of Appeal in Airbus (UK) Ltd v Webb [2008] IRLR 309 took the alternative view that s.98 ERA 1996 required a more “flexible” approach, stating that Diosynth did not specifically conclude that a tribunal could never consider previous conduct which resulted in the issue of a now expired warning.

The tribunal has recently had to consider this balancing act again in the case of NHS 24 v Pillar [2017] UKEATS/0005/16 which was reported in October 2017.

So, to which side did the tribunal lean on this occasion?

Pillar (“P”) was employed as a Nurse Practitioner by NHS 24, and was responsible for speaking to the public over the telephone and triaging them by asking appropriate questions to determine their medical emergency and priority. In August 2010 and July 2012, P had been found responsible for two Patient Safety Incidents (“PSIs”), neither of which led to formal disciplinary action but was dealt with by providing additional training to P as well as creating a development plan.

In December 2013, P again failed to ask appropriate questions which led to an incorrect referral of a heart attack patient to attend an out-of-hours GP clinic as opposed to calling 999. As a result, P was dismissed for gross misconduct. Details of the two previous PSIs had been included in the investigation report for the purposes of the disciplinary hearing.

P claimed unfair dismissal, arguing that it was not fair for the previous PSIs to have been included in the disciplinary hearing report because they had not led to disciplinary action at the time.

This was therefore the ultimate question: Should P’s previous conduct have been included in the investigation for dismissal?


The tribunal held that NHS 24 were entitled to treat the PSI in December 2013 as gross misconduct as there was a significant risk to patients, and that the decision to dismiss was in itself a reasonable one on the basis of the material before the decision-maker.

However, the tribunal concluded that the overall dismissal was unfair on the basis that it was unreasonable for the investigation to have included the details of the two previous PSIs in the report, and that it would have been sufficient for details of the training P received to have been included instead because the previous incidents did not lead to disciplinary action. P also argued that she had not been made aware at the time of the first two PSIs that further similar incidents would likely to be viewed as gross misconduct.

The tribunal’s view was that the dismissal had been procedurally unfair, with a lack of transparency as to what was being considered by NHS 24 in the lead up to the dismissal. Nonetheless, the tribunal reduced P’s compensation by 70% due to her contributory conduct.


NHS 24 appealed to the EAT who substituted a decision that the dismissal had in fact been fair. It was concluded that the previous warnings given to P were relevant to the dismissal. The EAT stated that the Burchell test in relation to the investigation stage was one of sufficiency, and that she was unaware of any case where it had been argued that an investigation failed this test because it included too much information. The EAT also reiterated that if the officer had only been provided information about the training P received previously, the decision would be lacking in the full context and background as to risks to patient safety.

The issue of fairness was also addressed, with the EAT acknowledging that this was a contentious point when taking into account past misconduct. However, it was stated that P had not actually challenged the decision that the dismissal had been fair on the basis of the material provided to the investigating officer. The EAT therefore held that there was no rational basis to exclude details of the previous PSIs from the report and it had been inconsistent of the tribunal to find that it should have been. It was therefore “perverse” to conclude that the inclusion of the material rendered the dismissal unfair.

The EAT also addressed P’s argument that if, as in Diosynth, an expired warning cannot be a determining factor in the decision, then surely conduct not treated as a disciplinary matter could never be considered such a factor too. The EAT distinguished that the expiry of the warning in Diosynth gave the employee a “false expectation” that it would no longer be a determining factor in future disciplinary matters, which made the subsequent dismissal in reliance on that warning unfair. However, in Pillar, no such expectation had been created as to whether the earlier PSIs would be relevant any future investigations so this argument had no basis.

Overall, the EAT substituted a finding that the dismissal by reason of gross misconduct was fair.


Pillar is a useful case as it clarifies the extent to which past conduct can be taken into account when an employer is deciding whether or not to dismiss an employee for misconduct. The EAT highlighted that it is always for the investigator to put together all relevant information and for the investigating officer to decide what to do with it. When a tribunal is assessing whether a dismissal was within the range of reasonable responses, it is the decision-maker’s state of mind that should be considered. The reasonableness of an investigation is relevant only where it results in an absence of proper information being put forward to the decision-maker.

The case also demonstrates the tricky interplay between procedural and substantive fairness in unfair dismissal cases, and provides some comfort to employers that tribunals will be reluctant to criticise a thorough investigation into gross misconduct. Finally, the case highlights the crucial point that including previous employee issues in any future investigation reports will not automatically render a dismissal of that employee unfair.

This article was written by Clare Gilroy-Scott, Partner, Employment, with assistance from Becky Minear, Trainee Solicitor.

This guide is for general information and interest only and should not be relied upon as providing specific legal advice. If you require any further information about the issues raised in this article please contact the author or call 0207 404 0606 and ask to speak to your usual Goodman Derrick contact.