Evidence Based Policing

The creation of a professional college for policing – part of the professionalisation agenda – is an opportunity to work towards becoming evidence based. This is something that has been made clear to me over the last two months and something I’ve been arguing for years, incidentally! We can look around and see that the College and individual police forces are trying to engage academics to bring research skills into policing and actively build that evidence base. West Midlands Police and the Metropolitan Police are just two forces hosting research fairs and inviting academics into their organisation with the aim of understanding what works in policing – and why?

Evidence is key to this – being the police, we should be concerned about the standard and quality of our evidence, shouldn’t we?! We are in criminal trials – why haven’t we been overly concerned with it terms of evaluating interventions? We see examples all around us where evidence is not being brought to bear on the claims we make – again and again, I see this in policing and mental health. For example, that the Centre for Mental Health, in writing various documents about liaison and diversion services, has said, “the evidence for liaison and diversion is just not there.” It doesn’t mean there is evidence of it not ‘working’ – whatever that means – it just means we haven’t (yet) gathered the data to show it works. Let me show you another example of where we’re not relying upon evidence.

I have repeatedly heard in the last two months that ‘street triage is saving officer time’ and other claims implying the same. You can see for yourself in the recent HMIC report ‘Core Business’ (2014, p116) – there is clear wording that the introduction of street triage has seen average assessment times for 136 fall ‘from eight hours to five’. We see the same report imply that street triage has reduced the time spent dealing with the consequences of section 136 detentions and an example is given of a force where over 13,000 hours of time was spent in 2013 waiting for assessments. (This amount of time has more than doubled in two years, incidentally – so we can all agree that the force concerned will want to do something about the ever-increasing tacit consumption of their resources by the mental health system.)

So does it actually save time? – and is that actually the point of it?!


Let me firstly suggest that the evidence you would want to gather around you concerning street triage will be highly dependent upon what you see as the point of it all. If you’re wondering about the time it saves you will need to know certain data; if there are other objectives, you’ll need other data. To test the proposition that it saves time, you would look to gather proper evidence about time spent on s136 and time spent undertaking new ‘street triage’ approaches.

  • Usage of section 136 prior to street triage
  • Usage of section 136 after the introduction of triage
  • Data about the average time spent by officers waiting for assessments
  • Data about the average time spent by officer assisting with detention and conveyance for those patients who are subsequently admitted to hospital after assessment
  • Data about time spent talking people home who were not admitted to hospital after the use of section 136.
  • Data about the time taken to deal with those incidents that previously would have involved use of section 136 MHA but will now be handled differently because of street triage – these jobs don’t go away, they are just handled differently and that takes time, too!
  • A clear understanding prior to the introduction of ‘triage’ of your areas s136 usage – how much of it is appropriate / inappropriate (you may have to define this yourself!);
    – how much of it represents ‘failure demand’ that could and should have been prevented from reaching the police in the first place;
    – how much of it represents ‘value demand’ which was perfectly proper, necessary use of the power which can now be handled differently because of the ability to engage the NHS far earlier.
  • If you’re really doing it properly, you may need to have a control area – where 136 usage and health funding or infrastructure is similar; where demographic and epidemiological data is similar; and where police training and resourcing are similar.

Then – and only then! – can you start to work out what impact street triage has had. The factors mentioned so far are only those you would need to start evaluating the impact on the police – and I’ve cut short the list of things you’d probably examine! I could go on and on if we start to think about the impact on health and social care. I repeat the point: you have to be asking yourself evaluation or research questions which address the issue “What are you trying to achieve by this scheme.” What is your objective or objectives?


This is really important to any evaluation: we need to be clear about what we’re actually trying to achieve because otherwise, how do you know whether you’ve succeeded? I heard recently that street triage is about ‘reducing the use of s136 – the end’. That’s fine – so if that’s what you’re trying to achieve, why are street triage schemes busy trying to do other things? Are they just over-resourced for their main purpose and kindly helping out with other extraneous ‘stuff’ or were they actually always there for a broader purpose? This question is really important and in no way flippant – because it affects the research and evaluation questions you ask yourself in relation to which you then draw on or seek out particular quantitative and / or qualitative data.

One other area told me “We’re trying to reduce the use of police cells.” Great – who could object to that? But please tell me why street triage is being done in areas that weren’t really using the cells anyway? It really must be about other, broader issues, otherwise areas like Birmingham wouldn’t be doing it at all – there was cell no usage to reduce! Reduction in the use of the power AND a reduction in reliance upon police cells? – still doesn’t explain why triage services then do things that achieve neither purpose and it doesn’t allow you to pose a proper research hypothesis that can be tested by proper inquiry.

If the activity of street triage schemes is undertaken somewhere other than in public streets, then it’s clear we must be trying to achieve something else – as well or instead. So what is it we are trying to do? That needs to be part of your evidence base as you assess these things. I admit that I personally have long since stopped judging schemes by what they say they are trying to do – I look instead at what they are actually spending their time doing.

Points about mental health street triage from various areas:, they are not responding in their multi-agency vehicle to most of the incidents they are contacted about; they are ‘dealing’ with incidents (often from afar) that are two-thirds of the time being hosted in people’s private homes; triage is seeking to avoid s136 usage but that was only ever a possibility in one-third of the incidents they attend and it seems to be occurring in areas regardless of whether the cells are being relied upon as a place of safety or not and irrespective of how that police force is perceived to be using the legislation in the first place.

So it’s about some else or something far broader, otherwise they wouldn’t be wasting their time on extraneous ‘stuff’. I’ll let you decide for yourself what you think it’s about!


What I do know is this: a police officer works for 2,080 hours a year (without working any overtime). So we can look at the various schemes and work out how many hours and pounds are being expended in providing it. (Let’s just stick with the police service, for now.) We can then look at the reductions in the use of the power and the use of the cells and work out the time and / or money saved. Let’s say, that a force with 1,000 detentions per year reduced s136 use by 40%, therefore avoiding 400 instances of two officers sitting in a place of safety pending assessment by an AMHP and a Doctor.

If the average wait was four hours per detainee after the half hour that the Royal College of Psychiatry standards suggest the police should spend handing over, we can then do the maths —

400 (avoided detentions) x 2 (police officers per detention) x 4 (hours spent waiting per detainee) = 3,200hrs saved. If your street triage scheme requires four full time equivalent police constables to deliver this, then you are expending 8,320 hours to achieve this. If you’re new improved arrangements also mean that you reduce the average wait from 4hrs to 3hrs, you can recalibrate that answer to suggest that 2,400hrs were saved, but that you also saved time for those detentions that still occured.

600 remaining detentions) x2 (police officers per detention) x 1 (hour saved by reduced waiting times) = 1,200hrs saved. So you can celebrate a total of 3,600hrs saved overall before moving on to ‘do the maths’ for the health and social care investment, versus saving and pose that back against your new operating model.

You could also work out the custody time saved if police cells were still being used. Where a street triage scheme brought about end of custody being used as a place of safety after previously having seen 250 detentions per year in the cells, averaging 10hrs each, you can calculate the implications and therefore the cost. Half an hour of a sergeant’s time saved per detainee booking them in, 5 minutes of a custody sergeant’s time for every subsequent hour in custody; one consultation by a police doctor; 5 minutes for every hour in custody for a detainee to be attended to by the custody assistants and 10hrs of someone undertaking level three or level four observations in custody. Forces model this time and cost different – for our purposes, I’m simply making the point that you could calculate it.


So if we’re trying to reduce s136 usage: Great, we’ve reduced it by 40% – job done. But we can’t claim in our hypothetical model that it has actually saved time. It is actually costing you time: because in order to save 3,600hrs overall, you are posting four FTE constables to the position and expending 8,320 hours of effort. This is a net loss totalling 4,720hrs – in other words it is more than two full time officers for the area where the scheme works. Well done.

Police time is FAR from being the only important thing in these issues, however – but the point I’m making is that we are claiming we need to be evidence based and claiming we save officer time when we actually seem to be spending it! But what price human dignity, less restrictive assessment options, faster responses to incidents of mental health crisis care by the health service? ……. ALL of these things are not easily measured in terms of hours and pounds so perhaps the emphasis should be on these positive outcomes and greater human dignity which is worth paying for (up to a point, given that budgets are finite).

I was talking to professionals today at a Crisis Care Concordat meeting asked them ‘How many people in your area ask for urgent help whilst in crisis?’ (accepting that ‘crisis’ isn’t defined.) No-one knew – it’s fair to say no-one knows in any area! But if we don’t know how many attempts are made to seek crisis support, how can we tell how many of these incidents were managed correctly, at the first time of asking? How can we evaluate whether crisis demand for the police today, is failure demand from attempts to access support via the CrisisTeam earlier today or yesterday. How many people who’ve attempted to access Accident & Emergency today, had attempted to access their GP or Community Mental Health Team before self-presenting?

No-one knows. So we probably need to take crisis care back even further and ask even more fundamental questions about what we’re trying to achieve before we can put street triage in it’s proper context and determine that it’s an answer to something – but only after it is properly evaluated and we have evidence, not anecdote!

IMG_0053IMG_0052Winner of the President’s Medal from
the Royal College of Psychiatrists.

Winner of the Mind Digital Media Award.


12 thoughts on “Evidence Based Policing

  1. Sorry, been thinking about this a lot but try to keep to two short points.
    1) Just read news reports on CQC report on s136. BBC news is saying people are being taken to cells rather than being treated. Not true, s136 does not result in ‘treatment’ it results in an assessment to see if you need to be admitted to hospital. If you are vanishingly lucky it may also result in recommendations to other services. So the debate on cells vs health based places of safety is a debate about where the detained person sits and waits. Unless they need emergency treatment that is all they will be doing. Treatment will only follow if they are admitted to hospital.
    2) My understanding is that if the police detain someone under s136 they are legally responsible for the safety of that person until the detention ends after an assessment. Certainly our local hospital place of safety doesn’t provide ‘safety’ to police standards for any but the most compliant detainee.
    3) In my experience of family members it can take years of engagement with MHS before receiving appropriate treatment. In that time crises result, if you are lucky in A&E or police interaction, if you are unlucky, in death. With appropriate treatment surely the aim should be that crises should be few and far between and intervention should be taking place before the police ever have to get involved. How many mental health ‘service’ users have a crisis plan which includes ‘call the police’ or ‘call an ambulance’? An interesting questions for the police to ask their mental health trusts….

    1. Regarding point 2) police have responsibility till another agency accepts that responsibility such as MH team or A&E medical team.

  2. Indeed,and the idea of what constitutes a “crisis” is key. Oncologists treat ppl with diagnosed cancer where treatment is available; historically MH services have treated ppl with severe mental illness (usually schizophrenia and severe bipolar). Now the expectation seems to be that everyone in distress is a candidate for treatment. When working as a therapist I am expected to determine if someone does actually have an emotional disorder or is merely sad, pissed off, overreliant on substances, antisocial etc. unfortunately in my other role in acute mental health care the growing expectation seems to be that everyone is “ill” and therefore treatable and that we are failing if they aren’t cured (bit like blaming the police when ppl keep reoffending ). Where is the evidence?

  3. You make an important wider point, too. The significance of how research is designed and shaped in addressing the important or salient questions. I remember being told years ago about how important were the questions I asked (as a manager). As you identify, their importance and the methods you adopt for answering them is not diminished whether you are applying them as a police investigator, a public inquiry member or a researcher

  4. This post was very helpful generally but especially to me as I am at the beginning of the fieldwork stage of an evaluation of mandated joint working in mental health services (which includes S136 and two other contexts – S117 and delayed transfers of care). I aim to use a realist evaluation approach to two case study sites, a shire county and a metropolitan authority area.
    This approach carries, among other things, an assumption that the circumstances in which an intervention takes place has a vital impact on how it works – and that therefore the intervention is likely to work differently dependent on those circumstances. So, I’m not sure that your idea of having a control area would be helpful. Although the ‘control area’ may not be using street triage, it may well be going about the operation of S136 in a different way as a result of the services in that area giving S136 a different priority or the ‘healthy’ (or otherwise) state of the all-important relationships between the police and health and social care staff.
    It would be very helpful to know what approach is being taken to evaluate the street triage pilots – hopefully it’s not just a statistical (numbers) approach but also how the triage scheme worked before during and after the pilots for all the key people involved.

  5. You mean policing hasn’t been evidence based! Ha Ha! What exactly are you saying here?!

    The police stated that someone diagnosed with borderline personality disorder, recurrent major depression and spent over a year in hospital didn’t suffer from mental health problems in a court of law. Very embarrassing for both the NHS (it was a witness statement from a DBT therapist) and the police.

    So embarrassing that the health trust wont acknowledge it. What would happen if the public found out?

    1. So it wasn’t the police who said they didn’t suffered from MH problems it was somebody else who provided evidence to that effect the the CPS presented it to the court! Add to this courts are open to the public feel free to relay the story to the public and see what happens when the public find out.

      1. With all due respect, I think you are missing the point. The charge was dropped, so that statement should not have been made, unless you feel differently?

  6. One bit of evidence not mentioned that could be mentioned regarding street triage is peoples (the people being detained or not) experience of the system as a whole. Surely thats the sort of evidence that needs to be included…. actually it seems to me to be the whole point.

    1. Didn’t I say above “what price human dignity ….” followed by a range of other benefits that amount to the same thing as your suggestion? Assuking it doesn’t take endless money to achieve it – it seems we may agree.

  7. Yes, we do agree. Its the same point. I should have been more explicit that what I was suggesting is that people who are detained should be asked explicitly about their experiences and that this data should be collected and monitored as part of the evidence base.

    Mental Health Services are notoriously bad at listening to the end user of their services. My surmise is that from an experiential point of view overall the police would come out looking quite good compared to so called mental health professionals.

    Their is much about mental health issues that end up being counter intuitive. The opportunity to be alone for a few hours in a police cell might be hell for some but a rare chance to reflect on life in peace and quite for others.

    Humans are funny old things…..

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s