Practicalities of introducing PbR in mental health – Clustering and initial assessment

This post is one in a series on Payment by Results in mental health, written for both professional and personal reasons. The full series is introduced and linked to here, including a post with all references in it.

Since 2010/11, there has been a considerable programme of work to support the introduction of Payment by Results into mental health. Below outlines the various areas that have been covered in the introduction and some of the issues arising. (Note that not all issues are dealt with as comprehensively as others.)


Recall that, in PbR for acute care, there are three fundamental features of Payment by Results: (1) classification; (2) currency; and (3) costing (DH, 2011a:18; CHE, 2009:45; NDTi, 2012:8).

Unlike PbR for acute care, mental health classification in PbR does not use the ICD-10 or OPCS-4 classification systems. Instead, mental health professionals rate service users using something called the Mental Health Clustering Tool (MHCT). This tool has 18 scales: the first 12 of these are the Health of the Nation Outcomes Scales (HoNOS), covering areas like depressed mood, problems with activities of daily living etc. Each scale is given a rating from 0 (no problem) to 4 (severe to very severe problem). The additional six scales are known as the Summary Assessment of Risk and Need (SARN), and take into account historical information (DH, 2011b:5).

The result of this process is that it enables clinicians to allocate people to a cluster.

In PbR for mental health, the cluster is the currency.

The key points to note with regard to clusters are as follows:

  • Clusters are groupings based on common characteristics such as level of need, and requiring similar resources to meet those needs through the provision of packages of care
  • There are 21 clusters in total. These are split into 3 super-clusters:
    • Non-psychotic (clusters 1-8)
    • Psychotic (clusters 10-17)
    • Organic (clusters 18-21)
    • (A full list of the what each cluster covers is included in a separate post. Cluster 9 is a blank cluster)
  • The clusters are mutually exclusive. A service user can only be allocated to one cluster at a time – if they transfer to a new cluster following a reassessment, the previous cluster episode ends
  • The same diagnosis could be associated with several clusters, since clusters reflect the assessed level of need
  • The mental health clustering booklet – which has so far been published annually by the Department of Health – helps clinicians to decide which cluster someone should be allocated to
  • If no match to cluster is possible, but the service user requires treatment (typically non-severe autism or learning disability), then a variance cluster (cluster 0) is used. Cluster 0 is sometime also used because of co-morbidities. The use of cluster 0 should be reducing over time as clinicians gain more confidence in clustering, and the clustering tool is further developed to take account of less frequently encountered complex needs
  • The final decision on which cluster to allocate a service user to rests with the mental health professional.
  • Clusters are based on a period of care over specific periods of time, which range from four weeks to 12 months before a review takes place
  • The clusters apply to both admitted patient care and care in a community setting
  • The clusters are designed to be setting independent, on the premise that people should be treated in the least restrictive care setting possible

The approach of establishing 21 clusters was initially developed by the NHS in Yorkshire and Humber and the North East. The approach started life as a clinical tool and then shifted to support the implementation of PbR in mental health (DH, 2010:4).

Clustering can happen at any one of three points (DH, 2011b:5-6):

  1. Initial assessment
  2. Scheduled reassessment
  3. Any reassessment following a significant change in need that cannot be met by the continuation of the current cluster care package.

Initial assessment

It is useful to note that initial assessments are paid for as a separate activity (DH, 2011b:3). This reflects the fact that initial assessment can be an intensive process requiring significant professional resource (DH, 2011b:10). For 2013/14, the assumption is that initial assessments will be the average of the number of user-professional contacts prior to clustering, which is set at two contacts or two working days for in-patients (DH, 2012a:41). Though it is preferred that this assessment is calculated for each cluster, the calculation for average assessment costs will be done across all initial assessments for all clusters in 2013/14.

It is also useful to note the areas not currently included within the scope of PbR in mental health:

  • CAMHS (though this is currently being piloted in 22 sites – see
  • IAPT
  • Forensic services
  • Secondary drug and alcohol misuse
  • LD services for non-mental health needs

MH services provided under a GP contract (DH, 2011b:15)


Published by


Man of letters & numbers; also occasionally of action. Husband to NTW. Dad of three. Friendly geek.

One thought on “Practicalities of introducing PbR in mental health – Clustering and initial assessment”

Leave a Reply

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

You are commenting using your 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