Dr Ollie Hart is a GP partner in the Sloan Medical Centre in Sheffield and clinical lead for the person-centred care work stream in the city.
Dr. Hart has been involved in an innovative pilot research project, based in his own inner city practice with 13,000 patients, to make the case for a more person-centred approach to how primary care can support people living with diabetes. In this blog for C4CC he shares his experiences.
The pilot was set up as first step towards how we can assess and segment people living with long-term conditions like diabetes, so that care can be tailored to their individual needs.
This avoids over-treating people who are doing well, and offers a more holistic, personalised approach for people who feel less confident, skilful or knowledgeable to look after their health.
It also enables health systems to focus a ‘medicalised approach’ on those most ready for input, avoiding waste and improving satisfaction all round for both staff and patients.
The new process
As part of their scheduled annual review (in their birthday month for ease of recall), people with diabetes are initially assessed for the traditional markers of disease control.
At the same time they are assessed for how they perceive their self-management skills, knowledge and confidence, using the Patient Activation Measure (PAM). The results are then used to segment people into one of four quadrants.
Letter group: Low medical complexity (appropriately controlled disease markers) and high activation (PAM Level 3 or 4), would be determined as doing fine, and be treated with a light touch. In practice this involves writing them a letter to congratulate them on how they are doing, and encouraging them to do more of the same.
Practice group: High complexity (one or more markers of disease control not met) and high activation. The working assumption is that with higher activation people are more ready for a conversation with a stronger focus on medical management and medical focused goals and activities. Therefore this group is invited into the practice for review with a doctor or nurse, who is given longer to have a coaching style consultation.
Wider support group: High complexity, low activation (PAM level 1 or 2). The working assumption here is that people in this group are more likely to struggle to interact well with a purely medical model.
They are more likely to be overwhelmed with other life issues, with health care low on their list of competing priorities, or feeling under confident or under-skilled to take on self-management.
They may well appreciate the more holistic approach that could be offered by a non-traditional provider, as an initial point of contact.
They can give them more time, and consider a wider range of issues beyond just health. This approach is sometimes referred to as ‘social prescribing’. At its best this involves a co-production process where patient and advisor consider the person’s individual needs and preferences, and matches them to the best resources and support available.
This might range from a series of 1:1 conversations with a health trainer, to cooking classes, to exercise opportunities, to review of benefits and finances.
It would still be important that people in this group are offered medical service input, but on the terms that best suit the patient.
There would be a strong focus on collaboratively agreed ‘small steps’ (health or otherwise), with the primary aim to build a patient’s activation as well as creating conditions for better health and wellbeing.
Prevention group: Low complexity, low activation. People in this group satisfy traditional medical targets but the assumption is their low level of activation puts them at risk of future poor disease control and they may well benefit from support to build their activation. This might include considering their health literacy, and their wider circumstances and coping skills.
Those people segmented into wider support or prevention groups are contacted by telephone to have a shared decision making process. Their results are explained and they are offered a range of available options depending in the local provision. Those managed in the Letter or Practice group are still offered access to social prescribing through signposting in the letter or via the practice team, if they wish.
Practice patient feedback was proactively sought via the patient participation group, through enquiry in face-to-face appointments, and at a diabetes patient group. This last group included 15 patients who had experienced the new quadrant model over the previous two months.
The feedback was that the patient body felt this was an appropriate and acceptable model to pursue. Some people who fell into the Letter group, identified the value of an appointment to help ‘maintain their motivation’. However, they acknowledged that this be seen as a lower priority in a system under strain.
Practice staff involved in this approach, particularly the nurses conducting the long-term condition reviews report that their consultations are more appropriate and satisfying.
I recognise that segmenting people in this way could be seen as an artificial ‘pigeon holing’ process that presents a reductionist approach to the complexity of people’s lives.
I also recognise that assessing people’s activation using a questionnaire has limits. These are discussed more fully in the longer paper.
However this approach is intended as a very deliberate ‘first step’ in implementing a personalised approach to LTC management.
Just like learning to drive a car, applying new processes requires simple easily understood steps towards a new way of doing things. This applies both to patients and practice teams.
The reality is that there is very little funding available for new innovations, and GP practices are reporting very high levels of workload.
Therefore attempts to embed new person centred approaches, must include identification of where it is fine to do less, to allow time and energy for new learning and practice. Focusing practice time on those most ready for medical input, allows practice staff to spend longer pursuing a more robust health coaching style approach, developing care plans that are meaningful to the patient.
The value of addressing the wider determinants of health, especially in those at lower levels of patient activation, also has to be recognized.
Evidence suggests that this often best delivered by community teams, closely networked into local non-medical services and support. This model systematically directs people most likely to benefit towards these services.
In summary, this segmenting approach seeks to safely and appropriately tailor care towards an individuals needs.
It seeks to make the most of finite resources and the range of people and organisations that can contribute to a holistic approach to health and wellbeing.