New findings: impact of patients' ability to manage their health conditions and NHS service use

Measuring the ability of patients to manage their health conditions

One way to measure patients’ ability to manage their health conditions is the Patient Activation Measure (PAM). This is a validated tool that measures the extent to which people feel engaged and confident, in taking care of their health conditions (Box 1). In 2016, NHS England agreed a five-year licence to use PAM with up to 1.8 million people as part of its work to provide personalised care for people with long-term conditions.

Box 1. How can you measure the ability of patients to manage their health conditions?

There are various methods available that measure aspects of patients’ ability to manage their own health. These include health literacy measures, which measure ‘the degree to which individuals have the capacity to obtain, process and understand basic health information and services needed to make appropriate health decisions’. However, there are many definitions and measures of health literacy, which is often complex to assess. Also notable are self-efficacy measures, which measure a patient’s belief in their ability to succeed in certain tasks and PAM, which measures a patient’s knowledge, skills and confidence in managing their own health and health conditions.

The PAM tool is licensed by the US company, Insignia Health LLC. Individuals are asked to complete a short survey and, based on their responses, they receive a score between zero and 100. The resulting score places the individual at one of four levels of activation, each of which is associated with a level of ability to manage their health.

We partnered with Islington Clinical Commissioning Group (CCG) to examine what these PAM questionnaires are telling us about how patients with different levels of knowledge, skills and confidence use the NHS.

Islington CCG sent PAM questionnaires to over 35,000 patients living with long-term conditions in their area, asking them 13 questions about their beliefs, confidence in managing health-related tasks and self-assessed knowledge. Examples of such statements include: ‘I am confident that I can tell whether I need to go to the doctor or whether I can take care of a health problem myself’, ‘I know what treatments are available for my health problems’, or ‘I am confident that I can tell a doctor my concerns, even when he or she does not ask’. The PAM questionnaire could be filled in with the help of a carer but it was intended for the use of the patient themselves. A variation of the PAM questionnaire is available to address the needs of carers specifically, but was not used in this study.

9,348 patients (25%) returned the questionnaires. They were grouped into four levels of ability to manage their health conditions as described in Box 1. Here are real-life examples for illustration:

Level 1:

Individuals tend to feel overwhelmed by managing their own health or health conditions, and may not feel able to take an active role in their own care. They may not understand what they can do to manage their condition better, and may not take their medication, attend preventative appointments, or see the link between healthy behaviours, such as smoking cessation, and good management of their condition. This group made up 22% of respondents in the Islington sample.

Level 2:

Individuals may be able to manage some aspects of their health (for example, take their medication and attend appointments) but still struggle in some aspects of their care, for example creating a care plan with their practitioner (19% of respondents).

Level 3:

Individuals appear to be taking action, for example setting goals for their health (such as adhering to a medically-advised diet) or creating a care plan with health care providers, but may still lack the confidence and skill to maintain these (46% of respondents).

Level 4:

Individuals have adopted behaviours and practices to manage their condition, such as good medication adherence, care planning or self-monitoring. Such individuals may be accessing support from online groups, peers and their community. Patients occasionally still need extra support, for example to recover from a relapse or a hospital procedure, or when other life stressors make it difficult to maintain their usual practices (13% of respondents).

Our research examined whether patients assessed as having a higher level of activation were less likely to use primary and secondary health care than people who were less activated. Previous evidence in England was from a smaller sample of people surveyed by phone and international evidence has come largely from the United States, which has a very different health system. A previous qualitative study (funded by the Health Foundation) examined the feasibility of using PAM in the NHS, but evidence was needed on whether there was an association between ability to manage health conditions, as measured by PAM, and use of services across primary and secondary care. Until now no quantitative research existed from the NHS.

Our study provides insight, using large linked datasets for the first time, into the relationship between a patient’s ability to manage their health condition and their health care utilisation across primary and secondary care.

What’s the relationship between the ability of patients to manage their own health and emergency hospital admissions?

On average, the patients who were least able to manage their health conditions were aged 67 – 58% had two or more long-term conditions and 39% lived with a mental health condition. Just over half (55%) lived in the most socio-economically deprived areas in Islington.

In comparison, patients who were more able to manage their health conditions tended to be slightly younger and healthier, but otherwise had a similar profile (Figure 2). On average, they were aged 64, with 50% having two or more long-term conditions and 24% having a mental health condition. Half (50%) lived in the most deprived areas of Islington.

This suggests that patients can have low levels of ability to manage their health conditions regardless of their age and whether they live in affluent areas or more deprived ones.

Figure 2: Characteristics of patients included in our research study

We were interested in whether the patients most able to manage their health conditions used the NHS differently to patients who were least able. In making these comparisons, we were mindful that many patient characteristics affect demand for health care, including the number of long-term conditions, age, and whether they live in a deprived area. Therefore, we took these kinds of characteristics into account by building a statistical model. Our models effectively estimate the association between emergency admissions and the ability to manage their health conditions for patients with similar age, health and socio-economic status.

After controlling for characteristics in this way, we found that the patients who were better able to manage their health conditions experienced fewer emergency admissions (Figure 3). Those most able (PAM level 4) experienced 38% fewer emergency admissions than the patients who were least able to manage their health conditions (PAM level 1).

We also found that those best able to manage their health conditions were 32% less likely to attend A&E (with no emergency admission) compared to those least able. They were also 33% less likely to have attended A&E for a minor condition that could be appropriately treated in primary care. Patients a level higher (PAM level 2) had 29% fewer emergency admissions than those least able.

Figure 3: Difference in emergency admissions (% reduction) compared to patients in PAM level 1

Figure 4: Difference in health care utilisation (% reduction) – patients with PAM level 4 compared to patients with PAM level 1

What is the association between ability to manage health conditions and how often patients use general practice and elective care?

Emergency admissions are not the only example of where patients are using more health care. General practice workload is also increasing and health care leaders are also concerned about the volume of patients attending emergency departments without being admitted and the length of stay once they are admitted to hospital.

We found similar patterns for general practice appointments, elective admissions and outpatient appointments to those we found for emergency admissions (Figure 4). For example, compared with those least able to manage their health conditions (PAM level 1), the next most able (PAM level 2) had 8% fewer GP appointments, and 10% fewer outpatient appointments, while the most able group (PAM level 4) had 18% fewer GP appointments and 19% fewer outpatient appointments.

We also found that, once patients were admitted to hospital for elective care (which is often necessary for care of long-term conditions), length of stay was 41% shorter among the patients most able to manage their health conditions compared with those least able. One potential explanation is that more able patients can be discharged home with fewer supports, as they may be better able to understand and manage their medication and care without additional support from community or social care services.,

We also found that patients with higher levels of ability to manage their health conditions were less likely to miss their appointments with GPs and with outpatient clinics (Figure 4). This is important for two reasons: first, patients who are less able to manage their health conditions may be missing out on the health care that they need to prevent their condition deteriorating. Secondly, missed appointments are also potentially a waste of NHS resources.

Do we find the same pattern in those with a mental health condition?

Almost one in three (30%) of patients in our study had a mental health condition, such as depression, often in addition to a physical health condition, such as diabetes or asthma. When we analysed their data specifically, we found that those most able to manage their health conditions experienced 49% fewer emergency admissions than those least able and had 32% fewer A&E attendances. These findings were established after controlling for clinical and demographic characteristics of the patients in the statistical model. They suggest that there is heightened potential to reduce emergency admissions by improving the ability of patients with mental health conditions to manage their wellbeing.

Strengths and limitations

Our study has many strengths, namely the large sample size, the fact that we had data about health care utilisation from across the NHS and that patients were registered for a long time with their GPs (median length of 15 years).

However, it has obvious limitations too. For example, information on the severity of people’s illness is not routinely recorded in health records and, while we have good information on levels of deprivation where patients live, we don’t have information on the individual support (such as family or friends) they can access. There is a chance that any association between self-management capability and health care utilisation is also influenced by support from families or social care that patients may have access to – something we were unable to measure. Finally, the questionnaire suffered from a low response rate, only 25% of patients were recorded as having returned it. The qualitative evaluation of PAM has also identified issues with the feasibility of administering the questionnaire using other methods – indicating that further work is needed to reach all patients. The low response rate and our analysis of respondents suggest that some patients, particularly those with high A&E attendance and low GP attendance, were less likely to return the questionnaire. This means that those who returned the questionnaire might not be comprehensively representative of the population living with long-term conditions in Islington.

What do these findings mean for patients and the NHS?

Patients living with long-term conditions, their advocates and many clinical experts have long recognised the importance of supporting patients to live well with their conditions, as part of high-quality, person-centred care. There is emerging evidence it may lead to patients having better clinical outcomes, improved wellbeing and more satisfaction with the care they receive. This alone is a reason that the NHS should invest in self-management support. However, when emergency admissions are rising, and demand for GP care outstrips an overstretched workforce, investing resources to support patients can be often overlooked. Our study demonstrates that this would be a mistake: it shows a clear link between the knowledge, skills and confidence a patient has to manage their health conditions, and their demand for health care, particularly emergency care.

If the group of patients who are currently least able to manage their conditions could be supported to manage their conditions as well as those who are currently most able, then the impact on the NHS could be as large as a reduction of 333,000–436,000 emergency admissions per year or 6% to 7% of the total number of these admissions. There could also be large reductions in the number of attendances at A&E, between 504,000 and 690,000 fewer per year or 5% to 6% of total attendances (Box 2). In our analysis in Islington, 32% of attendances at A&E were at the lowest level of activation, a quarter of which were for minor illnesses. This would equate to 2 million A&E attendances nationally, 544,000 of which could have been treated in primary care, suggesting that a large proportion of attendances could be avoidable. While it may never be possible for all patients to develop the skills and confidence to move to PAM level 4, some improvement in their ability – for example, moving up a PAM level – could have a noticeable impact on emergency admissions.

Box 2. Extrapolating our findings: methods

60% of patients admitted as an emergency in 2015/16 in England had at least one long-term condition – equal to 3.6 million emergency admissions per year.

Our analysis of patients with long-term conditions in Islington suggests that 32% of all emergency admissions are for patients at the lowest level of activation. Extrapolating this nationally, this translates to 1.1 million emergency admissions each year in England for these patients (32% of 3.6 million). Our analysis suggests that (controlling for patient and clinical characteristics) the patients most able to manage their health conditions had 38% fewer emergency admissions than those least able, or 29% fewer emergency admissions than those a level above. Therefore, there is potential to reduce the number of emergency admissions by between 333,000 and 436,000 per year, if patients who are currently least able to manage their health conditions were better supported so that they had the same ability as those better able to manage their health (29% and 38% of 1.1 million). This is equal to 6% to 7% of the total of admissions.

While exact figures are not available, if we assumed that 60% of the 10.9 million attending A&E without admission have a long-term condition, then this equates to 6.5 million emergency attendances. As those in the highest category of activation (PAM level 4) are 33% less likely to attend A&E, up to 690,000 A&E attendances could be avoided through increased support for self-management. More conservatively, those in PAM level 2 are 24% less likely to attend A&E, therefore up to 504,000 A&E attendances could be avoided if the least able (PAM level 1) were better supported to manage their condition. This is equal to 5% to 6% of the total of admissions.

However, strong caveats are attached to these estimates: first, we assume that our findings in Islington might be replicated across the country; secondly, we assume that all patients could become more activated and rely on effective interventions to achieve this. Finally, no intervention can be 100% effective so the full reduction is unlikely to be realised. That said, our selection of emergency admissions of patients with a long-term condition is conservative; more than 60% of patients admitted may have a long-term condition and there is also the possibility that activation could be improved within levels, as well as between them.

Box 3. Can self-management support always improve a patient’s ability to manage their health conditions?

The Health Foundation has previously published a practical guide to self-management support. This shows that providing information and access to tools (for example, online courses, electronic and written information material, patient access to medical records) is an important component of a health care system that encourages self-management support. However, such interventions alone may not improve a patient’s confidence and ability to put that information into practice. In contrast, other interventions, including health coaching and peer support, may improve a patient’s knowledge, skills and their confidence and, therefore, their ability to manage their health conditions.

Our study shows that there is a substantial proportion of patients whose ability to manage their health condition could be better supported and potentially improved. In the next section of this briefing we summarise some possible interventions and initiatives that could be introduced by commissioning services to unlock the potential of patients to self-manage.

Previous Next