Demanding better: The social determinants of health care demand

Our shared understanding of how health care should be delivered, both in the NHS and elsewhere, has traditionally been shaped by some underlying assumptions:

• that most health care is solely evidence based
• that it is delivered by those who understand medical science and technology to those who do not
• that health care is the principal contributor to the health and wellbeing of people and populations.

More recently, though, these assumptions have been challenged. A significant body of research tells us that patients’ preferences and levels of engagement can fundamentally change the way knowledge is applied and can influence the approaches that are used to prevent and co-manage illness.

Furthermore, the circumstances in which people are ‘born, grow, live, work and age’ – and the effects of those circumstances on people’s behaviour – contribute far more to health and wellbeing across the life course than health care itself.

Significant research attention is now being devoted to these ‘social determinants’ of health: to adverse childhood experiences and the links between deprivation, austerity and mortality. This work tends to focus on the importance of life circumstances in creating good health and often concludes that health care makes a relatively modest contribution.

Effects on the system

But these social determinants also have a significant impact on demand for health care. For example, socially deprived areas tend to have higher levels of multimorbidity, higher numbers and rates of admissions to hospital and extended stays once people are admitted. They also increase the level and complexity of the demand placed on primary care.

Social isolation is increasingly understood to be a major factor in determining demand for care, especially for the elderly. In a recent seminar convened by the Nuffield Trust on the social determinants of demand for health, a representative from one general practice reported that nearly half of all frequent users of that practice lived alone – compared to less than a quarter of the remainder of the practice population.

Another participant gave a powerful account of how much of his practice involves patients who are dealing with the impact of abuse during childhood, failed attachment in early life and trauma in later life resulting from domestic violence, incarceration or military service. The long-term effects of trauma are experienced as physical conditions as well as the psychological impact of depression, anxiety, self-harm and addiction.

All of these can be exacerbated by income insecurity and low levels of education. As Harry Burns has eloquently pointed out, poverty is a cause of these issues, but equally may be the result, so eliminating poverty is not alone enough to address these problems.

A different understanding

So if the determinants of health care demand are more complex than we have traditionally anticipated, are we responding in the right way? Is there a risk that our health systems are geared towards offering medical solutions that may not accurately reflect what is needed or what people want? Are there cases where needs can be met in less resource-intensive ways?

For example, problems associated with trauma can manifest as a range of symptoms that patients and clinicians attribute to physical conditions.

“Typically these patients will be under a gastroenterologist for their bowel symptoms; they will have had multiple CT and MRI scans… they may well have had pelvic surgery, especially if there is a history of sexual abuse. They will often be breathless and will have been investigated by cardiologists and respiratory physicians. They will probably be on a cocktail of dangerous and expensive opiates and will have long-term appointments at pain clinics and may well have had spinal surgery and injections, and they will keep attending and being investigated. I had one patient who attended A&E 60 times in a year and had 20 MRI scans… he had a history of being abused.” [GP participant at Nuffield Trust seminar]

Furthermore, the pressure for rapid access to general practice and the high levels of demand for other services can create incentives to find a diagnosis and ‘move the patient on’. A diagnosis may be required in order that the patient can claim benefits, so there is further incentive for both doctor and patient to agree on a label.

Social stigma can also lead to patients (and health professionals) interpreting issues as physical problems needing medical interventions. While for health professionals this may reflect a somewhat reductionist biomedical model of medicine, assigning a diagnosis can help to reduce the stigma and shame patients may be feeling:

“[GPs] are… gatekeepers between illness – what the patient feels – and disease – the label that they are then given. If you have a very biomedical mindset you will describe suffering as disease, and then the patient has a label.” [Participant at Nuffield Trust seminar]

And even when these conditions are correctly identified as being socially derived, GPs and other professionals may wish to ‘do something’ for a distressed person while in their presence, but may not have access to suitable non-medical responses.

The time and skills that professionals have available for engaging patients in goal-setting, shared decision making and motivational interviewing are limited. The resulting failure rates, especially in supporting behaviour change, are often demoralisingly high. Opening up conversations about these issues could mean that a 10-minute consultation turns into half an hour, and a relatively straightforward transaction can become a challenging act of emotional labour.

How best to respond

There are several implications of this.

For commissioners or integrated care organisations, the need to develop much richer linked data that can identify hot spots where loneliness, joblessness and other sources of deprivation are driving demand for health and other services is very important. Problems with information governance can be an obstacle here, but participants at our seminar from Kent, Camden, Manchester, Hampshire and Islington gave local examples of how this has been overcome.

Using the knowledge health and care teams have on the people they are working with is another powerful technique. The daily ‘huddle’ that many practices now operate is useful for this. The coding of GP records needs to capture social risk factors in ways that make them easy to find in a consultation or needs assessment.

For primary care and general practice, there is a need to build better responses to these social determinants of demand. Doing this could potentially take a lot of pressure off general practice. The growth of social prescribing, asset-based community development, more multidisciplinary models of general practice and, crucially, the inclusion of mental health in front-line primary care are all supported by a growing body of experience and research. In appropriate settings the inclusion of welfare, housing and other advice and support also makes a difference.

Many of the social causes of demand for health care are difficult to influence. But ensuring that there is a suitable response and that communities, patients and professionals have access to the service they need – rather than the one that is available – is a good first step.