The World Health Organization defines social determinants as “the conditions in which people are born, grow, work, live and age, and the wider set of forces and systems shaping the conditions of daily life”. Social determinants are non-medical factors that impact a person’s health.
In recent years, awareness has evolved, and experts agree the environment people live in, their socio-economic status, lifestyle and cultural background affect the health status of an individual far more than previously understood. Across many studies, authors agree that social determinants of health determine approximately 40% of an individual’s health outcomes, morbidity and mortality.1
Understanding social determinants can help predict negative outcomes
An understanding of social factors provides invaluable insight into improving a patient’s health status for predicting negative outcomes. Instances such as high-risk pregnancies, early hospital readmissions, poor compliance with prescribed medications, and susceptibility to chronic disease.
The structure of our current healthcare delivery system impacts health during episodes of injury or illness, while social determinants interact with health much earlier and on a day-to-day basis.
Addressing social determinants to achieve health equity
Addressing social determinants is crucial and is a primary approach to achieving health equity. Health systems and providers need options to address these issues once recognised in their patients.
Physicians can better support patients faced with social challenges by asking about their social history, providing them with advice, referring them to local support services, facilitating access to these services and acting as a reliable resource-person throughout the process.2
Once a social diagnosis has been made, social prescribing involves connecting patients with various support resources within and beyond the health system, such as local women’s groups, housing advocacy organisations or employment agencies.
Examples of social determinants
Social determinants directly impact every individual’s health. For some factors, the link between the factor and poor health may be clear cut; in other cases, the link has been established epidemiologically, though the mechanism may not be so readily apparent.
Examples of some social determinants include income and accumulated wealth, length of education – especially tertiary education, transport including ready access to healthcare services, social and community support, residential segregation, stable housing, community levels of crime and violence, recreational and leisure opportunities. Cultural issues including racial segregation are closely linked to these determinants, while also being impactful of themselves.
Many primary care visits are linked to the effects of stress arising from social factors such as financial worries, family responsibilities, marital relationships, and so on. A new area with the potential to add value for clinicians is the availability of additional information combined into summarised scores using machine learning. This approach combines raw data such as utility bills, traffic tickets, mortgage certificates, judgements and liens, address changes, and educational attainment into combined social determinant-based risk factors.
Read our white paper on Unlocking the power of machine learning, where we examine a patient scenario and review how social determinants like lack of disease education, lack of access to transport and low-income levels can lead to an unfavourable patient outcome.
How are health systems addressing the social determinants of health?
Healthcare organisations and front-line care providers should aim to connect patients with social needs to the relevant government and community resources and social services assets.
For instance, the establishment of dedicated clinics for Aboriginal people run by Aboriginal people, supporting efforts to address homelessness and so on. The key is for healthcare providers and provider organisations to encourage synergy among the various activities and invest in their sustainability.
Incorporating data on social determinants that address unmet social, financial, or economics needs of patient populations significantly improve clinician and care team understanding of the individual. In turn, addressing those needs should enable provider organisations to improve the overall health of a patient to a greater extent than a single-minded focus on purely clinical issues.
How can Orion Health help?
With the rise in availability of social determinants data, care providers now have access to a huge virtual library of relevant patient information they can use to better manage their individual patients and entire patient populations.
This is where machine learning can help. Machine learning can transform the way healthcare providers gain insights from data stored across a region including clinical and social determinants Orion Health data and use it to make better more informed decisions.
Our strengths are in our ability to combine data from a huge range and diversity of sources, combine it in ways that help clinicians make decisions and apply machine learning to the entire population.
Our Orion Health Intelligence platform uses machine learning and predictive modelling to analyse volumes of data and different data types and finds patterns and reason about data, enabling healthcare providers to move closer to personalised medicine.
Data-driven health helps clinicians and population health managers to quickly understand individual patients and their entire population of patients.
Interested to know more about how Orion Health Intelligence platform can help you?
1 Conference Paper: Hyojun Park, A., Roubal A., Roubal, B., Rudolph R., Booske C November 2013 Relative contributions of health determinants on health outcomes at the county level Conference: 141st APHA Annual Meeting and Exposition 2013.
2 Galea, S., Tracy, M., Hoggatt, K. J., DiMaggio, C., & Karpati, A. (2011). Estimated deaths attributable to social factors in the United States. American journal of public health, 101(8), 1456-1465.
