The coronavirus pandemic has highlighted differences in risk of poor outcomes between groups based on their ethnic background, level of deprivation, age and sex. Before the pandemic, inequalities in many other health conditions were clear. Living in socioeconomically deprived circumstances or being from certain ethnic minority groups is linked to having multiple long-term conditions. But we do not yet know whether these social and economic factors are also linked to poorer survival once a person has developed these conditions.
Improving care for people with multiple long-term conditions is a national priority. To target health care resources, we need to know whether there are particular groups of patients with multiple long-term conditions that have poorer prognosis and may benefit from targeted care.
This study will use information contained in health records to understand more about this. Analysts will use anonymised primary care records that have been linked to information on death certificates. They will test whether the time a person survives with multiple conditions depends on their level of deprivation, their ethnic background, their sex, or combinations of these. Differences may arise for many reasons. For example, people in deprived circumstances may find it more difficult to access health care, or more difficult to change their lifestyle to slow the progression of their health conditions. The information needed to understand why differences occur is not contained within primary care records but can be gathered by talking with people who are affected by the challenge of living with multiple conditions.
Prevalence of multiple long-term conditions is higher and age of onset lower among those who are socioeconomically disadvantaged, or from black, Asian and some other minority ethnic groups. Current evidence is mixed on whether deprivation, ethnicity, sex and their intersection modify prognosis once multiple conditions are established. This study will quantify the association between multiple conditions and mortality risk and test effect modifiers. This information could be used to identify sociodemographic groups with multiple conditions that have higher mortality risk and may benefit from targeted care. The coronavirus pandemic has made this issue even more pressing. People with multiple conditions are high users of health care and some may have been especially impacted by disruptions to health care services during the pandemic.
This observational follow-up study of a random sample of adults will address four questions. First, is number and type of multiple long-term conditions associated with mortality risk? Second, do sociodemographic factors moderate these associations? Third, are associations between sociodemographic factors and mortality explained by onset of new conditions, primary care activity, or continuity of care? Fourth, were these associations apparent during a period of “usual NHS care”, i.e. do associations remain if we consider only mortality before the start of the coronavirus pandemic?
Survival time based on deaths from all causes is the primary outcome. A secondary outcome is avoidable mortality through linkage to ONS registered deaths (using ONS 2019 list of ICD-10 codes for deaths from causes considered avoidable through timely and effective healthcare and public health interventions). Cox regression models will estimate hazard ratios for each demographic factor, multiple condition status and their interaction. Sequential adjustment for new conditions, number of GP consultations, and continuity of care during follow-up will be made to examine the contribution of these mediators to explaining differences in survival.
Primary outcome: Survival time (based on deaths from all causes)
Secondary outcome: Survival time (based on avoidable deaths)
Mai Stafford - Chief Investigator - The Health Foundation
Yannis Kotrotsios - Corresponding Applicant - The Health Foundation
Anne Alarilla - Collaborator - The Health Foundation
Arlene Gallagher - Collaborator - The Health Foundation
Elizabeth Crellin - Collaborator - The Health Foundation
Hannah Knight - Collaborator - The Health Foundation
Jay Hughes - Collaborator - The Health Foundation
Thomas Wagstaff - Collaborator - The Health Foundation
Yannis Kotrotsios - Collaborator - The Health Foundation
HES Admitted Patient Care;ONS Death Registration Data;Patient Level Index of Multiple Deprivation