Inequalities in waiting times for elective hospital care: what can routine health data and standard analytical methods tell us?

Abstract

Background/Aims Identifying health inequalities can require substantial data and analytical resources. A healthcare setting that may be particularly exposed to inequality is elective care, where hospital waiting times have substantially lengthened since the COVID-19 pandemic. This study aimed to investigate how routine health data and standard analytical methods could be used to identify inequalities in waiting times relating to age, sex, ethnicity and socioeconomic deprivation. Methods Decision trees were fitted to data on waiting times for 78 510 completed elective treatments from a large NHS organisation in England for the calendar year of 2021. Data were sourced from the Waiting List Minimum Dataset and combined with a local dataset on patient attributes. Analysis was stratified by treatment specialty and whether the treatment was delivered in an outpatient or inpatient setting. A decision tree was fitted to the data at each stratum to assess three questions: To what extent can variation in waiting times be explained by age, sex, ethnicity and socioeconomic deprivation? Which variables are most explanatory? In what ‘direction’ is this explanation (eg for sex, did male or female patients wait longer?). Results Across the elective hospital specialties assessed, a maximum of 12% of variation in waiting times could be explained by age, sex, ethnicity and socioeconomic deprivation. For all decision trees, age appeared as the most important explanative ‘branching’ variable in 54% of cases, followed by socioeconomic deprivation (2%) and sex (1%). Ethnicity was not a statistically significant explanatory variable. Where variation did exist, waiting times were longer for younger patients, female patients and those from areas with greater socioeconomic deprivation. Conclusions According to the approach taken in this study, there is little evidence of significant waiting time inequality dependent on sex, socioeconomic deprivation and ethnicity. Analysis of this nature does not confirm a causal association between younger age and longer waiting time, but instead highlights the need for further explorative analysis. Healthcare managers should be cautious about the use of routine health data and standard analytical methods in efforts to identify health inequalities.