Patient reported GP access and out of hours emergency department visits in children. A national population based study.

Talk Code: 

The problem

The National Health System in the United Kingdom is under major strain from rising numbers of emergency department (ED) visits. Children are among the largest users of EDs. Yet parents prefer their child to be seen by a general practitioner (GP) for first point of care when their child is acutely ill. Children's access to GPs could have diminished because of incentivised chronic disease management in adults creating pressure on the availability of appointments during normal surgery opening hours, while the majority of GPs do not provide out of hours (OOH) primary care. We aimed to investigate the relationship between patient reported access and ED visit rates in children comparing visits in and out of GP contractual hours. Our hypothesis was that children were more likely to visit EDs if they were registered with practices whose patients reported lower accessibility to their GP.

The approach

We carried out a cross sectional study of 9.5 million children aged <15 years between April 2011 and March 2012 registered with 8035 English practices. We investigated ED visit rates using Hospital Episode Statistics data. We defined out of hours (OOH) as visits between 6.30 pm and 8am Monday to Friday on weekdays, and all day on weekends and public holidays. We linked data, via the child's recorded practice, to national GP Patient Survey data. We assigned each practice to one of six ‘access' categories according to the proportion of patients who reported that they were able to get a GP appointment on their last contact. The lowest access category was <75%, the highest 95-100%. We applied a negative binomial regression model to estimate ED visit rate ratios (RR) according to patient reported access. We stratified by OOHs and adjusted for age of the child, sex, practice level deprivation.


The crude rate was 325 ED visits per 1,000 children (3,074,616/9,458,000). 56% of visits were out of GP contractual hours (1,722,899/3,074,616). Patient reported access ranged from 51% to 100% (median: 88%, interquartile range: 83% to 92%). Practices with lower reported access had higher ED visit rates. There was an 8% higher estimated rate in the lowest compared to the highest access group (adjusted rate ratio (95% CI): 1.08 (1.05 to 1.11). However, the relationship was only evident for OOHs visits.


Families registered with practices where it is reportedly difficult to see a GP are more likely to seek emergency healthcare for their child out of GP contractual hours. If this association is causal then over 60 thousand fewer children would visits EDs a year if all practices had more than 90% of patients reporting that they were able to get a GP appointment.


  • Elizabeth Cecil, King’s College, London, UK
  • Alex Bottle, King’s College, London, UK
  • Thomas Cowling, King’s College, London, UK
  • Azeem Majeed, King’s College, London, UK
  • Ingrid Wolfe
  • Sonia Saxena, King’s College, London, UK