Impact of hospital admission upon patterns of primary care prescribing

Date of Approval: 
2016-10-18 00:00:00
Lay Summary: 
Medicines are among the commonest treatments used by doctors to improve the health and well-being of patients. Admission to hospital can result in considerable changes to a patient's usual medicines. However, little is known in the UK about the nature of these medication changes, such as which patients and what types of medicines are most affected. We will examine how the numbers and types of medicines prescribed by GPs change before and after an admission to hospital in 100,000 adults who were hospitalised in 2014. We will find out what factors might affect any change in medicine use, including the reason for hospital admission, the urgency of the hospital admission, and length of stay in hospital. We will also take into account factors such as the age, sex and long-term health problems of the patient. This work will help to identify patients who are most likely to experience changes in their medicines after a hospital admission, and what types of medicines are most likely to be affected. Understanding these issues will help us design ways of improving the quality of prescribing and care for patients who have recently been discharged from hospital.
Technical Summary: 
Prescribing is a key therapeutic intervention offered by doctors, with the majority occurring in primary care. Being admitted to hospital can result in considerable changes to a patient's usual medicines, resulting in medication errors and discrepancies. Little is known of the nature of these changes in UK clinical practice, such as which therapeutic areas are affected most and which patients are most likely to experience such changes. Using data from 100,000 adult patients admitted to hospital in 2014, we will assess overall changes in all medicines immediately (<6 weeks) post-discharge. Secondary outcomes will include changes to specific therapeutic areas, potentially inappropriate prescribing, and further prescription changes by 6 months. Key exposures of interest will be reason for admission, urgency of admission, and length of hospitalisation. Multivariable regression will be used to model the association between outcomes and exposures, with adjustments made for age, gender, socioeconomic status, long-term morbidities, and GP surgery. This work will provide a valuable insight on prescribing at the primary-secondary care interface. Understanding these issues will inform the improvement of medicines reconciliation processes tailored to relevant therapeutic areas, and help target interventions to those individuals most likely to benefit.
Health Outcomes to be Measured: 
Pre-admission and post discharge long- and short-term prescribing. Primary outcomes: change in total number of medicines following admission, and number of changes in medication following admission. Secondary outcomes: examining change in number of medicines and changes in medication within specific therapeutic areas. The presence of inappropriate prescriptions changes following hospitalisation
Application Number: 

Rupert Payne - Chief Investigator - University of Bristol
Rachel Denholm - Collaborator - University of Bristol
Richard Morris - Collaborator - University of Bristol
Sarah Purdy - Collaborator - University of Bristol

HES Admitted Patient Care;Patient Level Index of Multiple Deprivation;Practice Level Index of Multiple Deprivation