New infections of HIV have decreased over the last decade, though, as care and management improves, the prevalence of people living with HIV (PLWH) has increased. HIV care is managed in sexual health clinics, but most PLWH are registered with a GP who are the first point of contact for most other conditions. To ensure patient safety when managing other conditions, it is advised for all HIV diagnoses and treatment regimens to be shared with the patient’s GP. In 2005, this occurred for only 51% of patients. As improvements have been made to data sharing, patient safety and ensuring accurate primary care records, data completeness may have improved. Understanding the completeness of HIV-related data within primary care will identify any potential gaps in data sharing between sexual health clinics and primary care, and improve our understanding of generalisability of HIV-related research using primary care records. Therefore, we will calculate the prevalence and incidence of HIV from CPRD and compare our findings to national statistics. Additionally, we will report how well antiretroviral therapy (ART) is recorded.
PLWH experience an increased prevalence and incidence of many co-occurring chronic conditions including cancer, cardiovascular disease, metabolic conditions, mental illness and neurocognitive disorders. We will investigate which conditions commonly occur (cluster) among PLWH and estimate which clusters are associated with hospitalisation and mortality. Understanding what conditions are commonly clustered together within PLHW and which are related to healthcare utilisation and death is imperative for adapting health services and improving care for aging PLWH.
The aim of this study is to compare prevalence and incidence of and completeness of medication for HIV in primary care settings to national statistics and to determine common comorbidity clusters and how they relate to hospitalisation and mortality in people living with HIV (PLWH).
Using data from CPRD of everyone aged 15 and over, we will calculate overall and annual prevalence of HIV between 2002 and 2020; two time points will be compared with the last two Public Health England reports (one prior to the COVID-19 pandemic and one since the start of the pandemic). Also using data from CPRD of everyone aged 15 and over, overall and annual incidence of HIV will be calculated between 2002 to 2020; annual incidence from 2011 to 2020 will be compared with the available data from Public Health England reports (data prior to 2011 is not available). The prevalence and incidence estimates will be standardised by age and sex to compare figures for these specific groups of PLWH with national statistics.
From all PLWH identified in the above analysis, we will calculate the proportion of PLWH with a record of any antiretrovirals and those with a record of the expected treatment regimen (i.e. two nucleoside reverse transcriptase inhibitors - typically prescribed in a combined pill - and one ritonavir-boosted protease inhibitor, nonnucleoside reverse transcriptase inhibitor or integrase inhibitor. These will be calculated separately for 2019 and 2020.
Using binary diagnostic codes, we will first determine comorbidity clusters using a latent class analysis then investigate the relationship between each of the comorbidity clusters with hospitalisation and mortality data using a Cox proportional hazards regression (time to event with 5 year follow up). All PLWH aged 18 years or above will be included in this analysis. Study period will be from 2002 and 2020.
- Prevalence and incidence of HIV using primary care records
- Completeness and type of antiretroviral therapy prescriptions recorded in primary care
- Prevalence of comorbidity clusters
- Clusters associated with all-cause hospitalisation in PLWH
- Clusters associated with all-cause mortality in PLWH
Tiffany Gooden - Chief Investigator - University of Birmingham
Tiffany Gooden - Corresponding Applicant - University of Birmingham
Jennifer Cooper - Collaborator - University of Birmingham
Krishnarajah Nirantharakumar - Collaborator - University of Birmingham
HES Admitted Patient Care;ONS Death Registration Data;Patient Level Index of Multiple Deprivation