Hospital Episode Statistics Dataset Research Articles

Prognostic factors for colchicine prophylaxis-related adverse events when initiating allopurinol for gout: retrospective cohort study.

Colchicine is commonly used to prevent flares when starting urate-lowering therapy for gout. Patients with gout are frequently concurrently prescribed other medications (such as statins) that may interact with colchicine, increasing the risk of adverse events. The aim of this study was to describe potential prognostic factors for adverse events in patients prescribed colchicine when initiating allopurinol. We conducted a retrospective cohort study in linked UK Clinical Practice Research Datalink and Hospital Episode Statistics datasets. Adults initiating allopurinol for gout with colchicine (01/04/1997-30/11/2016) were included. Potential prognostic factors were defined, and the likelihood of adverse events, including diarrhoea, nausea or vomiting, myocardial infarction (MI), neuropathy, myalgia, myopathy, rhabdomyolysis, and bone marrow suppression, were estimated. From 01/04/1997-30/11/2016, 13 945 people with gout initiated allopurinol with colchicine prophylaxis (mean age 63.9 (SD 14.7) years, 78.2% male). One quarter (26%, 95% CI 25% to 27%) were prescribed ≥1 potentially interacting medicines, most commonly statins (21%, 95% CI 20% to 22%). Statins were not associated with increased adverse events, although other drugs were associated with some adverse outcomes. Diarrhoea and MI were associated with more comorbidities and more severe CKD. People were given colchicine prophylaxis despite commonly having preexisting prescriptions for medications with potential to interact with colchicine. Adverse events were more common in people who had more comorbidities and certain potentially interacting medications. Our findings will provide much-needed information about prognostic factors for colchicine-related adverse events that can inform treatment decisions about prophylaxis when initiating allopurinol.

Association of day-case rates with post COVID-19 recovery of elective laparoscopic cholecystectomy activity across England.

The aim of this study was to investigate the safety of day-case laparoscopic cholecystectomy, and the association between day-case rates and, post the COVID-19 pandemic, recovery of activity to prepandemic levels for integrated care boards (ICBs) in England. This was a retrospective observational study of the Hospital Episodes Statistics (HES) data set. Elective laparoscopic cholecystectomies for the period 1 January 2019 to 31 December 2022 were identified. Activity levels for 2022 were compared with those for the whole of 2019 (baseline). Day-case activity was identified where the length of stay recorded in the HES was zero days. Data were available for 184,252 patients across the 42 ICBs in England, of which 120,408 (65.3%) were day-case procedures. By December 2022, activity levels for the whole of England had returned to 88.2% of prepandemic levels. The South West region stood out as having recovered activity levels to the greatest extent, with activity at 97.3% of prepandemic levels during 2022. The South West also had the highest postpandemic day-case rate at 74.9% of all patients seen as a day-case during 2022; this compares with an England average of 65.3%. At an ICB level, there was a significant correlation between day-case rates and postpandemic activity levels (r = 0.362, p = 0.019). There was no strong or consistent evidence that day-case surgery had poorer patient outcomes than inpatient surgery. Recovery of elective laparoscopic cholecystectomy activity has been better in South West England than in other regions. Increasing day-case rates may be important if ICBs in other regions are to increase activity levels up to and beyond prepandemic levels.

Factors associated with conversion from day-case to in-patient elective inguinal hernia repair surgery across England: an observational study using administrative data.

Elective primary inguinal hernia repair surgery is increasingly being conducted as a day-case procedure. However, some patients planned for day-case surgery have to stay in hospital for at least one night. The aim of this study was to identify the factors associated with conversion from day-case to in-patient management for elective inguinal hernia repair surgery. This was an exploratory retrospective analysis of observational data from the Hospital Episode Statistics dataset for England. All patients aged ≥ 17years undergoing a first elective inguinal hernia repair between 1st April 2014 and 31st March 2022 that was planned as day-case surgery were identified. The exposure of interest was discharged on the day of admission (day-case) or requiring overnight stay. The primary outcome of interest was 30-day emergency readmission with an overnight stay. For reporting, providers were aggregated to an Integrated Care Board (ICB) level. A total of 351,528 planned day-caseelective primary inguinal hernia repairs were identified over the eight-year study period. Of these, 45,305 (12.9%) stayed in hospital for at least one night and were classed as day-case to in-patient stay conversions. Patients who converted to in-patient stay were older, had more comorbidities, and were more likely to have bilateral surgery and be operated on by a low-annual volume surgeon. Post-procedural complications were strongly associated with conversion. Across the 42 ICBs in England, model-adjusted conversion rates varied from 3.3% to 21.3%. There was considerable variation in conversion to in-patient stay rates for inguinal hernia repair across ICBs in England. Our findings should help surgical teams to better identify patients suitable for day-case inguinal hernia repair and plan discharge services more effectively. This should help to reduce the variation in conversion rates.

Comparing reported management of ureteric stones between clinical audit and administrative datasets: An opportunity to streamline clinical audit

ObjectivesTo compare recorded patient management between a clinical audit and administrative dataset for patients presenting with ureteric stones in England and to assess the feasibility of using administrative data for routine audit. Patients and methodsThe British Association of Urological Surgeons conducted a clinical audit of all patients presenting as an emergency to 107 hospitals in England during November 2020 with ureteric stones. All patients were followed up until 31st March 2021 and in-patient and out-patient management received recorded. These clinical audit data were compared to those available from the English Hospital Episode Statistics (HES) administrative database covering the same time period. ResultsData were available for 2344 patients from HES, and 2050 patients admitted to the same 107 hospitals from clinical audit. The two cohorts were well matched for age (mean 47.2 years and 49.3 years respectively), but with a higher proportion of females in the HES dataset (42.2 % vs 30.1 %). Recorded treatment received was similar in both cohorts, other than for ureteroscopy, which was significantly under recorded in HES, most obviously following initial stent placement (17.2 % vs 26.0 % ureteroscopy as final management respectively). ConclusionsThe two data sources were generally well matched in terms of patient numbers, age and management. The higher number of patients and females in HES may be due to initial misdiagnosis of abdominal pain as ureteric stones in females. The reasons for discrepancies in recording of ureteroscopy are unclear and warrant further investigation. Administrative data can complement clinical audit data and streamline the audit process, but issues around data quality should be studied prior to use of administrative data for this purpose.

The Clinical and Economic Burden of Antibiotic Use in Pediatric Patients With Varicella Infection: A Retrospective Cohort Analysis of Real-World Data in England.

Varicella is a highly infectious disease, particularly affecting children, that can lead to complications requiring antibiotics or hospitalization. Antibiotic use for varicella management is poorly documented. This study assessed antibiotic use for varicella and its complications in a pediatric population in England. Data were drawn from medical records in the Clinical Practice Research Datalink and Hospital Episode Statistics data sets. The study included patients <18 years old with varicella diagnosed during 2014-2018 and 3-month follow-up available. We determined varicella-related complications, medication use, healthcare resource utilization, and costs from diagnosis until 3 months after diagnosis. We identified 114 578 children with a primary varicella diagnosis. Of these, 7.7% (n = 8814) had a varicella-related complication, the most common being ear, nose, and throat related (37.1% [n = 3271]). In all, 25.9% (n = 29 706 of 114 578) were prescribed antibiotics. A higher proportion of patients with complications than without complications were prescribed antibiotics (64.3% [n = 5668 of 8814] vs 22.7% [n = 24 038 of 105 764]). Mean annualized varicella-related costs were £2 231 481 for the study cohort. Overall, antibiotic prescriptions cost approximately £262 007. This study highlights high antibiotic use and healthcare resource utilization associated with varicella management, particularly in patients with complications. A national varicella vaccination program in England may reduce varicella burden and related complications, medication use, and costs.

Safety of colchicine and NSAID prophylaxis when initiating urate-lowering therapy for gout: propensity score-matched cohort studies in the UK Clinical Practice Research Datalink

ObjectivesTo determine the risk of adverse events associated with colchicine or non-steroidal anti-inflammatory drug (NSAID) prophylaxis when initiating allopurinol for gout.MethodsWe conducted two matched retrospective cohort studies in linked UK...

Secondary analysis of child hospital admission data for dental caries in London, UK: what the data tells us about oral health inequalities

ObjectivesDental caries is the most common reason for hospital admissions for children aged 6–10 years in England. The prevalence in the experience of hospital admission is not uniform across all...

Which people are most affected by changes to data linkage methodology? An exploration of patient, organisational and spatiotemporal characteristics in administrative hospital data in England

ObjectivesIn 2021, NHS Digital changed the process used to link records belonging to the same person across and within data collections. Our objectives were to identify patient, organisational and spatiotemporal characteristics associated with records impacted by this change and the implications for researchers using this data.&#x0D; MethodsWe used an observational cohort study of patients, aged 55 or less, with a secondary care contact recorded in any of the NHS Digital (now part of NHS England) curated Hospital Episode Statistics (HES) datasets between April 1997 and March 2021. We compared clusters of records assigned to each patient using the HES ID (old methodology using a three-step deterministic algorithm) and the Person ID (new methodology using a master patient spine). We used multivariable logistic regression to identify patient, organisational and spatiotemporal (such as area-level deprivation and year of first contact) characteristics associated with patients whose cluster had changed.&#x0D; ResultsOf 88 million hospital records in 2019, there were 18,968,711 distinct HES IDs and 18,717,142 distinct TPIs. Of the 12,701,169 HES IDs with more than one record, 145,948 (1.1%) were split into multiple Person IDs. Of the 12,999,671 Person IDs with more than one record, 483,091 (3.7%) were associated with two or more merged HES IDs. We will present an analysis using data covering the period April 1997 to March 2021 - 1.25 billion records - and present the characteristics associated with changes between linkage methods.&#x0D; ConclusionOur findings indicate that this change consolidated clusters, resulting in fewer distinct individuals in the data. Our findings will inform researchers about which groups of individuals are most likely to be affected by changes to linkage methodology. This is vital for understanding potential sources of bias due to linkage error.

Association between weight loss and health care resource utilization in adults living with obesity: Evidence from a UK primary care database.

We investigated the impact of intentional weight loss on health care resource utilization (HCRU) and costs among people with obesity. This retrospective, observational cohort study used data from the Clinical Practice Research Datalink (CPRD) GOLD database. Adults >18 years at index date [first recorded body mass index (BMI) of 30-50 kg/m2 between 2006 and 2015 with a further BMI record 4 years later] were assigned to an intentional weight loss cohort (-25% to -10% BMI change) or a stable weight cohort (-3% to +3%), based on their BMI change during a 4-year baseline period from index date. Evidence of intention to lose weight during the baseline period was required. Linked Hospital Episode Statistics datasets captured HCRU and costs over an 8-year follow-up period. Mixed effects models adjusted for demographics, total costs during baseline and baseline comorbidities were used. Baseline characteristics were similar between cohorts with weight loss (n = 8676) and stable weight (n = 44 519). Over follow-up, the weight loss cohort experienced a significantly lower mean annual increase in total costs [2.1% (95% confidence interval: 1.3-2.8)] than the stable weight cohort [4.3% (95% confidence interval: 4.0-4.6); p < .0001]. Weight loss was associated with a lower mean annual increase in multiple HCRU and cost components compared with maintaining a stable high weight. Our findings suggest that intentional weight loss of 10-25% is associated with lower HCRU and costs in the long term among individuals living with obesity, relative to stable weight.

A comprehensive and integrated hospital decision support system for efficient and effective healthcare services delivery using discrete event simulation

The difficulty that hospital management has been experiencing over the past decade in balancing demand and capacity needs is unprecedented in the United Kingdom. Due to a shortage of capacity, hospitals cannot treat all patients. We developed a whole hospital-level decision support system to assess and respond to the needs of local populations. We integrated a comparative forecasting approach and discrete event simulation modelling using Hospital Episode Statistics and local datasets. It is clear from the literature that this level of whole hospital simulation model has never been developed before (an innovative decision support system). First, the demands of all hospital specialties were forecasted, and the forecasts were embedded into the simulation model as input. Secondly, a simulation model was developed to capture the patient pathway of all specialties. The model integrates every component of a hospital to aid with efficient and effective use of scarce resources (e.g., staff and beds). As a result, the hospital can meet the increasing demand with its current resources. According to the scenario analysis, the hospital bed occupancy rate will reach the national target (i.e., 85%), and the total hospital revenue will increase by approximately 13%, with a 10% increase in A&E and outpatient and a 20% increase in inpatient demand. In conclusion, the hospital-level simulation model can become a crucial instrument for decision-makers to provide an efficient service for hospitals in England and other parts of the world.

Key causes and long-term trends related to emergency department and inpatient hospital admissions of homeless persons in England

BackgroundIt is estimated that approximately 300,000 people are experiencing homelessness in England. The aim of this study was to evaluate key causes and long-term trends of emergency departments (EDs) and in hospital inpatient admissions of persons experiencing homelessness in England.MethodsED and hospital inpatient admissions data were obtained from Hospital Episode Statistics (HES) covering all National Health Service (NHS) England hospitals. Anyone identified or declared to be experiencing homelessness during the service usage are recorded in HES datasets. Data were extracted for the 10-year study period and compared to the general population, which includes all patients attending the ED or admitted to inpatient care in England.ResultsDrug- and alcohol-related causes contribute to the most frequent reasons for attendance and admissions of persons experiencing homelessness in the ED and inpatient respectively. A total of 30,406 ED attendances were recorded for persons experiencing homelessness in the year 2018/2019 (+ 44.9% rise vs 2009/10) of which injuries and poisoning respectively represented 21.8% and 17.9% of all persons experiencing homelessness presentations to the ED. Poisoning (including drug overdose) represented only 1.9% of all attendances by the general population during the same study year (rate ratio vs general populations 9.2 95% CI 9.0–9.4). High mortality rates were observed in relation to presentations attributed to drug- and alcohol-related causes. A total of 14,858 persons experiencing homelessness inpatient admissions were recorded in 2018/2019 (+ 68.6% vs 2009/2010). Psychoactive substance use constituted 12.7% of all admissions in 2018/2019 compared to 0.4% of in the general populations (rate ratio: 33.3, 95% CI: 31.9–34.7). There was a 44.3% rise in the number of admissions related to poisoning in the study period amongst persons experiencing homelessness in England (vs 14.2% in general population).ConclusionMarked disparities around primary causes of ED and inpatient admissions were identified between persons experiencing homelessness and the general population. There is a continued need for prevention measures to reduce the prevalence of drug and alcohol, injury and poisoning-related admissions to the ED, enhanced service provision at the community level, and multisector collaborations. These initiatives should maximise opportunities for early interventions and improve outcomes for persons experiencing homelessness, including increased accessibility of healthcare and mental health services, particularly in areas that demonstrate increasing ED and inpatient attendance rates over time.

Induction of labour at 39 weeks and adverse outcomes in low-risk pregnancies according to ethnicity, socioeconomic deprivation, and parity: A national cohort study in England.

Ethnic and socioeconomic inequalities in obstetric outcomes are well established. However, the role of induction of labour (IOL) to reduce these inequalities is controversial, in part due to insufficient evidence. This national cohort study aimed to identify adverse perinatal outcomes associated with IOL with birth at 39 weeks of gestation ("IOL group") compared to expectant management ("expectant management group") according to maternal characteristics in women with low-risk pregnancies. All English National Health Service (NHS) hospital births between January 2018 and March 2021 were examined. Using the Hospital Episode Statistics (HES) dataset, maternal and neonatal data (demographic, diagnoses, procedures, labour, and birth details) were linked, with neonatal mortality data from the Office for National Statistics (ONS). Women with a low-risk pregnancy were identified by excluding pregnancies with preexisting comorbidities, previous cesarean section, breech presentation, placenta previa, gestational diabetes, or a baby with congenital abnormalities. Women with premature rupture of membranes, placental abruption, hypertensive disorders of pregnancy, amniotic fluid abnormalities, or antepartum stillbirth were excluded only from the IOL group. Adverse perinatal outcome was defined as stillbirth, neonatal death, or neonatal morbidity, the latter identified using the English composite neonatal outcome indicator (E-NAOI). Binomial regression models estimated risk differences (with 95% confidence intervals (CIs)) between the IOL group and the expectant management group, adjusting for ethnicity, socioeconomic background, maternal age, parity, year of birth, and birthweight centile. Interaction tests examined risk differences according to ethnicity, socioeconomic background, and parity. Of the 1 567 004 women with singleton pregnancies, 501 072 women with low-risk pregnancies and with sufficient data quality were included in the analysis. Approximately 3.3% of births in the IOL group (1 555/47 352) and 3.6% in the expectant management group (16 525/453 720) had an adverse perinatal outcome. After adjustment, a lower risk of adverse perinatal outcomes was found in the IOL group (risk difference -0.28%; 95% CI -0.43%, -0.12%; p = 0.001). This risk difference varied according to socioeconomic background from 0.38% (-0.08%, 0.83%) in the least deprived to -0.48% (-0.76%, -0.20%) in the most deprived national quintile (p-value for interaction = 0.01) and by parity with risk difference of -0.54% (-0.80%, -0.27%) in nulliparous women and -0.15% (-0.35%, 0.04%) in multiparous women (p-value for interaction = 0.02). There was no statistically significant evidence that risk differences varied according to ethnicity (p = 0.19). Key limitations included absence of additional confounding factors such as smoking, BMI, and the indication for induction in the HES datasets, which may mean some higher risk pregnancies were included. IOL with birth at 39 weeks was associated with a small reduction in the risk of adverse perinatal outcomes, with 360 inductions in low-risk pregnancies needed to avoid 1 adverse outcome. The risk reduction was mainly present in women from more socioeconomically deprived areas and in nulliparous women. There was no significant risk difference found by ethnicity. Increased uptake of IOL at 39 weeks, especially in women from more socioeconomically deprived areas, may help reduce inequalities in adverse perinatal outcomes.

Comparison of outcomes for patients with and without a serious mental illness presenting to hospital for chronic obstruction pulmonary disease: retrospective observational study using administrative data.

There are few data on the profile of those with serious mental illness (SMI) admitted to hospital for physical health reasons. To compare outcomes for patients with and without an SMI admitted to hospital in England where the primary reason for admission was chronic obstructive pulmonary disease (COPD). This was a retrospective, observational analysis of the English Hospital Episodes Statistics data-set for the period from 1 April 2018 to 31 March 2019, for patients aged 18-74 years with COPD as the dominant reason for admission. Patient with an SMI (psychosis spectrum disorder, bipolar disorder) were identified. Data were available for 54 578 patients, of whom 2096 (3.8%) had an SMI. Patients with an SMI were younger, more likely to be female and more likely to live in deprived areas than those without an SMI. The burden of comorbidity was similar between the two groups. After adjusting for covariates, SMI was associated with significantly greater risk of length of stay than the median (odds ratio 1.24, 95% CI 1.12-1.37, P ≤ 0.001) and with 30-day emergency readmission (odds ratio 1.51, 95% confidence interval 1.34-1.69, P ≤ 0.001) but not with in-hospital mortality. Clinicians should be aware of the potential for poorer outcomes in patients with an SMI even when the SMI is not the primary reason for admission. Collaborative working across mental and physical healthcare provision may facilitate improved outcomes for people with SMI.

P49 Epidemiology of skin cancer by ethnicity in England, 2013–2020: a cohort study

Abstract Skin cancer rates are increasing worldwide and predominantly affect people of white ethnicity. Although less common, skin cancers in people with skin of colour (SOC) are more likely to experience diagnostic delays and higher mortality. High-quality epidemiological data on skin cancers in SOC are limited. SOC and skin phototype are not recorded in registry data; however, the ethnic group can be used as a somewhat limited proxy. The aim of this retrospective national cancer registry cohort study was to describe skin cancer epidemiology by the ethnic group. We identified a cohort of patients diagnosed with skin cancer in England (2013–2020) from the National Disease Registration Service (NDRS). The ethnic group is routinely recorded as self-reported during hospital visits in Patient Administration System and Hospital Episode Statistics datasets collected by the NDRS. We analysed data by the broad ethnic group [White, Asian (excluding Chinese), Black, Chinese, mixed, other, unknown]. We report counts, incidence, stage at presentation (melanoma only), referral pathway, anatomical location and geographical variation. The first basal cell carcinoma (BCC) and cutaneous squamous cell carcinoma (cSCC) per patient were reported. More cases of skin cancers were diagnosed in Asian ethnicities than in Black and Chinese ethnicities, except for melanoma, which was more frequently diagnosed in Black ethnicities. Of 713 615 BCCs, 571 396 (80.1%) were reported in White, 799 (0.1%) in Asian, 634 (0.1%) in mixed, 413 (0.1%) in Black, 156 (0.02%) in Chinese and 3122 (0.4%) in people of ‘other’ ethnicities; 137 095 (19.2%) were of ‘unknown’ ethnicity. Of 244 072 cSCCs, 225 534 (92.4%) were in White, 609 (0.2%) in Asian, 354 (0.1%) in Black, 279 (0.1%) in mixed, 66 (0.03%) in Chinese and 1137 (0.5%) in people of ‘other’ ethnicity; 16 093 (6.6%) were of ‘unknown’ ethnicity. Of 111 676 melanomas, 102 273 (91.6%) were in White, 199 (0.2%) in Black, 180 (0.2%) in Asian, 125 (0.1%) in mixed, 25 (0.02%) in Chinese and 695 (0.6%) in ‘other’ ethnicities; 8179 (7.3%) were of ‘unknown’ ethnicity. This is the first national study of skin cancer types by ethnicity in England. National datasets allow for the better evaluation of rare subgroups of patients. The main limitation is the high proportion of unknown ethnicity reporting, which is higher for keratinocyte cancers than for other cancers, presumably due to their automated processing and lack of data sources (i.e. inpatient or multidisciplinary team datasets). These data will support a better understanding of unmet clinical needs in underserved populations.

P203 Comparison of coding of giant cell arteritis in primary and secondary care records: an observational feasibility study

Abstract Background/Aims Giant cell arteritis (GCA) is considered a medical emergency, and relapses are common, affecting around 50% of patients. Many of the early symptoms of GCA are non-specific and highly prevalent in older people, making the diagnosis challenging. The Clinical Practice Research Datalink (CPRD) is a large primary care database that provides the opportunity to study large numbers of people with GCA. General practitioners commonly identify suspected GCA and refer patients to secondary care for specialist evaluation and diagnostic confirmation. We sought to understand how primary and secondary care records correspond. Methods A retrospective observational study was undertaken using data from 2003 to 2020 by linking CPRD Aurum to Hospital Episode Statistics (HES) Admitted Patient Care (APC) and Outpatient (OP) data. We included all permanently registered patients aged 50 years and over in CPRD Aurum who could be linked to HES datasets. We identified individuals with a SNOMED code for GCA in CPRD Aurum and compared them to a list of patients who received an ICD-10 code for GCA in HES APC and OP datasets. We then restricted our comparison to the time-periods 90 and 180 days before or after the first record of GCA in the CPRD Aurum dataset (index code). Results We identified 18,806 individuals with a SNOMED code for GCA in the CPRD Aurum dataset between 2003 and 2020. 9755 (51.87%) of these individuals also received at least one secondary care code for GCA in the HES APC and OP data. 3914 (20.86%) and 4573 (24.37%) individuals received at least one secondary care code for GCA within 90 and 180 days, respectively, of the index primary care code. 7520 (58.96%) of those with two codes in primary care also had at least one code in secondary care, and 6090 (61.09%) of those with three codes. Conclusion Approximately one-half of individuals with an index code in primary care had a secondary care code for GCA. However, less than a quarter of patients had secondary care codes within 90 and 180 days of their index code in primary care. Given that it is unlikely that coding in either primary or secondary care is inaccurate, we conclude that this method of comparing GCA diagnoses is not a suitable approach to assess the accuracy of primary care coding. Further research could explore alternative approaches to understanding the quality of the coding of GCA in primary and secondary care data. Disclosure N. Padmanabhan: Other; NP is an NIHR Academic Clinical Fellow. S. Muller: Other; SM is part-funded by the NIHR Applied Research Collaboration West Midlands (ARC WM). J. Bailey: None. T. Helliwell: None. S.L. Mackie: Consultancies; SLM has provided consultancy on behalf of her institution for Roche/Chugai, Sanofi, AbbVie and AstraZeneca. Other; SLM has given talks on behalf of her institution for Pfizer, Vifor and UCB. In all cases fees were paid to her institution and no personal fees were received, SLM was supported by Roche to attend EULAR2019 and by Pfizer to attend ACR2021 virtually, SLM has been an investigator on clinical trials for Sanofi and GSK. As a research grant co-applicant, SLM has received research funding (partial salary support) from Vifor, paid to her institution. C.D. Mallen: Other; CDM is funded by the NIHR School for Primary Care Research and the NIHR Applied Research Collaboration West Midlands (ARC WM). E. Roddy: None.

Completeness, agreement, and representativeness of ethnicity recording in the United Kingdom’s Clinical Practice Research Datalink (CPRD) and linked Hospital Episode Statistics (HES)

BackgroundThis descriptive study assessed the completeness, agreement, and representativeness of ethnicity recording in the United Kingdom (UK) Clinical Practice Research Datalink (CPRD) primary care databases alone and, for those patients registered with a GP in England, when linked to secondary care data from Hospital Episode Statistics (HES).MethodsEthnicity records were assessed for all patients in the May 2021 builds of the CPRD GOLD and CPRD Aurum databases for all UK patients. In analyses of the UK, English data was from combined CPRD-HES, whereas data from Northern Ireland, Scotland, and Wales drew from CPRD only. The agreement of ethnicity records per patient was assessed within each dataset (CPRD GOLD, CPRD Aurum, and HES datasets) and between datasets at the highest level ethnicity categorisation (‘Asian’, ‘black’, ‘mixed’, ‘white’, ‘other’). Representativeness was assessed by comparing the ethnic distributions at the highest-level categorisation of CPRD-HES to those from the Census 2011 across the UK’s devolved administrations. Additionally, CPRD-HES was compared to the experimental ethnic distributions for England and Wales from the Office for National Statistics in 2019 (ONS2019) and the English ethnic distribution from May 2021 from NHS Digital’s General Practice Extraction Service Data for Pandemic Planning and Research with HES data linkage (GDPPR-HES).ResultsIn CPRD-HES, 81.7% of currently registered patients in the UK had ethnicity recorded in primary care. For patients with multiple ethnicity records, mismatched ethnicity within individual primary and secondary care datasets was < 10%. Of English patients with ethnicity recorded in both CPRD and HES, 93.3% of records matched at the highest-level categorisation; however, the level of agreement was markedly lower in the ‘mixed’ and ‘other’ ethnic groups. CPRD-HES was less proportionately ‘white’ compared to the UK Census 2011 (80.3% vs. 87.2%) and experimental ONS2019 data (80.4% vs. 84.3%). CPRD-HES was aligned with the ethnic distribution from GDPPR-HES (‘white’ 80.4% vs. 80.7%); however, with a smaller proportion classified as ‘other’ (1.1% vs. 2.8%).ConclusionsCPRD-HES has suitable representation of all ethnic categories with some overrepresentation of minority ethnic groups and a smaller proportion classified as ‘other’ compared to the UK general population from other data sources. CPRD-HES data is useful for studying health risks and outcomes in typically underrepresented groups.

Data quality and autism: Issues and potential impacts

IntroductionLarge healthcare datasets can provide insight that has the potential to improve outcomes for patients. However, it is important to understand the strengths and limitations of such datasets so that the insights they provide are accurate and useful. The aim of this study was to identify data inconsistencies within the Hospital Episodes Statistics (HES) dataset for autistic patients and assess potential biases introduced through these inconsistencies and their impact on patient outcomes. The study can only identify inconsistencies in recording of autism diagnosis and not whether the inclusion or exclusion of the autism diagnosis is the error. MethodsData were extracted from the HES database for the period 1st April 2013 to 31st March 2021 for patients with a diagnosis of autism. First spells in hospital during the study period were identified for each patient and these were linked to any subsequent spell in hospital for the same patient. Data inconsistencies were recorded where autism was not recorded as a diagnosis in a subsequent spell. Features associated with data inconsistencies were identified using a random forest classifiers and regression modelling. ResultsData were available for 172,324 unique patients who had been recorded as having an autism diagnosis on first admission. In total, 43.7 % of subsequent spells were found to have inconsistencies. The features most strongly associated with inconsistencies included greater age, greater deprivation, longer time since the first spell, change in provider, shorter length of stay, being female and a change in the main specialty description. The random forest algorithm had an area under the receiver operating characteristic curve of 0.864 (95 % CI [0.862 – 0.866]) in predicting a data inconsistency. For patients who died in hospital, inconsistencies in their final spell were significantly associated with being 80 years and over, being female, greater deprivation and use of a palliative care code in the death spell. ConclusionsData inconsistencies in the HES database were relatively common in autistic patients and were associated a number of patient and hospital admission characteristics. Such inconsistencies have the potential to distort our understanding of service use in key demographic groups.

The burden of selected vaccine-preventable diseases on the secondary care health system in England: Results from a five-year administrative healthcare dataset

ObjectivesThis study examined healthcare resource use (HCRU) for selected vaccine-preventable diseases (VPD) in secondary care in England. MethodsThe hospital episode statistics (HES) dataset covering all secondary care interactions within the English National Health Service (NHS) from 2015 to 2021 was used to identify and track HCRU for patients with a primary or secondary diagnosis for pertussis and Haemophilus influenzae type b (Hib), or a primary diagnosis only for hepatitis B, diphtheria, poliomyelitis, or tetanus. The first documented diagnosis during the study period (01/04/2015–31/03/2021) was the index event. Results7,274 patients with a total of 5,554,343 patient-days (mean follow up 1,491 days) were included. The total number of hospital admissions was 27,092 and total inpatient cost was £4,987,770, with hepatitis B making up ∼80 % of this. Mean outpatient hospital appointments per patient were highest for tetanus (4.00), but total outpatient A&E cost burden was highest for Hib (£643,343 [mean per attendance £144.57]). For patients 0–9 years of age (n = 1,917), pertussis (n = 1,547) and Hib (n = 313) were by far the most commonly coded diseases. Hepatitis B was the most common disease in adults of working age and Hib was most prevalent in adults of retirement age. Surprisingly, poliomyelitis was observed in the database potentially due to historic diagnoses and/or coding inaccuracy. Other discrepancies with surveillance data were noted. ConclusionsVPDs impose a large burden on the NHS, but there is potential to reduce this and improve public health by optimising vaccination schedules, improving access and ensuring high coverage rates.

Data consistency in the English Hospital Episodes Statistics database

BackgroundTo gain maximum insight from large administrative healthcare datasets it is important to understand their data quality. Although a gold standard against which to assess criterion validity rarely exists for...

Optimization of Timing of Maternal Pertussis Immunization From 6 Years of Postimplementation Surveillance Data in England.

England's third-trimester maternal pertussis vaccination, introduced in October 2012, was extended to the second trimester in 2016. Maternal vaccination provides high protection against infant disease, but routine second-trimester vaccination has not previously been assessed. National laboratory-confirmed pertussis case surveillance determined vaccination history, maternal vaccination history and hospitalization. Pertussis hospital admissions between 2012 and 2019 were extracted from the Hospital Episode Statistics data set. Vaccine effectiveness (VE) was calculated for pertussis case patients born between October 2012 and September 2018 using the screening method and matching with a nationally representative data set. Higher coverage was observed after earlier maternal vaccination with approximately 40% of pregnant women vaccinated ≥13 weeks before delivery. Cases and hospitalizations stabilized at low levels in younger infants but remained elevated in older infants, children, and adults. No deaths occurred in infants with vaccinated mothers after 2016. Of 1162 laboratory-confirmed pertussis cases in the study, 599 (52%) were in infants aged <93 days: 463 (77%) with unvaccinated and 136 (23%) with vaccinated mothers. The VE was equivalent in infants with mothers vaccinated at different gestational periods except in those with mothers vaccinated between 7 days before and 41 days after delivery. Children whose mothers were unvaccinated but with vaccination in a previous pregnancy had a VE against disease of 44% (95% confidence interval, 19%-75%). There was no increased disease risk after primary vaccination in children with mothers vaccinated at least 7 days before delivery. National policy recommending vaccination in the second trimester increased earlier maternal vaccine uptake with sustained high VE and impact against early infant disease.