Dataset Information

How do time trends in inhospital mortality compare? A retrospective study of England and Scotland over 17 years using administrative data.

ABSTRACT:

Objectives

To examine the trends in inhospital mortality for England and Scotland over a 17-year period to determine whether and if so to what extent the time trends differ after controlling for differences in the patients treated.

Design

Analysis of retrospective administrative hospital data using descriptive aggregate statistics of trends in inhospital mortality and estimates of a logistic regression model of individual patient-level inhospital mortality accounting for patient characteristics, case-mix, and country-specific and year-specific intercepts.

Setting

Secondary care across all hospitals in England and Scotland from 1997 to 2013.

Population

Over 190?million inpatient admissions, either electively or emergency, in England or Scotland from 1997 to 2013.

Data

Hospital Episode Statistics for England and the Scottish Morbidity Record 01 for Scotland.

Main outcome measures

Separately for two admission pathways (elective and emergency), we examine aggregate time trends of the proportion of patients who die in hospital and a binary variable indicating whether an individual patient died in hospital or survived, and how that indicator is influenced by the patient's characteristics, the year and the country (England or Scotland) in which they were admitted.

Results

Inhospital mortality has declined in both countries over the period studied, for both elective and emergency admissions, but has declined more in England than Scotland. The difference in trend reduction is greater for elective admissions. These differences persist after controlling for patient characteristics and case-mix.

Conclusions

Comparing data at country level suggests questions about the roles performed by or functioning of their healthcare systems. We found substantial differences between Scotland and England in regard to the trend reductions in inhospital mortality. Hospital resources are therefore being deployed increasingly differently over time in these two countries for reasons that have yet to be explained.

SUBMITTER: Aragon MJ

PROVIDER: S-EPMC5855481 | biostudies-literature | 2018 Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

How do time trends in inhospital mortality compare? A retrospective study of England and Scotland over 17 years using administrative data.

Aragón María José MJ Chalkley Martin M

BMJ open 20180221 2

<h4>Objectives</h4>To examine the trends in inhospital mortality for England and Scotland over a 17-year period to determine whether and if so to what extent the time trends differ after controlling for differences in the patients treated.<h4>Design</h4>Analysis of retrospective administrative hospital data using descriptive aggregate statistics of trends in inhospital mortality and estimates of a logistic regression model of individual patient-level inhospital mortality accounting for patient c ...[more]

PMID: 29467130

Similar Datasets

Project description:Scotland has disproportionately high rates of suicide compared with England. An analysis of trends may help reveal whether rates appear driven more by birth cohort, period or age. A 'birth cohort effect' for England & Wales has been previously reported by Gunnell et al. (B J Psych 182:164-70, 2003). This study replicates this analysis for Scotland, makes comparisons between the countries, and provides information on 'vulnerable' cohorts.Suicide and corresponding general population data were obtained from the National Records of Scotland, 1950 to 2014. Age and gender specific mortality rates were estimated. Age, period and cohort patterns were explored graphically by trend analysis.A pattern was found whereby successive male birth cohorts born after 1940 experienced higher suicide rates, in increasingly younger age groups, echoing findings reported for England & Wales. Young men (aged 20-39) were found to have a marked and statistically significant increase in suicide between those in the 1960 and 1965 birth cohorts. The 1965 cohort peaked in suicide rate aged 35-39, and the subsequent 1970 cohort peaked even younger, aged 25-29; it is possible that these 1965 and 1970 cohorts are at greater mass vulnerability to suicide than earlier cohorts. This was reflected in data for England & Wales, but to a lesser extent. Suicide rates associated with male birth cohorts subsequent to 1975 were less severe, and not statistically significantly different from earlier cohorts, suggestive of an amelioration of any possible influential 'cohort' effect. Scottish female suicide rates for all age groups converged and stabilised over time. Women have not been as affected as men, with less variation in patterns by different birth cohorts and with a much less convincing corresponding pattern suggestive of a 'cohort' effect.Trend analysis is useful in identifying 'vulnerable' cohorts, providing opportunities to develop suicide prevention strategies addressing these cohorts as they age.

Project description:Introduction"Big data" - including linked administrative data - can be exploited to evaluate interventions for maternal and child health, providing time- and cost-effective alternatives to randomised controlled trials. However, using these data to evaluate population-level interventions can be challenging.ObjectivesWe aimed to inform future evaluations of complex interventions by describing sources of bias, lessons learned, and suggestions for improvements, based on two observational studies using linked administrative data from health, education and social care sectors to evaluate the Family Nurse Partnership (FNP) in England and Scotland.MethodsWe first considered how different sources of potential bias within the administrative data could affect results of the evaluations. We explored how each study design addressed these sources of bias using maternal confounders captured in the data. We then determined what additional information could be captured at each step of the complex intervention to enable analysts to minimise bias and maximise comparability between intervention and usual care groups, so that any observed differences can be attributed to the intervention.ResultsLessons learned include the need for i) detailed data on intervention activity (dates/geography) and usual care; ii) improved information on data linkage quality to accurately characterise control groups; iii) more efficient provision of linked data to ensure timeliness of results; iv) better measurement of confounding characteristics affecting who is eligible, approached and enrolled.ConclusionsLinked administrative data are a valuable resource for evaluations of the FNP national programme and other complex population-level interventions. However, information on local programme delivery and usual care are required to account for biases that characterise those who receive the intervention, and to inform understanding of mechanisms of effect. National, ongoing, robust evaluations of complex public health evaluations would be more achievable if programme implementation was integrated with improved national and local data collection, and robust quasi-experimental designs.

Project description:BackgroundWhile opioid analgesics are often necessary for the management of acute postoperative pain, appropriate prescribing practices are crucial to avoid harm. The aim was to investigate the changes in the proportion of people receiving initial opioid prescriptions after hospital discharge following colectomy, and describe trends and patterns in prescription characteristics.MethodsThis was a retrospective cohort study. Patients undergoing colectomy in England between 2010 and 2019 were included using electronic health record data from linked primary (Clinical Practice Research Datalink Aurum) and secondary (Hospital Episode Statistics) care. The proportion of patients having an initial opioid prescription issued in primary care within 90 days of hospital discharge was calculated. Prescription characteristics of opioid type and formulation were described.ResultsOf 95 155 individuals undergoing colectomy, 15 503 (16.3%) received opioid prescriptions. There was a downward trend in the proportion of patients with no prior opioid exposure (opioid naive) who had a postdischarge opioid prescription (P <0.001), from 11.4% in 2010 to 6.7% in 2019 (-41.3%, P <0.001), whereas the proportions remained stable for those prescribed opioids prior to surgery, from 57.5% in 2010 to 58.3% in 2019 (P = 0.637). Codeine represented 44.5% of all prescriptions and prescribing increased by 14.5% between 2010 and 2019. Prescriptions for morphine and oxycodone rose significantly by 76.6% and 31.0% respectively, while tramadol prescribing dropped by 48.0%. The most commonly prescribed opioid formulations were immediate release (83.9%), followed by modified release (5.8%) and transdermal (3.2%). There was a modest decrease in the prescribing of immediate-release formulations from 86.0% in 2010 to 82.0% in 2019 (P <0.001).ConclusionOver the 10 years studied, there was a changing pattern of opioid prescribing following colectomy, with a decrease in the proportion of opioid-naive patients prescribed postdischarge opioids.

Project description:BackgroundDespite modern intensive care with standardized strategies against acute respiratory distress syndrome (ARDS), Pneumocystis pneumonia (PcP) remains a life-threatening disease with a high mortality rate. Here, we analyzed a large mixed cohort of immunocompromised patients with PcP, with regard to clinical course and treatment, and aimed at identifying predictors of outcome.MethodsThis was a single-center retrospective analysis in a tertiary care institution across 17 years. Diagnosis of PcP required typical clinical features and microbiological confirmation of Pneumocystis jirovecii. Epidemiological, clinical, laboratory and outcome data were collected from patient records.ResultsA total of 52,364 specimens from 7504 patients were sent for microbiological assessment (3653 with clinical suspicion of Pneumocystis pneumonia). PcP was confirmed in 240 patients, about half of them HIV positive (52%). The remaining subjects were either solid organ transplant recipients (16.3%) or suffered from malignancy (15.8%) or autoimmune diseases (11.7%). Of note, 95% of patients with PcP were not receiving chemoprophylaxis. Overall in-hospital mortality was 25.4%, increasing to 58% if ICU admission was required. Multivariable regression identified lactate dehydrogenase (LDH) as predictor of in-hospital mortality (adjusted OR 1.17 (95% CI 1.09-1.27), p < 0.0001). Mortality in LDH quartiles increased from 8% to 49%, and a cutoff value of 495 U/L predicted mortality with sensitivity and specificity of 70%. With regard to treatment, 40% of patients received trimethoprim-sulfamethoxazole at doses that were lower than recommended, and these patients had a higher mortality risk (HR 1.80 (95% CI 1.10-3.44), p = 0.02).ConclusionsPcP remains a life-threatening disease among immunocompromised patients. About half of patients with PcP do not have HIV infection. Initial LDH values might serve as a stratifying tool to identify those patients at high risk of death among patients with HIV and without HIV infection.

Dataset Information

How do time trends in inhospital mortality compare? A retrospective study of England and Scotland over 17 years using administrative data.

Objectives

Design

Setting

Population

Data

Main outcome measures

Results

Conclusions

Publications

How do time trends in inhospital mortality compare? A retrospective study of England and Scotland over 17 years using administrative data.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets