Bernard Silke

MRI in Acutely Ill Medical Patients in an Irish Hospital: Influence on Outcomes and Length of Hospital Stay

PURPOSEnMRI is an important diagnostic tool for acute medical admissions. Its relevance to in-hospital mortality and length of stay (LOS) has been examined at St Jamess Hospital in Dublin, Ireland.nnnMETHODSnAll patients admitted for medical emergencies from 2010 through 2012 were studied (18,534 episodes); any relationship between an MRI request, underlying diagnosis on any in-hospital death, and LOS was examined. Logistic regression with generalized estimating equations, adjusted for correlated observations (readmissions), odds ratio estimates, and zero-truncated Poisson regression for LOS were used.nnnRESULTSnMRI procedures were requested in 8.6% of episodes. The in-hospital mortality rate was significantly higher when MRI was performed (7.8% vs 4.6%, P < .001). The unadjusted odds ratio for in-hospital death during that episode was 1.74 (95% confidence interval, 1.26-2.37; P < .001) compared with episodes without MRI. The hospital stay for those MRI episodes was longer (median, 9.1 days; interquartile range, 4.0-26.8 days) than for non-MRI episodes (median, 5.8 days; interquartile range, 2.2-12.2; P < .001). Each unit increase in MRI waiting time (cutoffs set at 0, 1, 3, 7, and 14 days) gave an estimated increase of 1.12 days in hospital LOS, adjusted for illness severity and comorbidities.nnnCONCLUSIONSnMR imaging identified in a subgroup of emergency patients at higher risk of an in-hospital death. These patients have longer LOS attributable in part to procedure wait times, not merely to illness severity or comorbidities.

Fifteen-year outcomes of an acute medical admission unit

BackgroundThe Acute Medical Admission Unit (AMAU) model of care has been associated with improved short- and medium-term outcomes; whether these improvements are sustained remains unclear. We report on the 15-year outcomes of an AMAU in our institution.MethodsAll emergency medical admissions between 2002 and 2016 were examined and 30-day in-hospital mortality, admission rates, readmission rates and length of stay (LOS) assessed. We used logistic and Poisson regression and margin statistics to evaluate outcomes.ResultsThere were 96,305 admissions in 50,612 patients. By admission, the 30-day in-hospital mortality averaged 5.6% (95% CI 5.4 to 5.7%); there was a relative risk reduction (RRR) of 33.9% between 2002 and 2016, from 7.0 to 4.6% (pu2009=u20090.001), number need to treat (NNT) 41.9. By unique patient the 30-day in-hospital mortality averaged 10.5% (95% CI 10.3 to10.8%); there was a RRR of 61.7% between 2002 and 2016, from 15.1 to 5.8% (pu2009=u20090.001), NNT 10.7. The median LOS was 5.0xa0days (IQR 2.1, 9.8) and was unaltered over time. Deprivation status strongly influenced the admission rate/1000 population increasing from Q1 7.7 (95% CI 7.6 to 7.8) to Q5 37.8 (95% CI 37.6 to 38.0); this showed a slight trend to increase over time. Total readmissions increased as a function of time; early readmissions (<u20094xa0weeks) remained constant 10.5% (95% CI 9.6 to 11.3).ConclusionThe 30-day in-hospital mortality showed a linear trend to reduce over the 15xa0years following the institution of an AMAU; other key parameters were unaltered.

Effect of social deprivation on the admission rate and outcomes of adult respiratory emergency admissions

BACKGROUNDnPatients with respiratory disorders constitute a major source of activity for Acute Medicine. We have examined the impact of Socio-Economic Status (SES) and weather factors on the outcomes (30-day in-hospital mortality) of emergency hospitalisations with a respiratory presentation.nnnMETHODSnAll emergency respiratory admissions to St. James Hospital, Dublin, from 2002 to 2014 were evaluated. Patients were categorized by quintile of Deprivation Index, and evaluated against hospital admission rate (/1000 population) and 30-day in-hospital mortality. Univariate and multivariable risk estimates (Odds Ratios (OR) or Incidence Rate Ratios (IRR)) were calculated, using logistic or zero truncated Poisson regression as appropriate.nnnRESULTSnThere were 32,538 episodes in 14,093 patients, representing 39.5% of medical emergency episodes over the 13-yr period. Deprivation Quintile independently predicted the admission rate, with incidence rate ratios (IRR) of Q3 2.02 (95% CI: 1.27, 3.23), Q4 2.55 (95% CI: 1.35, 4.83) and Q5 5.68 (95% CI: 3.56, 9.06). The 30-day in-hospital mortality for the highest quintile was increased (pxa0<xa00.01), Q5 1.31 (95% CI: 1.07, 1.61). Particulate matter (PM10) was predictive for the top two quintiles (>17.2 and 23.8xa0μg/m3 respectively) with an OR for a worse outcome of Q4 1.22 (95% CI: 1.07, 1.40) and Q5 1.24 (95% CI: 1.08, 1.42). Weather (season) and the daily temperature did not affect the admission rate but were significantly associated with worse outcome.nnnCONCLUSIONnSocio-Economic Status influences the admission rate incidence and hospital mortality of respiratory emergency admissions; local environmental conditions (air pollution and temperature) appear only relevant to the mortality outcomes.

Deprivation status and the hospital costs of an emergency medical admission

BACKGROUNDnDeprivation has been shown to adversely affect health outcomes. However, whether deprivation increases hospitalisation costs is uncertain. We have examined the relationship between deprivation and the costs of emergency medical admissions in a single centre between 2008-2014.nnnMETHODSnWe calculated the total hospital costs of emergency admissions related to their deprivation status, based on area of residence (Electoral Division - small census area). We used truncated Poisson and quantile regression methods to examine relationships between predictor variables and total hospital episode costs.nnnRESULTSnOver the study period, 29,508 episodes were recorded in 15,932 patients. Compared with the least deprived (Q1), the incidence rate ratios (IRR) for annual costs were increased to Q3 1.15 (95% CI: 1.12, 1.19), Q4 2.39 (95% CI: 2.30, 2.49) and Q5 2.76 (95% CI: 2.68, 2.85). The margin statistic cost estimate per thousand population increased from 183.8 K€ in Q1 to 507.9 K€ in Q5. The total bed days/1000 population increased as follows (compared with Q1): Q3 IRR 1.41 (95% CI: 1.37, 1.45), Q4 1.96 (95% CI: 1.89, 2.03) and Q5 3.04 (95% CI: 2.96, 3.12). The margin statistic bed day estimate (/1000 population) increased from 218.7 in Q1 to 664.0 in Q5.nnnCONCLUSIONnDeprivation status had a profound impact on total hospital costs for emergency medical admissions. This was primarily mediated through a tripling of total bed days in the most deprived groups.

Age and the economics of an emergency medical admission-what factors determine costs?

Background: The ageing of the population may be anticipated to increase demand on hospital resources. We have investigated the relationship between hospital episode costs and age profile in a single centre. Methods: All Emergency Medical admissions (33 732 episodes) to an Irish hospital over a 6-year period, categorized into three age groups, were evaluated against total hospital episode costs. Univariate and adjusted incidence rate ratios (IRRs) were calculated using zero truncated Poisson regression. Results: The total hospital episode cost increased with age (P < 0.001). The multi-variable Poisson regression model demonstrated that the most important drivers of overall costs were Acute Illness Severity—IRR 1.36 (95% CI: 1.30, 1.41), Sepsis Status −1.46 (95% CI: 1.42, 1.51) and Chronic Disabling Disease Score -1.25 (95% CI: 1.22, 1.27) and the Age Group as exemplified for those >85 years IRR 1.23 (95% CI: 1.15, 1.32). Conclusion: Total hospital episode costs are a product of clinical complexity with contributions from the Acute Illness Severity, Co-Morbidity, Chronic Disabling Disease Score and Sepsis Status. However age is also an important contributor and an increasing patient age profile will have a predictable impact on total hospital episode costs.

Persons with disability, social deprivation and an emergency medical admission

BackgroundThe community level of disability and social deprivation may result in an emergency hospitalisation; we have examined the annual admission incidence rate for emergency medical conditions in relation to the community prevalence of such factors.MethodsAll emergency medical admissions (96,305 episodes in 50,612 patients) within the institution’s catchment area were examined between 2002 and 2016. The frequency of disability, level of full-time carers and unemployment for the 74 electoral divisions of the catchment area was regressed against admission rates; incidence rate ratios (IRR) were calculated using truncated Poisson regression.ResultsDisability was present in 12.1% of the catchment area population (95% CIu2009=u20099.7–15.0). The annual admission incidence rates/1000 population across disability quintiles for the more affluent areas increased from Q1 7.6 (95% CIu2009=u20097.4–7.8) to Q5 27.3 (95% CIu2009=u200927.0–27.5) and for the more deprived area from Q1 16.6 (95% CIu2009=u200916.4, 16.8) to and Q5 40.4 (95% CIu2009=u200940.1–40.7). Disability status influenced the overall admission IRR (compared with Q1/Q3) for Q4/Q5 1.11 (95% CIu2009=u20091.09–1.13) showing an increased rate of hospitalisation for the more deprived areas. Community disability levels interacted with local area unemployment and frequency of full-time carers; as they increased, a linear relationship between disability and the admission rate incidence was demonstrated.ConclusionLocal catchment area disability prevalence rates in addition to social deprivation factors are an important determinant of the annual incidence rate of emergency medical admissions.

Improved mortality outcomes over time for weekend emergency medical admissions

BackgroundMultiple studies have suggested an association between weekend hospital admissions and mortality. These have been limited by potential residual confounders and a lack of explanation of causation.AimWe previously attributed adverse weekend outcomes to higher acuity; we have re-examined this question for all emergency medical admissions to our institution from 2002 to 2014.MethodsWe divided admissions by a weekday or weekend (Friday to Sunday) hospital arrival. We utilised a multivariate logistic regression model, to determine whether the latter was independently predictive of 30-day in-hospital mortality.ResultsThere were 82,368 admissions in 44,628 patients over the 13-year period. Of admissions, 37.4% occurred at the weekend. The Acute Illness Severity Score, the Charlson Co-morbidity Index and the Chronic Disabling Disease Score were similar by a weekday or weekend admission. The multivariable logistic regression showed no increase in 30-day in-hospital mortality for weekend admissions, odds ratio 1.07 (95% confidence interval 0.98 to 1.16) (pxa0=xa00.11). Since the inception of the AMAU, the per patient mortality for a weekend admission has declined from 13.5% in 2002 to 4.4% in 2014. This represents a relative risk reduction of 67.9% with a number needed to treat of 10.8. Outcomes improved similarly for weekday and weekend admissions.ConclusionNo increase in 30-day in-hospital mortality for weekend admissions was found in this study. There has been a substantial reduction in mortality for both weekday and weekend admissions over time.

High Risk Subgroups Sensitive to Air Pollution Levels Following an Emergency Medical Admission

For three cohorts (the elderly, socially deprived, and those with chronic disabling disease), the relationship between the concentrations of particulate matter (PM10), sulphur dioxide (SO2), or oxides of nitrogen (NOx) at the time of hospital admission and outcomes (30-day in-hospital mortality) were investigated All emergency admissions (90,423 episodes, recorded in 48,035 patients) between 2002 and 2015 were examined. PM10, SO2, and NOx daily levels from the hospital catchment area were correlated with the outcomes for the older admission cohort (>70 years), those of lower socio-economic status (SES), and with more disabling disease. Adjusted for acuity and complexity, the level of each pollutant on the day of admission independently predicted the 30-day mortality: for PM10–OR 1.11 (95% CI: 1.08, 1.15), SO2–1.20 (95% CI: 1.16, 1.24), and NOx–1.09 (1.06–1.13). For the older admission cohort (≥70 years), as admission day pollution increased (NOx quintiles) the 30-day mortality was higher in the elderly (14.2% vs. 11.3%: p < 0.001). Persons with a lower SES were at increased risk. Persons with more disabling disease also had worse outcomes on days with higher admission particulate matter (PM10 quintiles). Levels of pollutants on the day of admission of emergency medical admissions predicted 30-day hospital mortality.

Social Factors Determine the Emergency Medical Admission Workload

We related social factors with the annual rate of emergency medical admissions using census small area statistics. All emergency medical admissions (70,543 episodes in 33,343 patients) within the catchment area of St. James’s Hospital, Dublin, were examined between 2002 and 2016. Deprivation Index, Single-Parent status, Educational level and Unemployment rates were regressed against admission rates. High deprivation areas had an approximately fourfold (Incidence Rate Ratio (IRR) 4.0 (3.96, 4.12)) increase in annual admission rate incidence/1000 population from Quintile 1(Q1), from 9.2/1000 (95% Confidence Interval (CI): 9.0, 9.4) to Q5 37.3 (37.0, 37.5)). Single-Parent families comprised 40.6% of households (95% CI: 32.4, 49.7); small areas with more Single Parents had a higher admission rate-IRR (Q1 vs. for Q5) of 2.92 (95% CI: 2.83, 3.01). The admission incidence rate was higher for Single-Parent status (IRR 1.50 (95% CI: 1.46, 1.52)) where the educational completion level was limited to primary level (Incidence Rate Ratio 1.45 (95% CI: 1.43, 1.47)). Small areas with higher educational quintiles predicted lower Admission Rates (IRR 0.85 (95% CI: 0.84, 0.86)). Social factors strongly predict the annual incidence rate of emergency medical admissions.

The dynamics of the emergency medical readmission — The underlying fundamentals

BACKGROUNDnHospital readmissions are a perennial problem. We reviewed readmissions to one institution (2002-2015) and investigated their dynamics.nnnMETHODSn96,474 emergency admissions (in 50,701 patients) to an Irish hospital over a 15-year period were studied, and patterns surrounding early (<28days) and late (any other) readmissions determined. Univariate and logistic or truncated Poisson regression methods were employed.nnnRESULTSnEarly readmission rate averaged 9.6% (95% CI: 9.4, 9.8) with a low/high of 8.4% (95% CI: 7.8, 9.1) and 10.3% (95% CI: 9.6, 11.0) respectively with no overall time trend. Early readmissions represented 20.1% (95% CI: 19.8, 20.5) of emergency medical readmissions. Median time to first readmission was 55weeks (95% CI: 13, 159), time to second was 35weeks (95% CI: 9, 98); by the 7th/8th readmissions, intervals were 13weeks (95% CI: 4, 36) and 11weeks (95% CI: 4, 30). Readmissions were older 67.1years (95% CI: 48.3, 79.2) vs. single admissions 53.9years (34.3, 72.4) and stayed longer - 5.8days (2.7, 10.6) vs. 3.9days (1.5, 8.0). Readmissions had more Acute Illness Severity, Charlson Co-Morbidity and Chronic Disabling Disease. Between 2002 and 2015 the logistic adjusted model of 30-day in-hospital mortality reduced from 6.1% (95% CI: 5.7, 6.5) to 4.4% (95% CI: 4.1, 4.7) (RRR 30.4%).nnnCONCLUSIONnEarly hospital readmission rate did not change over 15years despite improvements in hospital mortality outcomes. Readmissions have a consistent pattern related to patient illness and social characteristics; the fundamentals are driven by disease progression over time.