The dataset source, Public Health England, is an executive agency of the Department of Health & Social Care, part of England Government.
Liver disease is one of the top causes of death in England and people are dying from it at younger ages. Most liver disease is preventable and much is influenced by alcohol consumption and obesity prevalence, which are both amenable to public health interventions.
The hospital admissions on which are based the estimated rates include those for liver diseases recorded with ICD-10 codes B15-B19, C22, I81, I85, K70-K77, T86.4. Regular attendees have been excluded. Admissions have been included and counted according to the financial year in which the admission episode ended, and are presented by Local Authority of residence, aggregated into quinary age groups (0-4, 5-9,…, 80-84, 85+). A patient may be admitted more than once and admissions of patients with no fixed abode are have been included in the England totals.
The numerator data, based on the Hospital Episode Statistics from NHS Digital, for each quinary age group, have been divided by the denominator population data for each age group (based on the 2011 Census population from the Office for National Statistics) respectively to give age-specific admission rates for the area. These age-specific rates have been multiplied by the standard population for each age group respectively and aggregated across all the age groups to give the age-adjusted count of admissions for the area. This age-adjusted count of admissions has been divided by the total standard population for the whole age range included in the indicator, and multiplied by 100,000 to give the age standardized admission rate for the area. The standard population used to calculate age-specific standardized hospitalization rate is the 2013 European standard population. Local Authorities with between 1 and 5 admissions have been suppressed.
The confidence intervals are calculated using the Dobson & Byar’s method. Dobson & Byar’s method is used to calculate confidence intervals for directly standardized rates. A confidence interval is calculated for the observed total count of events using Byar’s method, which gives very accurate approximate confidence intervals for counts, based on the assumption of a Poisson distribution and is sufficiently accurate for counts as low as 5 (below 5, an exact method should be used, based on Poisson tables or the Chi-squared distribution). This interval is then weighted and scaled to give the interval for the standardized rate using the method described by Dobson.