A new Department of Health report has identified major variations in patient death rates between Irish hospitals for some common conditions such as heart attack and stroke.

However, while the Department has admitted that some of the variations may be due to quality of care and safety issues, it says data limitations and other technical factors not taken into account may call into question the reliability of the figures, and further analysis will be required.

While hospitals are not identified in this study, the Department of Health says further reviews of mortality rates will name individual hospitals. Experts have stressed that while the new death rate comparison was an important indicator of healthcare quality, a higher than average death rate recorded did not always mean inferior care beteween one hospital and another.

The Department stresses that the findings of the report 'should not be taken as making any inferences concerning quality of care in hospitals and certainly should not be interpreted as ranking hospitals with respect to the selected indicators'.

It says, however, that the collation of mortality figures in the report demonstrates the value of calculating quality indicators and has identified areas requiring further exploration in relation to clinical care and data collection.

Initial analysis in the report, entitled 'Health Care Quality Indicators in the Irish Health System', found that a small number of hospitals reported 30-day mortality rates higher than would be expected when compared to national rates, the report says.

The mortality comparisons between hospitals take into account the significant factor of age of patient admissions. The data collected was for the years 2008-2010 in 36 hospitals.

The statistics for age-standardised mortality include:

* The death rate for ischaemic stroke patients was nearly four times greater in the worst-performing hospital than in the best performing hospital (209.2 vs 61.0). Ischaemic strokes are strokes caused by blood clots.

* The death rate for haemorrhagic strokes was nearly three times greater in the worst-performing hospital than in the best performing centre (148.1 vs 53.5). Haemorrhagic strokes are strokes caused by bleeds.

* For heart attack patients, the mortality rate was nearly 10 times what it is in the worst performing centre versus the best (214.1 vs 26.4).

* For patients having hip fracture surgery, the difference in the age-standardised death rate was nearly threefold between the worst and best performing hospitals. (149.4 vs 57.0)

The centre with the highest mortality rate for heart attack patients is believed to be Roscommon General Hospital, in which emergency and many acute services have ceased since the figures were collated. The same hospital also had the highest age-standardised death rate for ischaemic stroke, according to the figures.

The heart attack death rate figures for Roscommon were previously quoted in public by Health Minister James Reilly, but their accuracy has been disputed.

The report said after the HSE communicated with hospitals with morality rates higher than accepted levels , they identified a number of problems with how they recorded mortality data, including incorrect recording of the principal diagnosis and issues with data coding.

The report says this highlighted issues in relation to data quality on hospital mortality.

It says the HSE's Quality and Safety Patient Directorate and the individual hospitals were communicated with about the mortality variations.

The report says a number of hospitals, on reviewing their data, found inconsistencies in the recording in medical records and/or in coding.

The report points out that there are limitations in the figures produced in terms of comparing the death rates among hospitals for heart attacks, ischaemic and haemorrhagic strokes, as the figures do not link the deaths directly to the principal diagnosis or procedure being undertaken at the hospital.

Also, the comparisons in this report do not take into account patient factors such as other illnesses the patient may have had in addition to the principal condition, or medication use which might have had an effect on the outcome.

Mortality rates included in the report also do not take into account factors occurring before or after treatment in hospital, including how easy it was for the patient to access the hospital, The figures also do not include deaths that occur within 30 days after discharge from hospital.

Other factors not taken into account in the report included social deprivation among patients admitted to hospitals.

The report says this type of confounding factor which may inflate mortality figures need to be taken into account in further reviews.

The Department has admitted that some of the data used in this report to compare death rates was unreliable due to lack of consistency in the documentation in hospital records.

Hospitals have also been ordered to improve the quality of their data on mortality rates, after issues with data quality were discovered as part of the review.

The Department says indicators as revealed in its report can be used as alerts or flags to identify areas of performance 'that may require further exploration', and they can also identify good practice that can be shared throughout the hospital system.

View the full report here