THE number of patients dying at two west of Scotland hospitals has triggered a red flag warning for the second time in a year after it emerged that mortality was "significantly higher" than the national average.

Experts at Healthcare Improvement Scotland, the watchdog for hospital standards, are consulting with health chiefs at NHS Greater Glasgow and Clyde over an unexpectedly high number of deaths earlier this year at the Vale of Leven hospital in West Dunbartonshire and the Royal Alexandra Hospital in Paisley, Renfrewshire.

Between January and March this year, 388 patients died within 30 days of admission to the hospitals, which are counted together under the ISD Scotland statistics. This compared to a predicted number of deaths of 345.

Analysis: Mortality scoring system was first to flag up Mid-Staffs scandal

The data emerged in the latest report on Hospital Standardised Mortality Ratios, a warning system used to flag up potential deficiencies in patient care. The system uses an algorithm to calculate the risk of dying across various patient subgroups based on factors such as age, sex, diagnosis, type of admission, number and severity of illnesses. This risk score is then applied to the particular case mix of patients admitted into different hospitals during a three-month period to predict how many deaths should have occurred.

This figure is then compared with the actual number of deaths reported to create a "mortality ratio". The combined ratio for Vale of Leven/Royal Alexandra was 1.13 - meaning mortality was 13 per cent higher than expected. In Scotland as a whole, average mortality during the period was 0.93 - seven per cent less than expected.

The Greater Glasgow sites were the only hospitals in Scotland to score a "red" rating, indicating that mortality was "significantly higher than the national average" - classed by ISD Scotland as above the upper control limit, which is three standard deviations from the Scottish average.

It is the second time in a year that they have been red-flagged. In January to March 2016, their mortality ratio was 19 per cent above the Scottish average.

A spokesman for Healthcare Improvement Scotland said: “Hospital mortality ratios are not in themselves a measure of quality of care. They can, however, be a valuable tool in drawing attention to potential areas of further review at local level. If a hospital is identified as having either a mortality ratio that is significantly high or that is rising, then we initiate a formal dialogue about this with the NHS board in question. This is to offer the NHS board advice and support in using such hospital mortality data, and also to seek assurance that the NHS board is taking appropriate steps to understand and respond to these figures."

He added that HIS was engaging was in contact with the health board about the data.

A spokeswoman for NHS Greater Glasgow and Clyde: "As a Board we are confident in the standards of care in the Royal Alexandra and Vale of Leven hospitals and in the improvement work which has been carried out by the clinical teams.

"However, as HSMR figures there remain higher than the national average, we are continuing to work with the clinical and managerial teams, supported by ISD and Healthcare Improvement Scotland, to identify factors that might be having a bearing on the HSMR figures.

"We will continue to work with Healthcare Improvement Scotland and will meet with them in the near future to discuss progress.

There are very active quality improvement programmes at these hospitals and a well developed set of arrangements which ensures ongoing monitoring and response to any quality issues which may emerge."

Meanwhile, Caithness General in the Highlands was given an amber rating after 39 patients died there during the first quarter of 2017, compared to 28 predicted.

Dr Rod Harvey, medical director for NHS Highland said: “We are aware of the increase in the standardised mortality ratio for the CGH for the most recent quarter and are in the process of reviewing all of the cases in more detail to better understand the cause. However, I can confirm that I have no concern over the standard of clinical care being provided at the hospital and am reassured to note that the total number of deaths for this quarter is exactly in line with the same quarter (Jan to March) for the previous two years.