Several hundred lives would be saved each year if some of New Zealand's highest public hospital death rates were brought down to those of the better-performing district health boards.
A Herald investigation has found that the Waikato District Health Board has had among the highest in-hospital "standardised" mortality rates in recent years.
In a five-year period, it had the highest rate for three years.
In those years, its standardised rate - an estimate that gives a fairer comparison than raw data - was significantly higher than the national average.
"Mortality as well as safety, we take it very seriously," said the DHB's chief operating officer, Jan Adams.
The Health Ministry has identified the Waikato rate as a problem, but has also commended the health board for the efforts it is making to get to the root of the issue.
Figures comparing hospital death rates are routinely available to the public in Britain and Canada, and Australia has endorsed using this kind of information as an indicator of healthcare quality and safety.
In New Zealand, the details have not been readily available, and the five-year figures for the Herald investigation were obtained from the ministry under the Official Information Act.
They show considerable variation between hospitals, but also an overall downward trend.
In the 2010/11 year, just under 8000 admitted inpatients and day patients - 1.48 per cent of total discharges - died within 30 days of admission to hospital.
This was down from 1.68 per cent in 2006/07.
Most hospital deaths cannot be avoided, but overseas research suggests several hundred a year in New Zealand may be preventable.
The ministry said comparing hospital mortality rates with other countries was complicated by different calculation methods, but Australian states tended to have a similar method and rate to New Zealand's.
The Auckland and Waitemata health boards generally had rates at the lower end nationally in the five years of data, while Counties Manukau was mostly in the mid-range.
Rotorua's Lakes DHB was also generally in the middle, while Northland ranged from mid-range to among the lowest. Bay of Plenty tended towards the upper end.
Further south, Capital & Coast was mostly among the lower-rate boards, but Canterbury usually had one of the higher rates.
Experts debate whether standardised hospital mortality rates can be used to evaluate and compare the quality and safety of healthcare in different hospitals.
But in its annual report last year, the ministry made an explicit link between hospital mortality and quality of healthcare.
After an epidemic of healthcare-induced harm was identified in New Zealand a decade ago, numerous schemes to improve the safety and quality of care have slowly taken root.
In a week-long series starting today, the Herald looks at some of these life-saving schemes.
Some, such as the surgical check-list, are simple, and some, such as changes to prevent common errors in the prescribing and dispensing of potentially toxic medicines, are more complex.
The Herald also talks to families who have lost loved ones to hospital errors that might have been avoidable.
They are bewildered by the paperwork mix-ups and overlooking of clinical information that have contributed to their personal tragedies and feel they have been denied justice.