Homeless die 20 years early - study

Some homeless families are being forced to live in their vehicles. Photo / Doug Sherring

Homeless people admitted to South Auckland's Middlemore Hospital die on average 16 to 20 years earlier than other New Zealanders, a new study has found.

The study followed 1182 Middlemore patients aged 15 to 74 who were tagged as having inadequate housing or poverty between January 2002 and December 2014, and found that the tagged patients died, on average, by age 63.5.

For comparison, New Zealand men can expect to live on average to age 79.5, and women to 83.2.

Study author Dr Simon Thornley, a Middlemore public health physician, said the tagged group was a tiny 0.3 per cent of all South Aucklanders aged 15 to 74, and all were sick enough to be admitted to hospital. But the gap in life expectancy was still much wider than he expected.

"Yes, there are major issues of mental health and addiction and metabolic disease, but even I was surprised by how low their life expectancy is compared with the general population. It's a select group, but 20 years younger, that's quite dramatic," he said.

"There's no doubt that those addiction issues are more common in this group than in the other group, and we know that addiction tends to make you less good at managing other areas of your life such as your finances and nutrition.

"But the stress of not having a home, and not being able to organise your life in the way that others can, is likely to exacerbate all those conditions."

He said the study, published today in the NZ Medical Journal, added to other evidence that investing in housing reduced future health costs.

"They tend to do better when they are provided with a house so that social services have a place to go to and care can be coordinated for them," he said.

Patients in the group were tagged as homeless or with inadequate housing, discord with neighbours, lodgers and landlords, problems related to living in residential institutions, lack of adequate food, low income, insufficient social insurance and welfare support, or other problems related to housing and economic circumstances.

Although it was a broad list, Thornley said the "vast majority" of tags were about poor housing, because problems were only tagged if they were "exacerbating conditions" for the patients' main health issues.

About two-fifths of the group were hospitalised for mental health issues and the rest for other medical conditions.

Forty per cent were European, 33 per cent Maori, 20 per cent Pacific and 7 per cent Asian. The Maori share of the homeless was twice the Maori share of the general population and they were also 68 per cent more likely to die during the study period than other tagged patients.

Almost half (47 per cent) of the tagged sample smoked cigarettes and that was by far the biggest single factor driving their higher death rate. The next-biggest drivers were diseases related to smoking cannabis, diabetes and alcohol-related illnesses.

Otago University housing and health researcher Dr Michael Baker noted that the study did not directly compare the tagged group with non-tagged patients admitted to Middlemore with the same illnesses.

"The analysis does show some limitations of what you can do with administrative data," he said. "A lot more effort needs to be put into looking at the health of this group."