Where do people die: what proportion use residential care?

Where do people die?

What proportion ever uses residential aged care?Joanna Broad, Michal Boyd, Martin Connollyin Freemasons Department of Geriatric Medicine,University of AucklandConference for General Practice, Auckland, September 20111People want to die at home Many reports, eg:UK Higginson 2000, Koffman 2004 USA Hays 2001 Korea Choi 2005 Japan Fukui 2011 Israel Iecovich 2009

In Australia, a survey showed 58% preferred to die at home Foreman 2006

If place of death was home, preference was not often met Agar 2008

2Well recognised that people in most societies express preference to die at home. Ageing in place strategies adopted by many governments.Social and demographic changes Traditional options for late life care less common: less informal care (family or neighbours) available esp. more women aged 45-65, in workforcemore older people living alonechanges in families, eg fewer children, live a distance awayalmost no live-in housekeeper or boarding arrangements? changing responsibility for/to parents Growth in retirement villages3In NZ & elsewhere, traditional models of care have declined, & life expectancy increased, making it harder to die at home.

More people now living in retirement villages than in LTC.

What proportion live in residential LTC?Of population, at any one time, 5-6% live in residential long term care:65+85+in Auckland, 2008 (OPAL)5.5%27.5%in NZ, 2006 (Census)5.5%25.4%in Australia, 2008 (AIHW)5.3%23.5%

OECD (2006) data for 65+s showed 8.2% NZ Japan3.0 Norway 5.6UK 4.2 Switzerland 6.5Australia 5.4Sweden6.8Denmark5.4 Netherlands 6.94Variations in reportsThese are all cross-sectional studies, ie at any one time. we do not know ever use marketing statement eg breast cancer that one in 10 women will be diagnosed with breast cancer at some time in their lives

Wish to recognise the widespread use of LTC, and therefore the topics importance, eg to inform.discussions about options to LTCany changes in admitting practicesthe impact of any funding changespersonal choices eg about retirement fundsto understand the sector better

Population rates in LTC in Auckland 2008 (OPAL)

5.5% of all 65+5Rates approx. double with each 5-year increase in ageIn over-85s, rates in care are particularly high Rates in residential LTC higher for women than for men

How will ageing impact.?

Source: MSD 201180+65-6975-7970-74In mid-2020s, numbers of 80+ will start to rise, & not stop ...

Should we anticipate a huge increase in demand for LTC? 6Where do people die?Classification of place of death using NZ Death Registrations7

8Top part of cause of death form - no boxes to tick for hospital or long term care facility. Has no information on usual residential address. Instructions for place of death box are please ensure that the name of the hospital or institution is entered in this field, not just the street address. Information is transferred onto Notification of Death form.At data entry, recognisable public hospitals are coded with facility code. Used for analysis & reporting.

Instructions say If the deceased died in a hospital or other institution, please ensure that the name of the hospital or institution is entered in this field, not just the street address.

Information goes to Dept. of Internal Affairs, & passed to Statistics NZ, MoH, MSD, transport, passports Usual address is not analysed or reported.9Births, Deaths, Marriages and Relationships Registration (Prescribed Information) regulations 1995 require place of death & usual residential address to be recorded

Two forms required for deaths in NZ:notification of death for registration medical certificate of causes of death (or order from coroner)Notification of Death for Registration usually completed by funeral director paper form or online. Info also goes to MoH, MSD, drivers licence office etc

Usual home address is not summarisedNZ Ministry of Health algorithm Developed an algorithm to classify place of death:hospital deaths identified from a facility code given during data entry => hospitalsearch for words e.g. rest, home, centre, lodge, hospital etc and not hospice => residential careif residence, unit, flat => private residenceif not otherwise classified e.g. traffic or work => private residenceFor policy development & advice to Minister10?for palliative care interestsNot published, best method to datehome very small proportion misclassifiedNZ MoH place of death for 65+s4 year 2003-2007, average %

Home & other includes private home, work place, road & traffic, public place & unknownRAC includes rest homes and all private hospitalsPublic acute care hospitals11For people aged over 65

Place of death of 65+s using published, online and requested data12Questions how well are families and patients supported during dying & close to death? - how well are they supported in expectations & understandings? - how well are physical & medical cares met? - how well are spiritual and emotional needs met?

Where do people live shortly before they die?Estimation for NZ Estimation for Australia

13What measures of ever use? Australiastudy in 2002 modeled annual probability of entering careest. 64% of 65+ women & 39% of 65+ men would use residential aged care before death Rowland 2002New Zealandno similar study undertaken in NZno national reporting of LTC admissions or discharges?use place of death or place of usual residence14Place of death not published in NZRoutine reports of admission and discharge from facilities is not required in NZWe set off to find what information is available.Estimate for AustraliaUsed annual summaries of discharged dead from LTC & acute hospital (AIHW)32% in LTC, 54% in hospitalOf all in-hospital deaths, 13% were LTC residents on leaveEst. total of ever use: 32% + 0.13*54 = 39% of 65+sProbably an underestimate e.g. if resident was discharged from LTC to hospital

15Back to first, underlying question what proportion ever live in residential aged care facilities? Australia first 32% died in care.Some die in acute hospital care having been admitted from rescare facility. 48% is a hugely different proportion than 5 or 6% at any one time.But is based on hospitalisations of Waitemata DHB residents, higher than overseas studies ?can get more data?

Estimate # 1 for New ZealandRACIP trial in WDHB tracked hospitalisations in those from residential LTC aged 65+ Boyd, in progressOf all in-hospital deaths for 65+s, 30% were from LTCEst. total living in residential LTC at the time of death: 38% + 0.30*34% = 48% of 65+s16Back to question what proportion ever live in residential aged care facilities? 38% died in care.Some die in acute hospital care having been admitted from rescare facility. Used RACIP data to estimate that proportion.48% is a hugely different proportion than 5 or 6% at any one time.But is based on hospitalisations of Waitemata DHB residents, higher than overseas studies ?can get more data?

Estimate # 2 for New ZealandFollow-up of OPAL cohort in Auckland hospitalisations & deaths Weighted to represent a 12-month RAC cohortOf all deaths Sep2008-09, est. 9% occurred in acute hospitalEst. total from residential LTC at the time of death: 38% * 1.11% = 42% of 65+s1740 - 50% in LTC in NZ? inadequate assessment before admission to LTC Weatherall 2004, Grant Thornton 2010LTC model prevails, & options not understood by older person or their families Jorgensen 2009home-based services poorly co-ordinated, inadequate, ineffective or not timely Weatherall 2004, Grant Thornton 2010? hospice-care is less available for older people, facilities provide de facto palliative care? reducing lengths of stay in AT&R units => failure to rehabilitate, false economy? LTC preference over episodes of acute care 18Some evidence, some opinionNo doubt othersBack to the future deaths will riseAfter 20-year stable period, number of deaths is now rising, though death rates continu to fall

Statistics NZ000sProjected rise in demand for beds of 78-110% by 2026 - ? unsustainable Grant Thornton 2010

2011198719Because long-term each stay is usually linked to a death, one indicator of demand for LTC is projected deaths. This graph from Statistics NZ.after a 25-year period of relative stability (yellow) about 26-27,000 numbers of deaths are rising (red)

Grant Thornton report in 2010 determined that the rise demand for LTC will be unsustainable - projected rise of 78%-110% in demand for beds by 2026, ie almost double in 15 years

In other countries & probably NZ, LTC use is related more to deaths than to age, And health of older people is improving, along with survival.

Where may that lead? Summary in NZ, high rates of use of LTC ~48% of 65+s die while resident in LTC arguably more compelling (vs 5-6%), more attention-demanding, with greater potential for advocacy or policyhelpful to explore other models of non-acute care, especially for end-of-lifeimpacts of changes to funding of LTCinform care provisionpersonal planning for retirement admitting practices to LTC2048% is a hugely different proportion than 5 or 6% at any one timeAcknowledgementsAnn Peut, AIHW, CanberraHe Chen, BeijingHongsoo Kim, SeoulMerryn Gott, Auckland

Be nice to your kids Theyll choose your nursing home Abraham Maslow

Deaths in residential care, doubling each 10 years of age23More questionsVery consistent - doubling with each additional 10 years of age

NZ at top in almost all age groups Implications how well do we support or train staff in residential aged care facilities to help the dying & their families? In future, population rates in LTC may move to older ages

24Improved health of older people

Perhaps rates will move up say 5 years ?

NZ Population pyramids

50 years on 80 years on1951 25Growing proportion are reaching very old ageAgeing of population means more older people, and more oldest oldDemographers tell us there will be more people living alone esp. widowed or never marriedCompared to others, those living alone are more likely to move to LTC

OutlineWhere do people die?Reporting deaths in NZPlace of death in NZ International comparisons

Where do people live before they die?Estimates for NZ & AustraliaImplications & work in progress

26Talk a little about residential long term care & importance to consider future at this time. What is known about residential LTC utilisation in NZ . Describe reporting of deaths. Describe place of death in NZ, and what it tells of utilisation. Way ahead work in progress.Invite audience to walk a trail together

High proportion of deaths in LTC? or Low proportion in hospital?If estimates correct, compared to other countries, NZhas high proportion of deaths in private home & LTC& low proportion in hospitalMaybe LTC is in some sense the best option for some to age in place ....not necessarily the family home or living alonecould be a facility of their choosing, familiar surroundingsResidence in LTC may reduce deaths in hospital preventive? protective?or a question of access?27If accepting of these data, then can ask why are hospital deaths in NZ so lowQuestion of access to acute or secondary hospital care if living in LTC?

Reporting preference to die at home Preference for home death may not reflect care when disabled or very old Freid 1999Difficult to survey preferences in some cultures Choi 2005 Poor-moderate congruence between preferred place of death and actual place of death Bell 2009Post-bereavement interviews - care-givers think actual place of death appropriate Brazil 2005Important for care-givers and physicians to know preferred place of death Cohen 2011

28Or you can ask why is LTC in NZ so high with the advent and proliferation of institutional care, traditional models of care have declined. Perhaps this trend needs to be rebutted somewhat, and social policies developed that support that role more? What are options? We hear about initiatives from overseas. Market & environment considerationsLittle provision for: short-term live-in rehabilitation after acute episode & perhaps => hasty decision to movecommunity care eg adult day-care or seniors centressheltered housing eg from local bodies, & few small apartments & flats (though LTC is more than a housing option) LTC bed provision high

29Or you can ask why is LTC IN NZ so highPossible factors might includeIn comparison to other countries, little provision for Little data, esp for trends over timeExcerpt from A guide for doctors and coroners on the provision of information on deaths to the New Zealand Health Information Service (New Zealand Health Information Service, Wellington, 2001)Completing the certificate.Place of deathIf the deceased died in a hospital or other institution, please ensure that the name of the hospital or institution is entered in this field, not just the street address.Place of death of 65+s using published data

31No data for Australia broken down by these age groups. Several questionsAre the data believable?If true, what are messages? DISCUSS Some differences in definitions in UK, hospital data include a few NHS care homes ? NZ very high in rescare, but low in hospital deaths. Maybe high rescare is not such a bad thing?

Australia - impact of deaths in hospital from LTC

MenWomenAT&R utilisation for 85+s

1.3% 7.4% 3.9% 35Not only are there few facilities for rehab, but utilisation of public AT&R services is not optimal. This is chart of 85+s compares utilisation of AT&R in Northern Region, with population increase and to medical/surgical increase.