Long Term Care System: Is There Proximity between Supply and Demand in France?

Amélie Carrère, Université Paris Dauphine

Context: In France, during the 1960’s, people have moved from rural to urban areas leading to the development of urban centres that concentrate economic activity. Some people still live in areas faraway from these centres: 5% in 2010. In the same year, elderly people lived more in these isolated areas (8%). They therefore are at higher risks of living in cities providing little economic activity and probably few formal care or medical facilities useful to stay longer at home. Nevertheless, areas with higher prevalence of disability are also areas where the supply for long term care is higher. Is the supply for home care services related to the need of assistance and to the entrance in a specialized institution? If so, what are the reasons?

Method: I used CARE survey (n=15000 individuals aged 60 and over living either in the community or in specialised institution). I classified the supply into ten categories: home care services, nursing-homes, “residences-autonomies” (dwellings for old people close to nursing-home but providing no medical assistance), temporary dwellings, general practitioners, nurses, dentists, physiotherapists, ophthalmologists and pharmacists. For the first four categories, I calculated distances between the person’s place of residence (or previous place of residence if living in nursing home) and the closest facility. For the other facilities, I used the Local Potential Accessibility (LPA) which is a local indicator of accessibility to health care at a city level. I conducted multi-level logistic regressions to identify the role of healthcare provision on the risk of having disability and of living in a specialised institution.

First results: Some counties have specialised their supply in nurses and their risk of disability is larger, in the community population. Others have specialised their supply in nursing-home and their risk of disability is lower in the community population.


CONTEXT

Most of Europeancountries and OECD counties are facing the ageing of their population. Theshare of the population above 65 was 9.6% in 1960 and 18.4% in 2014 in Europeanunion (28 countries). It is predicted to grow to 28.4% by 2060 in Europeanunion (EuropeanCommission2014) and in France (Blanpain and Buisson2016). Elderly people areat higher risks of having chronic conditions and having disabilities (Ankri and Mormiche 2002; Orfila, Ferrer et al.2003; Von Strauss, Aguero Torres et al. 2003; Dos Santos andMakdessi 2010; OECD 2015). Needs of peoplewith disabilities are large and multiple: medical and nursing care, domestichelp and personal care, assistive devices and housing adaptations. As thenumber of elderly is increasing, the number of elderly people needingassistance is expected to grow.

Simultaneous,urban centres that concentrate economic activity have expanded. But there werestill 5% of the French population living in an urban centre or its periphery in2010. And elderly people live more in these areas (8%). They therefore are athigher risks of living in cities providing little economic activity andprobably few formal care or medical facilities useful to stay longer at home.Nevertheless, (Coldefy, Com-Ruelle etal. 2011) showed that notliving in a urban centre does not mean being away from medical providers.

Thispresentation addresses the following question: is the supply for home careservices related to the need of assistance and to the entrance in a specialisedinstitution?

Articles aboutdeterminants of disability are numerous but little is known about the role ofthe proximity of domestic help and personal care. Literature of healtheconomics focuses mostly on the role of primary care on the emergency care visits(Lang, Davido et al.1997; Wolinski, Liu et al. 2008; He, Hou et al. 2011; Feng, Coots et al. 2013; Ono, Schoenstein et al. 2014;Or and Penneau 2017).

METHOD

First of all, Icompared the need and the supply analysed at a county level. Severaldefinitions of need of assistance are computed using “Vie Quotidienne et Santé”(VQS) survey (n=166800 individuals older than 60 years old living in thecommunity): physical, cognitive and sensory functional limitations, bathingrestrictions and unmet needs for bathing (Desai, Lentzner et al.2001). I have selecteddifferent supply indicators at a county level from multiple sources: number ofplaces in personal care for 100 persons (Finess administrative data); number ofnurses for 100 persons (Adeli administrative data); number of places in nursinghome for 100 persons (EHPA survey), average price of professional services(CARE survey), average price of nurses (Sniiram administrative date) and medianprice for a room in a nursing home (CNSA survey).

Second of all, Ianalysed determinants of need for assistance using VQS data base and amulti-level logistic regression to identify the role of factors associated withindividuals demand and with organisation of health care provision. The firstlevel of this regression is individual and the second is a county level. Takinginto account county dimension is essential because counties organize geriatricschemes of their territory (Billaud,Bourreau-Bubois et al. 2013)(Ramos-Gorand 2016). I used the supply indicators quotedpreviously.

Finally, I useda French Survey named CARE “Capacités, Aides et REssources des seniors” (n=15000individuals aged 60 and over living either in the community or in nursing-home).I classified the supply into ten categories: home care services, nursing-homes,“residences-autonomies” (dwellingsfor old people close to nursing-home but providing no medical assistance),temporary dwellings, general practitioners, nurses, dentists, physiotherapists,ophthalmologists and pharmacists. For the first four categories, I calculateddistances (in kilometres) and duration (in minutes) between the person’s placeof residence (or previous place of residence if living in nursing home) and theclosest facility. These facilities are geolocated using FINESS database. Forthe other facilities, I used the Local Potential Accessibility (LPA) which is alocal indicator of accessibility to health care at a city level, inspired byacademic literature (Radke and Mu 2000; Luo and Wang 2003; Luo and Qi 2009) andclose to density indicator (Mizrahi and Mizrahi2011).I conducted multi-levellogistic regressions to identify the role of healthcare provision on the riskof having disability and of living in a nursing home.

RESULTS

Prevalence ofbathing restrictions by county is drawn in the first map of Image 1. Some counties have specialised theirsupply: some have high level of restrictions, numerous nurses and few nursinghome places and other have low level of restrictions, few nurses and numerousnursing home places.

With CARE survey(Table 1), the risk of havingdisabilities is higher in areas where accessibility to nurses is high. On thecontrary, the risk of having disabilities is lower in areas where the price ofnurses is high.

Presented in Session 1233: Posters