Background

With widespread population ageing, many Western countries can expect a substantial growth in demand for labour-intensive long-term care (LTC) services (OECD 2021). At the same time, older segments of the population are growing increasingly diverse as immigrants ‘age in place’ (Rechel et al. Reference Rechel, Mladovsky, Ingleby, Mackenbach and McKee2013). Despite immigrants’ access to health care being a crucial element of integration and a fundamental human right, there is limited knowledge on the utilization of LTC services among immigrants (Bolster-Foucault et al. Reference Bolster-Foucault, Vedel, Busa, Hacker, Sourial and Quesnel-Vallée2024; Nielsen et al. Reference Nielsen, Krasnik and Rosano2009).

While the share of the foreign-born population in Norway is higher than the Organisation for Economic Co-operation and Development average (OECD 2022), large-scale immigration to Norway is a relatively recent phenomenon and the number of immigrants falling into the older age categories remains low. However, according to the official population projections, the number of immigrants over the age of 80 will increase fourfold within 20 years and more than tenfold by 2060 (Thomas and Tømmerås Reference Thomas and Tømmerås2024). This represents an increase from around 13,000 today, to over 145,000 by 2060 (Thomas and Tømmerås Reference Thomas and Tømmerås2024).

Generally speaking, immigrants in Western countries, including Norway, have lower average risks of mortality than their native-born peers (Wallace et al. Reference Wallace, Thomas, Aburto, Pallesen, Mortensen, Syse and Drefahl2022). This general health advantage has been linked to the so-called healthy immigrant effect, where immigrants are thought to be positively selected on health (and health-conducive socio-economic characteristics) relative to non-migrants in the sending country, while requirements linked to immigrant admission policies at the destination may add a further layer of positive selection (Elshahat et al. Reference Elshahat, Moffat and Newbold2022). Negative selection into return migration might also contribute to apparent health advantages among immigrants relative to the native-born population, particularly if less healthy and less economically advantaged immigrants are those that are more likely to return to their country of origin (a process described as the ‘salmon bias’) (Abraído-Lanza et al. Reference Abraído-Lanza, Dohrenwend, Ng-Mak and Turner1999). However, the existence of a salmon bias effect has been challenged by recent empirical results from Sweden, with lower risks of emigration observed among immigrants with co-morbidities (low, moderate and high) relative to those without co-morbidities (Dunlavy et al. Reference Dunlavy, Cederström, Katikireddi, Rostila and Juárez2022).

Of particular relevance to a focus on older immigrants, who typically have long durations of residence, is evidence that immigrant health advantages decline with time spent in the destination country. In Norway, for instance, studies ranging from hospital admission rates (Elstad Reference Elstad2016) to mortality risks (Syse et al. Reference Syse, Dzamarija, Kumar and Diaz2018) observe a general trend of convergence towards the native-born population as immigrants’ durations of stay increase. For some immigrant groups, health can be significantly worse than the national average regardless of duration of residence, and especially with long duration of residence (Kjøllesdal et al. Reference Kjøllesdal, Gerwing and Indseth2023). A process of unhealthful acculturation, wherein immigrants adopt risky/unhealthy behaviours and lifestyles more prevalent in the destination context, is one perspective as to why immigrants experience health deterioration over time. For instance, meta-analyses of health risk behaviours point to the importance of specific origin and destination contexts and differences according to the behaviours studied; however, several North American and European studies do at least highlight generally increased risks of smoking among immigrant populations with longer durations of residence (Juárez et al. Reference Juárez, Honkaniemi, Gustafsson, Rostila and Berg2022). An alternative perspective points to (cumulative) individual, social and system-based stressors as having deleterious effects on immigrants’ health over time. Disadvantaged socio-economic status, language difficulties, discrimination and difficulties navigating housing, labour market and health systems have been noted as examples of stressors that can accumulate over time with negative implications for immigrants’ health trajectories (Elshahat et al. Reference Elshahat, Moffat and Newbold2022). It has also been noted that experiences of such stressors may themselves underlie observed increases in health risk behaviours among immigrants over time. For example, stressors may engender the uptake of risky/unhealthy behaviours, while limited material resources can restrict access to certain types of physical activity and diet that are associated with healthier lifestyles (Juárez et al. Reference Juárez, Honkaniemi, Gustafsson, Rostila and Berg2022). The typical associations between older immigrants’ durations of stay, health and health-care utilization are thus likely to be moderated by a wide range of social determinants (Marmot Reference Marmot2005), including socio-economic status and education, health literacy and practices, access to care, and lived or perceived discrimination (Diaz et al. Reference Diaz, Kumar, Gimeno‐Feliu, Calderón‐Larrañaga, Poblador‐Pou and Prados‐Torres2015; Edberg et al. Reference Edberg, Cleary and Vyas2011).

A further important source of heterogeneity relates to immigrants’ country of origin, which is known to be closely associated with initial reasons for immigration. Notably, over the period 1990–2023, the large majority (79 per cent) of immigrations from combined African and Asian origins were linked to refugee and family reunification reasons, with just 7 per cent of immigrations from Africa and Asia undertaken for labour reasons (Statistics Norway 2024a). Refugees are often exempt from the typical health benefits we associate with immigrants upon arrival, and they may even experience health deteriorations because of circumstances before or during the transfer (Syse et al. Reference Syse, Dzamarija, Kumar and Diaz2018). Non-Western immigrants also have typically lower levels of educational attainment and weaker labour-market attachment and earnings than Western-origin immigrants in Norway (Bratsberg et al. Reference Bratsberg, Raaum and Røed2018). While this may point to relatively poor health outcomes, refugees are invited to a general health assessment upon arrival in Norway, which could promote a higher uptake of formal health and care services and follow-up when such needs exist (Haj-Younes et al. Reference Haj-Younes, Strømme, Igland, Abildsnes, Kumar, Hasha and Diaz2021).

Among older age immigrants, country of origin has also been used to proxy cultural distance between origin and destination contexts and to recognize differences in sensitivities to and experiences of health-care systems, as well as heterogeneities in the role and expectations of family in care-giving (Hestevik et al. Reference Hestevik, Jardim and Hval2022). Independent of any underlying differences in care needs, such factors could bear relevance for disparities in LTC utilization by immigrant background. In a small-scale study of municipal care among older immigrants in Sweden, Hovde et al. (Reference Hovde, Hallberg and Edberg2008a) noted how non-Nordic immigrants found difficulties in access to culturally congruent care and in knowledge and interactions with health-care systems, while health-care personnel appeared to have less knowledge about non-Nordic immigrant patients than their Nordic-born peers. Family members of non-Nordic immigrants were also found to play a more significant role in health-care provision, where they were more often registered as carers (Hovde et al. Reference Hovde, Hallberg and Edberg2008b). A more recent qualitative study of older Pakistani women living in Oslo, Norway, documented a range of issues that were reported to have influenced preferences, experiences and negotiations between formal and informal care, which ranged from perceptions about the quality of formal care and fears of being left behind in residential care homes, to concerns about mixing with other cultures and genders and feelings of disloyalty and shame in relation to being cared for by outsiders (Arora et al. Reference Arora, Straiton, Bergland, Rechel and Debesay2020).

From this perspective, where the presence of family members (e.g. adult children and partners) may indicate access to informal familial support systems, we might expect a higher likelihood that family care substitutes for formal health and social care among older immigrants, and perhaps particularly for those from more culturally distant origins. Likewise, those who do transition into LTC might be a select immigrant subgroup that, for instance, utilizes LTC only when the situation is severe. With that said, an increasing openness to home health-care services among older immigrants with non-European backgrounds has been noted as another form of adaptation, wherein professional care provision might play a supportive rather than a substitutive role (Horn Reference Horn, Torres and Hunter2023).

More generally, previous studies find the presence of partners (and parenthood) to be positively associated with health and longevity (Kravdal et al. Reference Kravdal, Grundy, Lyngstad and Wiik2012; Perelli-Harris et al. Reference Perelli-Harris, Hoherz, Addo, Lappegård, Evans, Sassler and Styrc2018), while unpartnered and/or childless older people have been shown to have higher rates of LTC utilization than those with partners and/or children (Grundy and Jitlal Reference Grundy and Jitlal2007; Syse et al. Reference Syse, Artamonova, Thomas and Veenstra2022; van Der Pers et al. Reference van Der Pers, Kibele and Mulder2015). There is very little research on the role of family in influencing the intensity of LTC, but it has been suggested that close family members could play an important role in helping to support older relatives’ navigation, interaction and follow-up in health-care systems (Smith et al. Reference Smith, Dixon, Trevena, Nutbeam and McCaffery2009; Syse et al. Reference Syse, Artamonova, Thomas and Veenstra2022).

In the Norwegian context, the initial evidence suggests that immigrants are less likely than natives to use LTC services, and if they are awarded services, they tend to receive fewer hours of care (Dzamarija Reference Dzamarija2022). However, these prior findings are descriptive in nature and do not account for socio-demographic differences between immigrants and natives, or potential variations across immigrant subgroups. They are, however, in line with existing research on possible under-use (Hansen Reference Hansen2014), especially among those with short durations of residence (Emami and Ekman Reference Emami and Ekman1998; Qureshi et al. Reference Qureshi, Schumacher, Talarico, Lapenskie, Tanuseputro, Scott and Hsu2021). Against this backdrop, academic and policy reports have considered whether eligible immigrants might under-utilize LTC services owing to potential discrimination in health system evaluations, differing levels of knowledge of immigrant groups among health-care professionals, differences in preferences or a lack of health literacy when navigating complex care systems (Hestevik et al. Reference Hestevik, Jardim and Hval2022; Khatri and Assefa Reference Khatri and Assefa2022).

The Norwegian setting

Norway sits within the Nordic tradition of extensive and universal welfare provision (Esping‐Andersen Reference Esping‐Andersen1999), providing highly subsidized health and care services (either home-based or institutionalized) to all residents regardless of age (Molven and Ferkis Reference Molven and Ferkis2011). The services are predominantly publicly financed through general taxation and rationed according to needs (Molven and Ferkis Reference Molven and Ferkis2011); public sources account for 85 per cent of total health expenditure (Debesay et al. Reference Debesay, Arora, Bergland, Borch, Harsløf, Klepp and Laitala2019). Care services are administered at the municipal level. To mitigate geographical disparities in service provision, statutory regulations require all municipalities to adhere to predefined minimum standards of welfare services, including primary health and care services. With that said, residents are not entitled to specific services, and municipalities must decide on the type and scope of the service that is warranted to meet the corresponding individual needs of residents (Molven and Ferkis Reference Molven and Ferkis2011). Previous variance component analyses have found very little variation in uptake of LTC at the municipality level (Syse et al. Reference Syse, Artamonova, Thomas and Veenstra2021).

Home health care (HHC), which includes home nursing and other forms of health care (e.g. physiotherapy, occupational therapy and (re)habilitation services) received in the patient’s home, is the most commonly awarded LTC service among older individuals in Western countries (OECD 2021) and is the preferred form of care, when appropriate, in Norway (Ministry of Health and Care Services 2011). More than 80 per cent of all LTC users in Norway depend on HHC (Statistics Norway 2023). Within the Norwegian health-care system, HHC is defined as an LTC service and, according to the data analysed in this article, around 90–92 per cent of persons receiving HHC continued to receive it in the following year. An additional attractiveness about the analysis of HHC is that it enables us to assess variations in both transitions into care and intensity of care, with the latter not possible in analyses of institutionalized care. Thus, via a full-population longitudinal study, we examine patterns in HHC uptake and intensity among immigrants and natives in Norway, adjusting for differences in immigrant background and socio-demographic characteristics.

Norway’s highly universalistic service provision, coupled with its defamilialized policy environment (Albertini and Pavolini Reference Albertini and Pavolini2017), should allow individuals access to formal care relatively independently from their own economic or familial resources. Still, recent research indicates fewer transitions to institutionalized LTC among older individuals with resourceful partners and adult children, in contrast to those without partners or children, or those with partners and children with less advantageous characteristics (Syse et al. Reference Syse, Artamonova, Thomas and Veenstra2022). To our knowledge, no prior research has been performed to determine how patterns in HHC uptake and intensities might vary by immigrant background characteristics. Given the limited body of existing research, we pose three broad research questions (RQs):

Data and methods

Discussion and concluding remarks

To our knowledge, no prior research has attempted to determine how HHC utilization (transition risks and intensities) might vary by immigrant background characteristics. Given the limited body of existing research, we posed three broad explorative research questions. RQ1 asked: ‘Does the likelihood of transitioning into HHC differ by immigrant background characteristics […] and, if so, in what ways?’. We found that immigrants were less likely to transition into HHC, but transition risks among immigrants trended towards the propensities of the native-born population as durations of stay in Norway increased. No immigrant group had a higher transition risk than natives, but natives and non-Western immigrants with a short duration of residence had similar risks. A closer look at the latter group revealed that 70 per cent of those who transitioned into HHC were recently arrived refugees. RQ2 asked: ‘For those who are in receipt of HHC, are there differences in the intensity of care by immigrant background characteristics?’. Adjusted model results suggested that variations in HHC intensities by immigrant background characteristics were small and not significant for all groups except non-Western immigrants, where we observed fewer hours of care per day on average than for natives. RQ3 asked: ‘Do the patterns associated with the effects of socio-demographic characteristics (e.g. the presence of family and individual socio-economic resources) on HHC utilization differ according to immigrant background characteristics?’. For HHC transitions, the largest variations in the effects of socio-demographic characteristics were observed among the native-born population, with generally less variation by socio-demographic background characteristics among the immigrant groups. Across all country-of-origin groups, higher transition likelihoods were observed for individuals with lower education, lower income, living alone and residing in less urban areas. Associations with childlessness varied: it was linked to higher relative transition propensities among natives and Nordic immigrants, but to lower relative propensities among Western-origin and Eastern European immigrants. Among non-Western immigrants, childlessness appeared to have little influence on transition propensities. For HHC intensity, where only non-Western immigrants were observed to receive significantly fewer hours of care than natives, subsequent analysis of interaction effects indicated that this difference was entirely contingent on living alone: only Non-Western immigrants living alone had significantly fewer hours of care than natives (living alone or otherwise).

These findings on family composition are particularly interesting in the context of previous qualitative research (Hovde et al. Reference Hovde, Hallberg and Edberg2008b), which highlighted a more prominent role of families in health-care provision among non-Nordic immigrants. From this perspective, our expectations might have run in the opposite direction, with fewer familial resources translating into higher HHC uptake particularly among the non-Nordic immigrant groups. Finding lower transition propensities to be associated with childlessness might instead point to the relevance of children in supporting older immigrants’ navigation, interaction and follow-up with formal health-care systems. This finding may also point to potential barriers to accessing care among older immigrants when limited familial support networks exist.

Whether the lower likelihood of HHC utilization among immigrants reflects reduced needs, differences in preferences, potential discrimination in health system evaluations or a lack of health literacy when navigating complex care systems is not clear from this study. Since the likelihood of transitioning into HHC was so much lower among immigrants than natives, one might anticipate that immigrants who do transition receive more intensive care owing to greater needs. Our findings do not support this. No immigrant subgroup had higher estimated HHC intensities than natives.

On the one hand, the higher HHC transition rates associated with long durations of residence could suggest that disadvantageous ‘acculturation’ or cumulative stressors play a role in promoting worse health outcomes among immigrant populations. With that said, additional sensitivity checks suggest that the inclusion of health status does little to adjust associations by immigrant background or country group of origin. On the other hand, this pattern could also reflect adaptation processes in terms of reduced language barriers and an improvement in knowledge of the Norwegian health and care system, including knowledge of eligibility and rights to care and changes in expectations in terms of who is to deliver care (Horn Reference Horn, Torres and Hunter2023; Park et al. Reference Park, Lee and Kang2018). The relatively high HHC transition rates among Nordic-origin immigrants, as compared to the other immigrant groups, support this interpretation. More generally, our findings align with those of studies examining other forms of public health service utilization in Norway (general practitioner (GP) care, emergency primary care services and institutional care), where immigrants generally use fewer health services compared to natives, and where immigrants from high-income countries and those with longer durations of residence tend to have usage more similar to natives (Debesay et al. Reference Debesay, Arora, Bergland, Borch, Harsløf, Klepp and Laitala2019; Diaz and Kumar Reference Diaz and Kumar2014; Dzamarija Reference Dzamarija2022).

Although most people in Norway report that they receive adequate (medical) care (OECD 2021), population ageing and a more diverse older-age population mean that increased efforts will be required to ensure that disparities in LTC utilization do not occur among certain groups of older persons living at home. Since the allocation of care services requires an active approach, where those in need of care (often with assistance from relatives) are responsible for informing health services about their needs, either by making an appointment with the GP or by applying directly to home health services, it will be important to ensure that older immigrants and their relatives are sufficiently informed about, and able to use, such approaches. This comes as a responsibility not only for immigrants and their families but also for health-care services, given previous findings of difficulties in accessing culturally congruent care among immigrants and lower levels of knowledge about (non-Nordic) immigrants among health-care personnel (Hovde et al. Reference Hovde, Hallberg and Edberg2008a; Khatri and Assefa Reference Khatri and Assefa2022).