Social position and geriatric syndromes among Swedish older people: a population-based study

Older people typically have both age-dependent conditions, like functional decline, and chronic ones, like cardiovascular diseases [1, 2]. Their co-occurrence challenges outcome specific clinical and public health interventions [2, 3]. Modifiable factors like health behavior and social isolation, but also less modifiable factors like social position play a role in the occurrence of these conditions [1, 4, 5]. While there are several social position measures that have a documented direct association with health for both general and chronic conditions, e.g. level of education or income [1, 4, 6, 7], less is known about how they relate to age-dependent conditions.

One group of age-dependent conditions that warrants closer investigation in that respect is geriatric syndromes (GSs) (e.g. loss of hearing and vision or urinary incontinence) [7‐9]. Presence of geriatric syndromes marks older peoples’ frail state. Acting on these syndromes may prevent serious deterioration of existing chronic conditions, as well as the decline in functional disabilities and dependence [2, 10‐12]. Time of onset and severity of GSs may be influenced by older people’s health, social stress and health-related behaviors, all of which are closely related to their social position [1, 5, 12‐14]. Various measures of social position like education, income, but also civil status have different effect on GSs [1, 8, 13, 15]. However, knowledge on the association between social position and geriatric syndromes is scarce and an investigation into a combination of GSs and measures of social position may provide valuable overview into their relationship. To our knowledge, no previous study has considered this association in the context of Sweden.

GSs have been defined as multifactorial health conditions that visualize and contribute to older peoples underlying frailty [10, 16]. As such they present an increased level of vulnerability for situational challenges due to aging processes and accumulated impairments [10]. While GSs are pluriform and prevalent among community-dwellers [17, 18], they are difficult to manage clinically [5, 10, 17, 19], and put older people at increased risk not only for developing new chronic conditions, but also for hospitalization and/or institutionalization [12]. Various GSs share risk factors and etiologies like older age or impaired mobility [10, 18]. Specific GSs, like falls and urinary incontinence, have been extensively studied and their relation with future health outcomes, such as increased mortality, and decreased quality of life is well established [10, 18, 20, 21]. Other GSs, like the course of functional decline and insomnia have been studied to a lesser extend [10, 22]. However, studies combining several types of GS are rare [12, 18, 19], as consensus on a clear universal definition for GSs is lacking [10, 12]. Studying GSs as a set, rather than individual or specific GSs, helps to get a better understanding of other determinants involved such as health factors and social factors [23]. This approach also helps to study the influence of social position on GSs, which may identify health inequalities in regards to the existence of GSs among older people living in society [10].

In this study, we therefore aim to determine the prevalence of GSs and the associations between social position and GSs among older community-dwellers, taking into account health status, health-related behavior and social stress.

In total, 12,333 (70.0%) older community-dwellers reported at least one GS. Table 1 shows insomnia as the most commonly reported GS (39.1%). GSs were more prevalent among women, except for severe hearing loss (men 22.3% vs. women 19.6%). Further, a higher prevalence of GSs was found among older community-dwellers with obesity or underweight, and those indicating to have a chronic condition (Table 2). GSs were particularly prevalent among those who rated their health as bad or very bad (n = 1199, 97.9%).

Among the measures of social position, GS were most prevalent among older people with financial stress (87.7%) (Table 3). All measures of social position were significantly associated with presence of GS (Table 3), when adjusting only for age and sex. Financial stress showed the strongest association with GSs (OR, 3.33; 95% CI, 2.75–4.03).

Financial stress remained strongly associated with GSs after adjustments for health status, i.e. chronic conditions and health behavior. Primary and secondary school education, and unmarried older people did not remain associated with GSs (Table 4) after taking chronic conditions and health behavior into account. Additional adjustment for social stress attenuated the associations of social position with GSs, but remained statistical significant, except for older people born outside the Nordic countries. Financial stress was most strongly associated with the presence of GSs even in the adjusted models (adj. OR, 2.59; 95% CI, 2.13–3.15). Financial stress was also associated with most types of geriatric syndromes (Additional file 2: Table S1.).

Our study shows that GSs were highly prevelant (70.0%) among older community-dwellers in Stockholm County. The observerd prevalance varied acrossdifferent GSs. Older community-dwellers with self-reported financial stress, those not owning housing, as well as those that were widowed had the highest prevalence of geriatric syndromes. Presence of GSs was associated with all five measures of social position i.e. civil status, country of origin, level of education, type of housing and financial stress even after adjustment for age, sex and health status (i.e. chronic conditions). These associations between social position measures and geriatric syndromes remained after additional adjustments for health behavior and social stress, except for measures like education, i.e. primary and seconday school education, civil status, i.e. being unmarried and country of origin, i.e. being born outside of the nordic countries. Financial stress was by far most strongly associated with GSs. Those reporting financial stress were more than twice as common to experience GSs than those who did not.

All GSs and all participants aggregated, the high prevalence we observed compares to some extent to that reported in previous population-based studies on GSs [12, 17, 18], with estimations up to 49.9% [18],76.3% [17] and 80.5% [12], compared to our 70.0%. The difference between studies can be a reflection of differences in either the specific GSs combined, e.g. inclusion of polypharmacy [12] or the population groups, some focusing on female community-dwellers e.g. Women’s Health Initiative Observational Study [17], people aged 75 years and older [12] or community-dwellers and nursing home residents [18]. Our study includes seven of the most common GSs [10, 17, 18, 24, 30].

In regard to specific types of GSs, varying prevalences are also noted between our results and other studies [31, 32], including vision and hearing impairment [18] and falls [17]. However, prevalences vary for some other specific GSs, e.g. the prevalence of urinary incontinence was 29.3% [17] and 9.3% [18] compared to our 26.4%, which partly can be explained by the use of different definitions e.g. use of pads [18] vs. presence of urinary leakage.

When it comes to differences across social measures, it is of note that earlier studies have also shown an association between financial stress and age-dependent health conditions [4, 8, 13, 30, 33‐35], but not yet to prevalent GSs in general, a potential indicator of upcoming new health conditions [10, 17, 18]. A previous study has shown a high prevalence of GSs among people in poverty [13], and studies on specific types of GSs have shown an increased risk with financial stress and also type of housing [34, 35]. A range of factors can explain the association between financial stress and GSs, among which health risk behavior [14], health status [1, 30] and living circumstances [35, 36]. But, it is also of note, that in our study, the association remains strong after taking all these factors into account.

The fact that older people born outside the Nordic countries have higher odds for GSs echoes earlier studies showing poorer health and higher prevalence of GSs in this population group [23, 37]. While we adjusted for contributing factors like social stressors [37], it remains unclear whether the association is a matter of higher physical vulnerability [37], less use of in-hospital care due to difference in illness presention, health illiteracy or communicaion barriers [38] or both. This population group may require particular attention, as prevention of GSs may decrease the risk of new chronic conditions and disability [2, 10, 18].

Poorer health among single persons, divorced and widowed, has been demonstrated in various studies [1, 5, 15]. In our study we also observe a significant association with GSs even after adjustments for chronic conditions, health behavior and social stressors (excluding living alone to avoid overadjustment when analysing civil status).

To date, study results on the association between education and GSs are mixed [1, 5, 13, 18]. We find a significant association between education and GSs in the crude analyses, but not in the adjusted ones suggests that health status, health behavior and social stress contribute largely to explain the association. With the majority of older people, in this study, holding a primary school degree as their highest educational attaintment, the low level of discrimination between levels of education may also affect the results [39].

Besides the large sample size, another strengths were the availability of different self-reported geriatrics syndromes in this population of older swedish community-dwellers from the Stockholm County Council, Public Health Survey. The same protocol was applied throughout the different survey years [24]. While self-reported data is prone to misclassification, it is more suited to capture the prevalence of GSs among older community-dwellers compared to registry information as many GSs are normally not officially coded in registers [40]. While other health surveys only capture some GS, the Stockholm County Council Public Health Survey is quite unique as the questions identify all specific GSs commonly reported in other GSs studies [10, 12, 23, 31, 32].

However, self-reports on GSs are also prone to underreporting, as older people may be less likely to report undesirable conditions [41], or consider them as “normal” age-related conditions. It is difficult to asses the affect of underreporting on our results, but we may have underestimated the magnitude of specific single GSs.

Further limitations concern non-participation bias in the survey. Very disabled older community-dwellers may not be fit enough to respond to the survey due to physical or cognitive impairments leading to an underestimation of the prevalence of GSs. This may weaken the external validity of our study and also for example older people born outside the Nordic countries are underpresented among the participants [24]. Despite lower response-rate among some subgroups, in terms of morbidity the Stockholm County Council Public Health survey data has been shown to be generally comparable to that of the Swedish population aged 65–80 years [42]. Those aged 80 years or older tended to be even slightly healthier than the population [42].

Our study does not allow any conclusions on causality or trajectories, as it investigates cross-sectional associations in terms of prevalence. However, some measures for social position like level of education that are stable over time, especially at the age of 65 and above, can be assumed to precede single or a set of GSs. Yet for other meassures like financial stress and type of housing a potential reverse causality could exist. A longitudinal study would be warranted to disentangle these effects and assess the risks of low social position in developing GSs.

In this large population-based study, there is a high prevalence of GSs among older Swedish community-dwellers. Lower social position, especially the presence of financial stress, was associated with a higher risk of GSs independent of health behaviour and health status. More attention is required on social position when adressing,delaying or even preventing GSs among older community-dwellers.