Sage Journals: Discover world-class research

Abstract

Objective:

While near-centenarians (95–99) and centenarians are the fastest growing sectors of the population in many countries, few studies have investigated their psychological health. We aimed to compare levels of psychological distress and life satisfaction in individuals aged 95 or above (95+) with younger age groups and identify the factors associated with psychological distress and life satisfaction in near-centenarians and centenarians.

Methods:

We assessed the physical, cognitive, social and psychological health of 207 participants aged 95+ in the Sydney Centenarian Study. Psychological distress and life satisfaction were rated on the Kessler Psychological Distress Scale (K10) and Satisfaction with Life Scale, respectively. Cross-sectional univariate comparisons were performed with participants aged 70–90 years from the Sydney Memory and Ageing Study. Factors associated with psychological distress and life satisfaction among Sydney Centenarian Study participants were examined using multiple regression analyses.

Results:

In Sydney Centenarian Study and Memory and Ageing Study, mean K10 scores were 15.3 (±5.9) and 13.4 (±3.6), and clinical levels of psychological distress (K10 ⩾ 20) were 19% and 7%, respectively. Sydney Centenarian Study participants demonstrated significantly higher levels and rates of psychological distress (t = 3.869, p < 0.001; χ² = 27.331, p < 0.001). In Sydney Centenarian Study, more psychotropic medications and having fewer relatives and friends were associated with higher psychological distress. Sydney Centenarian Study participants reported significantly higher levels of life satisfaction than Memory and Ageing Study participants, mean scores 6.0 (±1.5) and 5.6 (±1.3); t = 5.835, p < 0.001. Lower Mini-Mental State Examination scores and having fewer relatives and friends were associated with lower life satisfaction in Sydney Centenarian Study.

Conclusion:

Despite showing higher levels of psychological distress in the prior 4 weeks than younger age groups, near-centenarians and centenarians remained highly satisfied with their overall lives. The identification of risk and protective factors for psychological distress and life satisfaction provides opportunities for interventions to maintain good psychological health in this vulnerable population.

Keywords

Aged 80 and over centenarians mental health psychological distress life satisfaction

Introduction

The near-centenarian and centenarian (95+) population has become the fastest growing sector in many developed countries. In 2015, the estimated 451,000 centenarians in the world are projected to reach 3,676,000 in 2050 (United Nations, Department of Economic and Social Affairs Population Division, 2015). In Australia, the 2016 census data reported 3569 centenarians (Australian Bureau of Statistics (ABS), 2017) and the Australian Institute of Health and Welfare predicted that there will be 15,700 in 2034 and more than 100,000 in 2084 (Australian Institute of Health and Welfare, 2015). Although the 95+ population continues to rise, research into this population remains significantly limited. Studies usually explore the physical health, cognition and functional ability of centenarians; very few specifically investigate their psychological health.

Estimates of the prevalence of depression in near-centenarians and centenarians range from 12.8% to 39.2% (Nyqvist et al., 2017; Tafaro et al., 2002). Anxiety prevalence has been examined in only two studies which reported rates of 45.4% in Portuguese centenarians and 9.5% in Australian centenarians (Ribeiro et al., 2015; Richmond et al., 2011).

Compared to younger age groups, those 95+ have been reported to have higher (Bergdahl et al., 2005; Jang et al., 2004; Nyqvist et al., 2017; Scheetz et al., 2012), similar (Cohen-Mansfield et al., 2013; Dello Buono et al., 1998; von Heideken Wagert et al., 2006) or lower depression scores or rates (Kiljunen et al., 1997).

Positive psychological health, defined by an individual’s level of life satisfaction, has been associated with longevity (Jopp et al., 2016). Life satisfaction is a broad construct that relates to overall well-being, encompassing aspects of emotional and mental health (MacDonald et al., 2009). Findings on the levels of life satisfaction in centenarians vary. Although most studies have reported that life satisfaction is high in centenarians, it remains uncertain whether centenarians are more satisfied with life than the younger cohorts (MacDonald et al., 2009). Martin et al. (1996) reported that 64.5% of participants in the 100+ group rated their life satisfaction as good, compared to 82.8% of participants in the 80–89 group and 81.3% of participants in the 60–69 group; only 1.3% of centenarians stated that their life satisfaction was poor while over 30% rated it as fair. Others have found higher life satisfaction in centenarians compared to octogenarians and nonagenarians (Dello Buono et al., 1998).

Very few studies have investigated factors associated with the psychological health of the oldest-old. Table 1 summarises the current literature on the associations of psychological morbidity and life satisfaction in the near-centenarian and centenarian population. There is significant heterogeneity in the small body of literature on the physical, functional, cognitive, psychiatric and social correlates of psychological health.

Table 1.

An overview of the literature on the correlates of psychological morbidity and life satisfaction in near-centenarians and centenarians.

	Psychological morbidity	Life satisfaction
Age (95+)	No significant association between age and depression (Forsell et al., 1995; Jopp et al., 2016; Kato et al., 2016; Ribeiro et al., 2018)	No significant association between age and life satisfaction (Jopp et al., 2016)
Sex	Sex was not significantly associated with depression (Forsell et al., 1995; Kato et al., 2016; Margrett et al., 2010; Richmond et al., 2011). No significant association was found between sex and anxiety (Ribeiro et al., 2015; Richmond et al., 2011). No significant sex differences for psychological distress (Sachdev et al., 2013)	No significant relationship was found between sex and life satisfaction (Jopp et al., 2016; Sachdev et al., 2013)
Education	Mixed evidence: No significant association between education and depression (Jopp et al., 2016; Kato et al., 2016; Ribeiro et al., 2018). However, a significant association between education and depression has been found (Tafaro et al., 2002)	No significant association between education and life satisfaction (Jopp et al., 2016)
Living situation	Mixed evidence: Centenarians with high levels of care such as assisted living or institutionalisation have shown increased levels of depressive symptoms (Kato et al., 2016; Margrett et al., 2010). However, living alone and institutionalisation did not significantly predict the number of depressive symptoms (Forsell et al., 1995)	No significant relationship was found between residence type and life satisfaction (Jopp et al., 2016)
Number of illnesses	Mixed evidence: A higher number of medical conditions have been found to be a significant predictive factor of anxiety (Ribeiro et al., 2015). Depression scores have been associated with the number of chronic diseases reported (Tafaro et al., 2002) No association was found between a history of medical illnesses and depressive symptoms (Kato et al., 2016)	No significant correlation was found between satisfaction with life and the number of medical illnesses (Araújo et al., 2017; Jopp et al., 2016)
Activities of daily living (ADL)	Mixed evidence: No significant relationship was found between ADL and depression or anxiety (Richmond et al., 2011). However, depression has been found to be positively correlated with greater functional impairment measured by ADL (Forsell et al., 1995; Ribeiro et al., 2018; Tafaro et al., 2002)	Mixed evidence: A significant correlation has been found between higher life satisfaction scores and less functional impairment measured by ADL (Araújo et al., 2017; Jopp et al., 2016). ADL was not significantly associated with life satisfaction but has shown to have an indirect effect in the presence of negative affect (Bishop et al., 2012)
Instrumental activities of daily living (IADL)	Mixed evidence: IADL is highly associated with individual differences in depression (Jopp et al., 2016). Others have found no significant correlation between IADL and depression score (Tafaro et al., 2002)	Less functional impairment measured by IADL score has been associated with higher life satisfaction scores (Araújo et al., 2017; Jopp et al., 2016)
Cognition	Mixed evidence: A negative correlation has been reported between cognition and depressive symptoms (Davey et al., 2013). However, other research has reported no significant correlation (Margrett et al., 2010). Many studies have reported no association between Mini-Mental State Examination (MMSE) score and depression levels (Forsell et al., 1995; Jopp et al., 2016; Kato et al., 2016; Tafaro et al., 2002)	Mixed evidence: A significant relationship has been found between greater life satisfaction and higher levels of cognitive functioning (Araújo et al., 2016). However, others reported that cognition was not significantly correlated with life satisfaction (Bishop et al., 2012; Jopp et al., 2016)
Psychotropic medications	The number of psychotropic drugs was positively associated with depressive symptoms (Forsell et al., 1995)	No evidence yet
Psychiatric history	There was no association between a past psychiatric disorder and depression (Forsell et al., 1995)	No evidence yet
Social support	Mixed evidence: Greater social relationships and social support have been associated with lower depression scores (Tafaro et al., 2002; Toyoshima et al., 2018). The number of relatives was negatively associated with depression (Jopp et al., 2016). However, a significant relationship was not found between social functioning and depression (Richmond et al., 2011) Mixed evidence: Loneliness was found to be a significant predictive factor of anxiety (Ribeiro et al., 2015). However, (Richmond et al., 2011) reported that social functioning is not associated with anxiety	Higher life satisfaction has been associated with greater social resources, with health impairment identified as the mediating factor (Bishop et al., 2006) The number of children was found to be a significant predictor of life satisfaction (Jopp et al., 2016). Greater support from relatives was significantly associated with higher life satisfaction (Jopp et al., 2016)

ADL: activities of daily living; IADL: instrumental activities of daily living.

These discrepancies in findings may be explained in part by methodological differences: failure to obtain a representative sample, the challenging nature of age validation, low participation rates and samples restricted to community-dwellers. The major centenarian studies over the past decade have taken place in Europe and America with few occurring in Australia. This study intended to address some of these limitations of the literature.

The Sydney Centenarian Study (SCS) is an ongoing population-based study that aims to obtain a better understanding of successful cognitive ageing by investigating the physical and mental health in Australian residents aged 95 and above (Sachdev et al., 2013). Cross-sectional comparisons were made with their younger counterparts aged 70–90 years from the Sydney Memory and Ageing Study (MAS) (Sachdev et al., 2010).

Aims and hypotheses

The aims of this study were as follows:

To examine psychological distress and life satisfaction in persons aged 95–106 years (SCS) and make cross-sectional comparisons with persons aged 70–90 years (MAS).

To investigate the risk and protective factors of psychological distress and life satisfaction in near-centenarians and centenarians (SCS).

Based on the most common findings (Bergdahl et al., 2005; Jang et al., 2004; Martin et al., 1996; Nyqvist et al., 2017; Scheetz et al., 2012; Yi and Vaupel, 2002), we hypothesised that those 95+ would show higher levels of psychological distress and be less satisfied with life compared to the younger cohorts. Hypotheses were not formulated for the second aim as there were even fewer studies on the predictors of psychological health in centenarians.

Methods

Participants

SCS

Participants were individuals aged 95 or above from seven local government districts in Sydney, Australia. Sampling strategies comprised invitations using the compulsory Australian electoral roll and Medicare lists on which 99.9% of Australians are registered (Department of Human Services, 2018), along with media campaigns, contacting aged care facilities and other recruitment strategies to achieve as comprehensive a coverage as possible. Written informed consent was given by either the participant or their next of kin. Age validation was achieved using multiple forms of evidence including birth certificates, old passports and marriage certificates. SCS collected data from 343 participants age 95 to 106 years at baseline. When we excluded those with dementia to enable comparison with MAS participants, 207 SCS participants remained. Data were also collected from 156 informants at baseline, who had regular contact with participants (at least 1 hour of contact per week). Further details are available in Sachdev et al. (2013).

Sydney MAS

At baseline, MAS collected data from 1037 participants recruited from the electoral roll, who were non-demented community-dwelling adults aged 70 to 90 years. However, five participants were removed from the sample because they also participated in SCS after they had reached the age of 95, leaving a total sample of 1032. Data were collected from 974 informants at baseline. Informants had at least 1 hour of contact per week with participants. All participants provided written informed consent. Inclusion criteria were participants who had a sufficient level of English proficiency to ensure completion of neuropsychological testing. Exclusion criteria were individuals who had a diagnosis of dementia, a Mini-Mental State Examination (MMSE) score of <24 adjusted for age, educational level and non-English speaking background status (Anderson et al., 2007), psychotic symptoms, bipolar disorder, schizophrenia, multiple sclerosis, progressive malignancy, motor neuron disease or other conditions that prevented the completion of assessments. Further details are available in Sachdev et al. (2010). Both SCS (HC12313) and MAS (HRECO9382) were approved by the Human Ethics Committees of the University of New South Wales and the South Eastern Sydney and Illawarra Area Health Service.

Measures

Methodological harmonisation was attempted when designing SCS and MAS. Both studies used the same measurement tools, enabling comparisons between the two different cohorts.

Psychological distress

The Kessler Psychological Distress Scale (K10) is a 10-item, self-report screening tool for non-specific psychological distress (Kessler et al., 2002). Participants (not informants) are asked how often they have felt distressed in the past four weeks. Each item is scored from 1 to 5, with a minimum score of 10 and a maximum score of 50. Higher scores indicate greater levels of psychological distress. The K10 has been significantly correlated with Diagnostic and Statistical Manual of Mental Disorders Fourth Edition (DSM-IV) mood and anxiety disorders (Slade et al., 2011). A score of ⩾20 suggests clinical levels of psychological distress and a high likelihood of having a mental disorder (Anderson et al., 2013). The K10 is a well-validated tool and in the original study, it demonstrated very high internal consistency with a Cronbach’s α of 0.93 (Kessler et al., 2002). In this study, Cronbach’s α for the combined SCS and MAS cohort was high at 0.83.

Life satisfaction

Life satisfaction of participants was measured by the third item from the Satisfaction with Life Scale (SWLS), which is a 7-point Likert-type scale with a score ranging from 1 to 7 (Diener et al., 1985). A higher score indicates greater levels of subjective well-being. The single-item life satisfaction measure has shown high reliability (Lucas and Brent Donnellan, 2012) and has been strongly correlated with the five-item SWLS (r = 0.75) (Kobau et al., 2010). Internal consistency of the five-item scale is high with a Cronbach’s α of 0.82 in the original study (Diener et al., 1985). The SWLS has been used across different age groups and has been specifically validated among elderly persons (Pavot et al., 1991).

Demographic statistics

Demographic statistics collected for both SCS and MAS comprised age, sex, education level, highest educational qualification attained, family and living conditions, marital status, non-English speaking background status and ethnicity.

Physical, functional, cognitive, psychiatric and social health (SCS only)

Physical health was assessed by calculating the number of current and lifetime illnesses of participants. This was measured using a questionnaire which had a list of common medical illnesses including stroke/transient ischaemic attack, heart attack, angina/blood vessel problems, high blood pressure, high cholesterol, type 2 diabetes, thyroid problems, epilepsy and meningitis/encephalitis. Both participants and informants responded to the questionnaire. If only the participant or the informant answered, this response was recorded. If there were discrepancies between participant and informant data, a diagnosis was recorded if either the participant or informant reported a clinically diagnosed illness. Functional ability was measured by activities of daily living (ADL) via the Lawton and Brody scale, which was completed by an informant (range: 7–35) (Lawton and Brody, 1969). A higher score indicates reduced function, with a cut-off score >7 suggesting impairment (Lawton and Brody, 1969). The Bayer Activities of Daily Living Scale, which was completed by an informant, assessed instrumental activities of daily living (IADL) (range: 1–10) (Hindmarch et al., 1998). A higher score suggests greater functional difficulties and a cut-off score of ⩾3 is usually employed (Folquitto et al., 2007). Cognitive health was assessed using the Addenbrooke Cognitive Examination-Revised (ACE-R) (range: 0–100) (Mathuranath et al., 2000). The questionnaire can be divided into five domains: attention (range: 0–18), fluency (range: 0–14), language (range: 0–26), memory (range: 0–26) and visuospatial (range: 0–16). This test includes a clock-drawing task and embeds the MMSE (range: 0–30) (Folstein et al., 1975).

Psychiatric history was recorded based on an open-ended question ‘what diagnosis was given?’ which was provided to both the participant and informant. A senior psychogeriatrician (H.B.) assessed participant and informant responses to determine the validity of the diagnoses, and the answers of ‘nervous, grievance counselling, grief, change of life, cognitive impairment, nerves and stress’ were not considered psychiatric diagnoses. Answers that were considered valid psychiatric diagnoses are demonstrated in Figure 1.

Figure 1.

Psychiatric history in Sydney Centenarian Study participants.

Discrepancies between participant and informant data were resolved using the same method as above. Psychotropic medications were defined as any medication which belonged to the categories of sedatives, hypnotics, antianxiety agents, antipsychotic agents or antidepressants. A list of current medications was obtained for participants. Social support was measured by the number of relatives (not in the household) and friends that a participant was in regular contact with (visit, phone or write to at least once a month) and the number of confidants, as defined by a relative or friend with whom a participant could talk about private matters and call on for help. The concept of regular social contact is based on the Lubben Social Network Scale (Lubben, 1988).

Missing data

Multiple imputations were performed (30 repetitions) to replace item-level missing data for ADL, MMSE, ACE-R and K10. Thresholds for imputations were determined based on the distribution of missing items for each variable: cases were excluded if more than 29% of items were missing for ADL, if more than 27% of items were missing for MMSE, if more than 20% of items were missing for ACE-R, or if more than 20% of items were missing for K10. Imputations were performed using the MICE package in R (v.3.4.3).

Statistical analyses

Statistical analyses were performed using IBM^™ SPSS^™ 25. For SCS participants, sociodemographic, cognitive and physical profiles were compared between sexes and between near-centenarians and centenarians using independent samples t-tests/Mann–Whitney U tests for continuous variables and χ²/Fisher’s exact tests for categorical variables. Differences in the mean K10 and SWLS scores between SCS and MAS were assessed using independent samples t-tests. K10 scores for SCS and MAS were positively skewed and SWLS scores for SCS and MAS were negatively skewed. We performed log transformations to reduce skewness to between 1 and –1 and so more closely approximate the normal distribution. Predictors for the K10 and SWLS were analysed for the SCS cohort using multiple regression analyses. Multicollinearity was assessed using the variance inflation factor scores obtained from the output of the regression analyses. The sample size was not large enough to include all predictors in the regression model simultaneously without imputations. Therefore, a series of preliminary analyses using domain-specific models (i.e. physical and functional health variables in a single regression model) were initially constructed to find predictors to be included in a final combined model. Predictors from each domain-specific model with p-values ⩽0.2 were included in a combined model. A reduced, final model was then constructed by only including predictors from the combined model with a p ⩽ 0.1, to obtain a more parsimonious model and avoid a large cases-to-predictor ratio.

Results

Sociodemographic characteristics

The sociodemographic characteristics of SCS and MAS participants are shown in Tables 2 and 3 respectively.

Table 2.

Demographics of the Sydney Centenarian Study sample.

			Sex						Age group
	Whole sample		Male		Female				95–99.99 years		100+ years
	n	%	n	%	n	%	Statistic	p	n	%	n	%	Statistic	p
n	207	100.0	70	33.8	137	66.2			188	90.8	19	9.2
Age (M, SD)	97.1	1.9	97.1	2.0	97.1	1.9	t = –0.205	0.838	96.7	1.3	101.5	1.9	t = –11.083	<0.001
Years of education (M, SD)	10.5	3.1	11.0	3.8	10.2	2.7	t = 1.515	0.133	10.5	3.1	10.2	3.5	t = 0.509	0.611
Highest educational qualification^a
Low level^b	86	60.6	29	55.8	57	63.3	χ² = 2.445	0.295	79	59.8	7	70.0	χ² = 0.401	0.818
Middle level^c	37	26.1	13	25.0	24	26.7			35	26.5	2	20.0
High level^d	19	13.4	10	19.2	9	10.0			18	13.6	1	10.0
Total	142	100.0	52	100.0	90	100.0			132	100.0	10	100.0
Type of dwelling
Private dwelling	154	74.4	59	84.3	95	69.3	χ² = 5.431^e	0.028	140	74.5	14	73.7	χ² = 0.006^e	1.000
Assisted living	53	25.6	11	15.7	42	30.7			48	25.5	5	26.3
Total	207	100.0	70	100.0	137	100.0			188	100.0	19	100.0
Marital status^a
Married or de facto	35	21.1	21	35.0	14	13.2	χ² = 12.707	0.005	32	20.8	3	25.0	χ² = 3.615	0.306
Single	9	5.4	1	1.7	8	7.5			7	4.5	2	16.7
Divorced	2	1.2	1	1.7	1	0.9			2	1.3	0	0.0
Widowed	120	72.3	37	61.7	83	78.3			113	73.4	7	58.3
Total	166	100.0	60	100.0	106	100.0			154	100.0	12	100.0
NESB status^f
ESB	154	74.4	52	74.3	102	74.5	χ² = 0.001^e	1.000	136	72.3	18	94.7	χ² = 4.544^e	0.049
NESB	53	25.6	18	25.7	35	25.5			52	27.7	1	5.3
Total	207	100.0	70	100.0	137	100.0			188	100.0	19	100.0
Ethnic background^a
Caucasian	189	96.9	63	94.0	126	98.4	χ² = 6.560	0.087	174	96.7	15	100.0	χ² = 0.516	0.915
Asian	2	1.0	1	1.5	1	0.8			2	1.1	0	0.0
Pacific Islander	1	0.5	0	0.0	1	0.8			1	0.6	0	0.0
Other	3	1.5	3	4.5	0	0.0			3	1.7	0	0.0
Total	195	100.0	67	100.0	128	100.0			180	100.0	15	100.0

Numbers sum less than 207 when data were missing.

Low level = primary school or intermediate certificate.

Middle level = leaving certificate or diploma.

High level = bachelor’s or postgraduate degree.

Fisher’s exact test.

NESB = non-English speaking background, ESB = English speaking background.

Table 3.

Demographics of the Sydney Memory and Ageing Study sample.

	Whole sample		Sex				Statistic	p
	Whole sample		Male		Female
	n	%	n	%	n	%
n	1032	100.0	464	45.0	568	55.0
Age (M, SD)	78.8	4.8	78.7	4.7	78.8	4.9	t = –0.278	0.781
Years of education (M, SD)	11.6	3.5	12.3	3.8	11.0	3.0	t = 5.721	<0.001
Highest educational qualification
Low level^a	435	42.2	155	33.4	280	49.3	χ² = 27.990	<0.001
Middle level^b	254	24.6	124	26.7	130	22.9
High level^c	343	33.2	185	39.9	158	27.8
Total	1032	100.0	464	100.0	568	100.0
Type of dwelling
Private dwelling	1006	97.5	452	97.4	554	97.5	χ² = 0.000	1.000
Other community dwelling^d	26	2.5	12	2.6	14	2.5
Total	1032	100.0	464	100.0	568	100.0
Marital status^e
Never married	132	12.8	71	15.4	61	10.8	χ² = 149.495	<0.001
Married or de facto	428	41.6	273	59.1	155	27.3
Divorced or separated	124	12.1	46	10.0	78	13.8
Widowed	345	33.5	72	15.6	273	48.1
Total	1029	100.0	462	100.0	567	100.0
NESB status^f
ESB	869	84.2	382	82.3	487	85.7	χ² = 1.986	0.135
NESB	163	15.8	82	17.7	81	14.3
Total	1032	100.0	464	100.0	568	100.0
Ethnic background^e
Caucasian	1011	98.3	449	97.4	562	99.1	χ² = 7.390^g	0.033
Asian	10	1.0	7	1.5	3	0.5
Mixed	3	0.3	1	0.2	2	0.4
Other	4	0.4	4	0.9	0	0.0
Total	1028	100.0	461	100.0	567	100.0

Low level = Primary school or incomplete high school.

Middle level = Complete high school or certificate/diploma.

High level = Incomplete or complete bachelor’s or postgraduate degree.

Other community dwelling = Hostel villa, retirement home or other.

Numbers sum less than 1032 when data were missing.

NESB = Non-English speaking background; ESB = English speaking background.

Fisher’s exact test.

Psychological health

SCS participants had significantly higher K10 scores than MAS participants (Table 4). There were no sex differences in K10 score in SCS, while in MAS, women (M = 13.7, SD 3.8) had significantly higher K10 scores than men (M = 13.2, SD 3.5; U = 115,959, p < 0.05). On the K10, 19% of SCS and only 7% of MAS participants had clinical levels of psychological distress, that is, score ⩾20 (χ² = 27.331, p < 0.001).

Table 4.

Psychological distress and life satisfaction of the Sydney Centenarian Study and Sydney Memory and Ageing Study.

	SCS			MAS			T	p
	n	M	SD	n	M	SD	T	p
K10	166	15.3	5.9	1015	13.4	3.6	3.869	<0.001
SWLS	172	6.0	1.5	994	5.6	1.3	5.835	<0.001

K10: Kessler Psychological Distress Scale; higher scores indicate greater psychological distress; SWLS: Satisfaction with Life Scale; higher scores indicate greater life satisfaction; SCS: Sydney Centenarian Study; MAS: Memory and Ageing Study.

On the SWLS, SCS participants’ scores were significantly higher than MAS participants’ scores (Table 4). There were no sex differences in SWLS score in SCS or MAS.

The SCS: physical, functional, cognitive, psychiatric and social status

The physical, functional, cognitive, psychiatric and social characteristics of SCS participants are shown in Table 5. There was evidence of functional impairment (mean ADL score of 10.2 was well above the cut-off point of >7). Men were less functionally impaired, scoring much lower on Bayer-ADL than women and had a higher mean ACE-R total score than women, performing significantly better on the language and visuospatial subscales. On average, near-centenarians had significantly more illnesses compared to centenarians. There were no other significant differences between men and women and between near-centenarians and centenarians.

Table 5.

Predictor variables for the Sydney Centenarian Study.

SCS	Whole sample			Sex				Statistic	p	Age group				Statistic	p
	Whole sample			Male		Female				95–99.99 years		100+ years
	n	M	SD	M	SD	M	SD			M	SD	M	SD
History of medical illnesses^a	201	2.1	1.3	2.1	1.3	2.1	1.4	t = –0.162	0.872	2.2	1.3	1.4	1.2	t = 2.191	0.030
ADL	152	10.2	4.2	9.6	3.7	10.5	4.4	t = –1.187	0.237	10.0	4.0	12.3	6.3	t = –1.133	0.285
Bayer-ADL	155	2.8	1.9	2.4	1.3	3.1	2.0	t = –2.479	0.014	2.8	1.9	3.1	1.9	t = –0.428	0.669
MMSE	192	25.7	2.7	25.9	3.2	25.5	2.5	t = 0.784	0.434	25.7	2.6	24.7	3.7	t = 1.496	0.136
ACE-R total	180	74.3	10.7	77.7	10.4	72.6	10.5	t = 3.042	0.003	74.4	10.9	73.1	8.4	t = 0.478	0.633
ACE-R attention	180	16.5	1.6	16.7	1.7	16.5	1.6	U = 3.145	0.152	16.6	1.6	16.1	1.5	U = 1.020	0.128
ACE-R fluency	180	7.2	2.8	7.7	2.9	6.9	2.8	t = 1.757	0.081	7.2	2.9	6.8	2.4	t = 0.596	0.552
ACE-R language	180	20.8	3.3	21.6	3.2	20.4	3.3	t = 2.366	0.019	20.8	3.4	20.8	2.5	t = –0.016	0.987
ACE-R memory	180	17.4	5.1	18.2	5.1	17.0	5.0	t = 1.536	0.126	17.4	5.1	17.4	4.6	t = –0.058	0.953
ACE-R visuospatial	180	12.4	2.3	13.5	2.1	11.9	2.3	t = 4.503	0.000	12.5	2.4	11.9	2.3	t = 0.907	0.365
Past psychiatric history, N (%)	21	10.3	–	9.0	13.0	12.0	8.9	χ² = 0.853^b	0.465	21.0	11.2	0.0	0.0	χ² = 1.992^b	0.382
Number of psychotropic medications	179	0.4	0.7	0.3	0.6	0.5	0.8	t = –1.399	0.164	0.4	0.7	0.5	0.9	t = –0.538	0.592
Number of relatives and friends	163	8.7	7.1	8.4	6.9	8.9	7.2	t = –0.412	0.681	8.8	7.2	7.5	6.1	t = 0.600	0.549
Number of confidants	163	3.7	3.9	3.8	3.6	3.7	4.0	t = 0.185	0.853	3.7	3.9	3.5	3.4	t = 0.163	0.871

SCS: Sydney Centenarian Study; ADL: Activities of Daily Living; higher scores indicate worse functioning; Bayer-ADL: Bayer Activities of Daily Living; higher scores indicate worse functioning; MMSE: Mini-Mental State Examination; higher scores indicate greater memory; ACE-R: Addenbrooke Cognitive Examination-Revised; higher scores indicate greater cognitive functioning.

Number of illnesses affirmed from a list of nine possible illnesses.

Fisher’s exact test.

Risk and protective factors of psychological distress and life satisfaction

We analysed the predictive factors of psychological distress and life satisfaction in SCS participants using linear regression models by domains of demographics, physical functioning, cognition, psychiatric health and social resources. As MMSE and ACE-R had a correlation coefficient of r = 0.72 (see Supplemental Appendix 2), producing variance inflation scores of greater than 2, these predictor variables were entered in separate cognitive health models (cognitive models 1 and 2 in Table 6) to reduce multicollinearity.

Table 6.

Regression models for psychological distress and life satisfaction in the Sydney Centenarian Study.

	K10				SWLS
	B	β	p	R²	B	β	p	R²
Demographic model				0.01				0.03
Age	0.00	−0.07	0.41		0.00	0.03	0.74
Gender (1 = male, 2 = female)	−0.01	−0.04	0.65		0.02	0.05	0.56
Education	0.00	−0.05	0.54		0.00	−0.04	0.61
Living situation (1 = private dwelling, 2 = assisted living)	−0.02	−0.05	0.55		−0.11	−0.18	0.02
Physical functioning model				0.02				0.06
History of medical illnesses	0.01	0.07	0.42		0.00	0.02	0.83
ADL	0.00	−0.01	0.93		−0.01	−0.14	0.16
Bayer-ADL	0.01	0.13	0.21		−0.02	−0.13	0.16
Cognitive model 1				0.00				0.04**
MMSE	0.00	0.01	0.87		0.02	0.20	0.01
Cognitive model 2				0.00				0.03*
ACE-R	0.00	0.00	0.96		0.00	0.16	0.04
Psychiatric health model				0.13**				0.04*
Number of psychotropic medications	0.06	0.29	0.00		−0.06	−0.17	0.04
Past psychiatric history (0 = no, 1 = yes)	0.07	0.16	0.05		−0.07	−0.08	0.32
Social resources model				0.05*				0.06**
Number of relatives and friends	0.00	−0.22	0.02		0.01	0.20	0.03
Number of confidants	0.00	−0.03	0.76		0.00	0.07	0.45
Combined model				0.16**				0.11**
Living situation (1 = private dwelling, 2 = assisted living)	–	–	–		−0.08	−0.13	0.09
MMSE	–	–	–		0.02	0.19	0.02
Number of psychotropic medications	0.06	0.27	0.00		–	–	–
Past psychiatric history (0 = no, 1 = yes)	0.06	0.13	0.12		–	–	–
Number of relatives and friends	0.00	−0.20	0.02		0.01	0.19	0.02

Selection of variables for the combined regression models was performed by including all variables with a p ⩽ 0.2 in the separate domain-specific models. A stricter threshold of p ⩽ 0.1 was then applied to reduce the number of predictors. The symbol “–” signifies that the variable was not selected for the combined regression model. Combined model for psychological distress: N = 136. Combined model for life satisfaction: N = 153. Listwise deletion was used.

ADL: Activities of Daily Living; higher scores indicate worse functioning; Bayer-ADL: Bayer Activities of Daily Living; higher scores indicate worse functioning; MMSE: Mini-Mental State Examination; higher scores indicate greater memory; ACE-R: Addenbrooke Cognitive Examination-Revised; higher scores indicate greater cognitive functioning; K10: Kessler Psychological Distress Scale; higher scores indicate greater psychological distress; SWLS: Satisfaction with Life Scale; higher scores indicate greater life satisfaction.

p < 0.01; *p < 0.05.

Demographic, physical functioning and cognitive health variables were not significantly associated with K10 score. By contrast, higher K10 scores were significantly associated with more psychotropic medications and fewer relatives and friends. The combined model with these two variables explained approximately 16.3% of the variance in psychological distress.

Life satisfaction was not significantly correlated with demographic, physical functioning and psychiatric health variables. Higher MMSE scores and regular contact with relatives and friends were significantly associated with higher life satisfaction. The combined model explained 11.1% of the variance in life satisfaction.

Discussion

The present study aimed to examine the psychological health of near-centenarians and centenarians by making cross-sectional comparisons with younger cohorts and investigating the risk and protective factors associated with psychological distress and life satisfaction. To the best of our knowledge, this is the largest study to examine psychological distress in those 95+ and the first study to investigate this using the K10.

As predicted and consistent with four previous studies (Bergdahl et al., 2005; Jang et al., 2004; Nyqvist et al., 2017; Scheetz et al., 2012), persons aged 95+ reported significantly higher levels of psychological distress than the younger counterparts aged 70–90 years. One reported explanation is that risk factors for psychological morbidity increase during the ageing process as the oldest-old experience greater loneliness, impaired mobility, declines in physical and cognitive function and institutionalisation (Stek et al., 2004). For example, 25.6% of the SCS cohort resided in assisted living facilities compared to 1.9% of the MAS cohort.

Our findings that 81% of SCS participants scored below the K10 cut-off of ⩾20 are consistent with reports from Australia and the United States. Richmond et al. (2011) reported that 86% of Australian centenarians did not describe depression; Jopp et al. (2016) from the United States reported that 80% of those 95+ did not have clinical levels of depressive symptoms, as assessed using a cut-off of ⩾8 on the 15-item Geriatric Depression Scale (GDS-15).

Inconsistent with our hypothesis, near-centenarians and centenarians reported significantly higher levels of life satisfaction compared to the younger age groups. This is a particularly interesting finding given that those 95+ were more psychologically distressed. Previous research has also reported these seemingly paradoxical results. Scheetz et al. (2012) found that although centenarians had higher scores than sexagenarians and octogenarians on the Geriatric Depression Scale (GDS-30), more reported being in good spirits than the younger age groups. Possible theoretical explanations are that high levels of life satisfaction may be a survivor effect or may occur as part of an adaptation process: centenarians accept the inevitable loss of function and reduction in activity and become content with their lives (Dello Buono et al., 1998).

The design of the instruments used to measure psychological distress and life satisfaction may also explain this seeming paradox. The K10 measures psychological distress over the past 4 weeks whereas the SWLS assesses satisfaction with a person’s life as a whole. Near-centenarians and centenarians may be psychologically distressed in the short-term, whereas in the long-run, they remain optimistic and satisfied with their overall lives. Furthermore, it is important to note that the K10 and SWLS measure different constructs. We found moderate negative correlations between K10 and SWLS in both the SCS and MAS cohort (see Supplemental Appendix 2).

Our findings from the SCS and MAS comparisons contradict previous research by Martin et al. (1996) and Yi and Vaupel (2002), who reported that centenarians were less satisfied with their lives than younger cohorts. These studies utilised different measurement tools to assess life satisfaction: Martin et al. (1996) employed the Older Americans Resources and Services (OARS) questionnaire and Yi and Vaupel (2002) asked a single item ‘how do you rate your life at present?’ with only four levels of responses available instead of the 7-point Likert-type scale used in this study. It is also possible that different sampling strategies, cultural diversity, age ranges, and inclusion and exclusion criteria for the studies may explain the variation in findings.

In line with previous findings, our study revealed that more psychotropic medications were associated with higher psychological distress (Forsell et al., 1995), although the direction of causality remains unclear. It may be that psychotropic drugs worsen psychological distress through medication interactions and side-effects; that those with high levels of psychological distress are more likely to be prescribed psychotropic drugs or that the association is bidirectional.

We found that regular contact with fewer relatives and friends was associated with greater psychological distress. These results confirm findings by Jopp et al. (2016) that a lack of support from relatives and friends significantly predicted levels of depressive symptoms, although they contradict Richmond et al. (2011) who found no association between levels of depression and social functioning. However, social support was operationalised differently in the two studies. Richmond et al. (2011) measured social functioning by assessing marital status, participation in social activities and time spent with families whereas this study and that by Jopp et al. (2016) used the number of relatives and friends as an indicator of social resources.

Surprisingly, we did not find that living situation was a significant predictor of psychological distress, unlike Margrett et al. (2010) who reported that institutional residence was associated with higher depression levels. However, depression and psychological distress are two different albeit overlapping concepts and comparisons should be treated with caution. Similar to Jopp et al. (2016) in the Fordham Centenarian Study, we found no association between residence type and life satisfaction. However, a large majority of the sample in both studies (74%) lived in a private dwelling, possibly implying a selection bias.

In our findings, ADL and Bayer-ADL did not emerge as significant predictors, in contrast to previous literature. One possible reason for the difference is that functional health was assessed by the informant, whereas previous studies (Jopp et al., 2016; Ribeiro et al., 2018) have only relied on participant responses, which has been associated with greater reporting bias (Jekel et al., 2015). Consistent with previous research (Forsell et al., 1995; Jopp et al., 2016; Kato et al., 2016), we found that cognition was not a significant predictor of psychological distress in near-centenarians and centenarians.

In line with previous findings, greater cognition was a significant predictor of higher life satisfaction. Higher cognitive functioning may facilitate individuals to perceive the positive aspects of life and adapt to the processes of ageing, thereby improving their satisfaction with life (Enkvist et al., 2013).

Regular contact with more relatives and friends was significantly associated with higher life satisfaction. Family members and friends often provide meaningful and enriching experiences that help maintain high levels of life satisfaction in the very old (Jopp et al., 2016).

Strengths and limitations

The current study is the largest sample of near-centenarians and centenarians to be studied in Australia. A further strength is the stringent age validation method to ensure the reliability of participants’ ages. The use of multiple sources of evidence to verify age safeguarded the study against inaccurate reporting.

Despite use of a mixed population and convenience sample, the entire SCS cohort from which the cognitively intact sample was drawn, was reasonably representative as compared to 2011 ABS census data (ABS, 2011) for the same geographical areas sampled. There were no significant differences in the proportion of females (71.7% in SCS vs 76.4% in the ABS data, p = 0.08), proportion of individuals in the 100+ age group (12.0% in SCS vs 8.6% in the ABS data, p = 0.08) proportion of individuals living in a private dwelling (58.6% in SCS vs 54.7% in the ABS data, p = 0.02) or in the educational qualifications attained (17.6% had a tertiary education, 18.1% had a leaving certificate and 64.3% did not complete a leaving certificate in SCS, compared to 15.6%, 22.6% and 61.8%, respectively, in the ABS data, p = 0.33).

However, the current findings cannot be generalised to all persons aged 95+ as the present study was restricted to seven local government areas in south-east Sydney, Australia. In addition, we only included centenarians without dementia to ensure the reliability of the psychological self-report measures. Utilising cognitively intact centenarians increases the reliability of self-report information about mental health status. Consequently, our findings may not characterise centenarians with cognitive impairment or dementia. Recruitment bias may have influenced the comparison of SCS and MAS. In the MAS study, participants were recruited solely from the electoral roll but had a response rate of 19.9% (1772 out of 8914 people) (Sachdev et al., 2010). Therefore, the findings may be susceptible to the healthy volunteer bias.

Although the minimum requirement for informants in this study was 1 hour of contact a week with participants, informants were relatives who knew participants well. In addition, the average time of contact was significantly higher in both SCS (7.6 hours) and MAS (8.3 hours).

A limitation of this study is the cross-sectional design, which does not allow causal relationships between variables to be inferred as well as the directions of associations. It is also unclear whether the differences in levels of psychological distress and life satisfaction between SCS and MAS are a result of age-related changes or cohort effects.

Although we used the Lubben Social Network Scale as a framework for defining regular social contact, regular contact may vary across cultures as demonstrated by Georgas et al. (2001). Finally, our regression model only explained 11% in the variance of life satisfaction, indicating that other important factors influence life satisfaction such as the extraversion personality trait and an optimistic outlook on life (Kato et al., 2016).

Conclusion

This study adds to the small body of research on the mental health of near-centenarians and centenarians. Our findings have considerable implications for the development of services tailored towards the oldest-old and their specific needs. Despite the growing population of centenarians, many developed countries including Australia have few policies adapted to this unique group of individuals. Given that the 95+ population suffer higher rates of psychological distress than younger age groups, policies could focus on prevention through enhancing social networks.

Future research should continue to examine the prevalence and risk factors of psychological morbidity and life satisfaction in the oldest-old due to the conflicting results. Other facets of psychological health should be examined, including behavioural and psychological symptoms of dementia and apathy, as these concepts have not been investigated in those at the extreme ages, as well as personality variables. Longitudinal data could examine the relationship between declines in psychological health and time to death. Longitudinal analysis can also provide insight into whether the variations in psychological health are caused by age or cohort differences. The findings of this study are an important step in determining the specific needs of the oldest-old, who are the fastest growing segment in Australia and other developed countries and point the way to improved care for this vulnerable group.

Supplemental Material

Supplemental – Supplemental material for The psychological health of 207 near-centenarians (95–99) and centenarians from the Sydney Centenarian Study

Supplemental material, Supplemental for The psychological health of 207 near-centenarians (95–99) and centenarians from the Sydney Centenarian Study by Adrian Cheng, Yvonne Leung, John D Crawford, Fleur Harrison, Perminder Sachdev and Henry Brodaty in Australian & New Zealand Journal of Psychiatry

Footnotes

Acknowledgements

The authors would like to thank the study coordinators and research assistants of SCS and MAS at CHeBA for data collection and Dr Kristan Kang for data management support.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: The Sydney Centenarian Study was funded by the National Health and Medical Research Council (NHMRC) of Australia (Project Grant 630593 and Programme Grant 568969). The Sydney Memory and Ageing Study was funded by a NHMRC Programme Grant (350833).

ORCID iDs

Adrian Cheng

Yvonne Leung

Henry Brodaty

Supplemental material

Supplemental material for this article is available online.

References

Anderson

Sachdev

Brodaty

, et al. (2007) Effects of sociodemographic and health variables on mini-mental state exam scores in older Australians. American Journal of Geriatric Psychiatry 15: 467–476.

Anderson

Sunderland

Andrews

, et al. (2013) The 10-item Kessler Psychological Distress Scale (K10) as a screening instrument in older individuals. American Journal of Geriatric Psychiatry 21: 596–606.

Araújo

Ribeiro

Paúl

(2017) The role of existential beliefs within the relation of centenarians’ health and well-being. Journal of Religion & Health 56: 1111–1122.

Araújo

Ribeiro

Teixeira

, et al. (2016) Successful aging at 100 years: The relevance of subjectivity and psychological resources. International Psychogeriatrics 28: 179–188.

Australian Bureau of Statistics (ABS) (2011) Census of population and housing. Available at: www.abs.gov.au/ (accessed 26 October 2018).

Australian Bureau of Statistics (ABS) (2017) 2071.0 – Census of population and housing: Reflecting Australia: Stories from the census, 2016. Available at: www.abs.gov.au/AUSSTATS/abs@.nsf/DetailsPage/2071.02016. (accessed 10 March 2018)

Australian Institute of Health and Welfare (2015) Australia’s welfare 2015. Available at: www.aihw.gov.au/getmedia/d523980b-d67c-4b8b-869f-563ab8568453/AW15-6-1-Australians-aged-85-years-and-over.pdf.aspx (accessed 10 March 2018).

Bergdahl

Gustavsson

Kallin

, et al. (2005) Depression among the oldest old: The Umea 85+ study. International Psychogeriatrics 17: 557–575.

Bishop

Martin

Poon

(2006) Happiness and congruence in older adulthood: A structural model of life satisfaction. Aging & Mental Health 10: 445–453.

10.

Bishop

Martin

Randall

, et al. (2012) Exploring life satisfaction in exceptional old age: The mediating role of positive and negative affect. Clinical Gerontologist 35: 105–125.

11.

Cohen-Mansfield

Shmotkin

Blumstein

, et al. (2013) The old, old-old, and the oldest old: Continuation or distinct categories? An examination of the relationship between age and changes in health, function, and well-being. International Journal of Aging and Human Development 77: 37–57.

12.

Davey

Dai

Woodard

, et al. (2013) Profiles of cognitive functioning in a population-based sample of centenarians using factor mixture analysis. Experimental Aging Research 39: 125–144.

13.

Dello Buono

Urciuoli

De Leo

(1998) Quality of life and longevity: A study of centenarians. Age and Ageing 27: 207–216.

14.

Department of Human Services (2018) 2017-2018 Annual report. Canberra, ACT, Australia: Commonwealth of Australia, Australian Government.

15.

Diener

Emmons

Larsen

, et al. (1985) The Satisfaction with Life Scale. Journal of Personality Assessment 49: 71–75.

16.

Enkvist

Ekström

Elmståhl

(2013) Associations between cognitive abilities and life satisfaction in the oldest-old: Results from the longitudinal population study Good Aging in Skåne. Clinical Interventions in Aging 8: 845–853.

17.

Folquitto

Bustamante

SEZ

Barros

, et al. (2007) The Bayer: Activities of daily living scale (B-ADL) in the differentiation between mild to moderate dementia and normal aging. Revista Brasileira de Psiquiatria 29: 350–353.

18.

Folstein

McHugh

(1975) ‘Mini-mental state’: A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research 12: 189–198.

19.

Forsell

Jorm

von Strauss

, et al. (1995) Prevalence and correlates of depression in a population of nonagenarians. British Journal of Psychiatry 167: 61–64.

20.

Georgas

Mylonas

Bafiti

, et al. (2001) Functional relationships in the nuclear and extended family: A 16-culture study. International Journal of Psychology 36: 289–300.

21.

Hindmarch

Lehfeld

de Jongh

, et al. (1998) The Bayer activities of daily living scale (B-ADL). Dementia and Geriatric Cognitive Disorders 9: 20–26.

22.

Jang

Poon

Martin

(2004) Individual differences in the effects of disease and disability on depressive symptoms: The role of age and subjective health. International Journal of Aging and Human Development 59: 125–137.

23.

Jekel

Damian

Wattmo

, et al. (2015) Mild cognitive impairment and deficits in instrumental activities of daily living: A systematic review. Alzheimer’s Research & Therapy 7: 17.

24.

Jopp

Park

Lehrfeld

, et al. (2016) Physical, cognitive, social and mental health in near-centenarians and centenarians living in New York City: Findings from the Fordham Centenarian Study. BMC Geriatrics 16: 1.

25.

Kato

Zweig

Schechter

, et al. (2016) Positive attitude toward life, emotional expression, self-rated health, and depressive symptoms among centenarians and near-centenarians. Aging & Mental Health 20: 930–939.

26.

Kessler

Andrews

Colpe

, et al. (2002) Short screening scales to monitor population prevalences and trends in non-specific psychological distress. Psychological Medicine 32: 959–976.

27.

Kiljunen

Sulkava

Niinisto

, et al. (1997) Depression measured by the Zung depression status inventory is very rare in a Finnish population aged 85 years and over. International Psychogeriatrics 9: 359–368.

28.

Kobau

Sniezek

Zack

, et al. (2010) Well-being assessment: An evaluation of well-being scales for public health and population estimates of well-being among US adults. Applied Psychology: Health and Well-Being 2: 272–297.

29.

Lawton

Brody

(1969) Assessment of older people: Self-maintaining and instrumental activities of daily living. Gerontologist 9: 179–186.

30.

Lubben

(1988) Assessing social networks among elderly populations. Family and Community Health 11: 42–52.

31.

Lucas

Brent Donnellan

(2012) Estimating the reliability of single-item life satisfaction measures: Results from four national panel studies. Social Indicators Research 105: 323–331.

32.

MacDonald

Martin

Margrett

, et al. (2009) Correspondence of perceptions about centenarians’ mental health. Aging & Mental Health 13: 827–837.

33.

Margrett

Martin

Woodard

, et al. (2010) Depression among centenarians and the oldest old: Contributions of cognition and personality. Gerontology 56: 93–99.

34.

Martin

Poon

Kim

, et al. (1996) Social and psychological resources in the oldest old. Experimental Aging Research 22: 121–139.

35.

Mathuranath

Nestor

Berrios

, et al. (2000) A brief cognitive test battery to differentiate Alzheimer’s disease and frontotemporal dementia. Neurology 55: 1613–1620.

36.

Nyqvist

Cattan

Conradsson

, et al. (2017) Prevalence of loneliness over ten years among the oldest old. Scandinavian Journal of Public Health 45: 411–418.

37.

Pavot

Diener

Colvin

, et al. (1991) Further validation of the Satisfaction with Life Scale: Evidence for the cross-method convergence of well-being measures. Journal of Personality Assessment 57: 149–161.

38.

Ribeiro

Duarte

Teixeira

, et al. (2018) Frailty and depression in centenarians. International Psychogeriatrics 30: 115–124.

39.

Ribeiro

Teixeira

Araújo

, et al. (2015) Predictors of anxiety in centenarians: Health, economic factors, and loneliness. International Psychogeriatrics 27: 1167–1176.

40.

Richmond

Law

Kay-Lambkin

(2011) Physical, mental, and cognitive function in a convenience sample of centenarians in Australia. Journal of the American Geriatrics Society 59: 1080–1086.

41.

Sachdev

Brodaty

Reppermund

, et al. (2010) The Sydney Memory and Ageing Study (MAS): Methodology and baseline medical and neuropsychiatric characteristics of an elderly epidemiological non-demented cohort of Australians aged 70-90 years. International Psychogeriatrics 22: 1248–1264.

42.

Sachdev

Levitan

Crawford

, et al. (2013) The Sydney Centenarian Study: Methodology and profile of centenarians and near-centenarians. International Psychogeriatrics 25: 993–1005.

43.

Scheetz

Martin

Poon

(2012) Do centenarians have higher levels of depression? Findings from the Georgia Centenarian Study. Journal of the American Geriatrics Society 60: 238–242.

44.

Slade

Grove

Burgess

(2011) Kessler Psychological Distress Scale: Normative data from the 2007 Australian national survey of mental health and well-being. Australian and New Zealand Journal of Psychiatry 45: 308–316.

45.

Stek

Gussekloo

Beekman

, et al. (2004) Prevalence, correlates and recognition of depression in the oldest old: The Leiden 85-plus study. Journal of Affective Disorders 78: 193–200.

46.

Tafaro

Cicconetti

Zannino

, et al. (2002) Depression and aging: A survival study on centenarians. Archives of Gerontology and Geriatrics 35: 371–376.

47.

Toyoshima

Martin

Sato

, et al. (2018) The relationship between vision impairment and well-being among centenarians: Findings from the Georgia Centenarian Study. International Journal of Geriatric Psychiatry 33: 414–422.

48.

United Nations, Department of Economic and Social Affairs Population Division (2015) World population prospects: The 2015 revision. New York: United Nations.

49.

von Heideken Wagert

Gustavsson

JMC

Lundin-Olsson

, et al. (2006) Health status in the oldest old: Age and sex differences in the Umea 85+ study. Aging Clinical and Experimental Research 18: 116–126.

50.

Vaupel

(2002) Functional capacity and self-evaluation of health and life of oldest old in China. Journal of Social Issues 58: 733–748.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.28 MB