Subjective Cognitive Decline May Be a Stronger Predictor of Incident Dementia in Women than in Men

Abstract

Background/Objective:

Subjective cognitive decline (SCD) has often been associated with an increased risk for subsequent dementia. However, sex-specific associations are understudied until now.

Methods:

Cross-sectional and longitudinal associations over a follow-up period of up to 13 years were investigated in a sample of participants without objective cognitive impairment at baseline (n = 2,422, mean age = 79.63 years). Logistic regression and Cox proportional hazards models were conducted.

Results:

Women less frequently reported SCD without worries (p < 0.001), but tended to report more often SCD with worries (p = 0.082) at baseline compared to men. In models adjusted for age, education, cognitive status, and depressive symptoms, SCD at baseline increased the risk for subsequent dementia (p < 0.001), and this effect was less pronounced in males (interaction sex×SCD: p = 0.022). Stratified analyses showed that SCD increased the risk for subsequent dementia in women (HR = 1.77, p < 0.001), but not in men (HR = 1.07, p = 0.682). Similar results were found in analyses with SCD without and with worries, except that SCD with worries also predicted subsequent Alzheimer’s disease (AD) in men (p = 0.037).

Conclusion:

At baseline, men reported more SCD without worries and women tended to report more SCD with worries. SCD in women was more strongly associated with subsequent dementia. SCD without and with worries was related to incident dementia and AD in women, whereas in men only SCD with worries increased the risk for AD, but not for all-cause dementia.

Keywords

Alzheimer’s disease dementia gender sex subjective cognitive decline subjective memory decline subjective memory impairment

INTRODUCTION

The interest in subjective memory impairment, or more general, subjective cognitive decline (SCD) as a possible early sign of incipient dementia has increased during the last decade. It can be defined as the self-report on perceived cognitive decline in memory or other cognitive domains such as executive function or attention. A meta-analysis found a significantly increased conversion rate to dementia in individuals with SCD without objective cognitive impairment [1]. An international working group called the Subjective Cognitive Decline Initiative (SCD-I) has been established to advance related research [2]. According to a recent research framework [3], SCD in individuals with normal cognitive performance and a positive amyloid-β (tau-) biomarker profile constitutes stage 2 of the numeric clinical Alzheimer’s disease (AD) staging.

Cross-sectionally, SCD is more strongly related with affect and mood such as depression and anxiety (for example, [4, 5]), rather than with objective cognitive function. Hence, SCD is often not a good indicator of current cognitive deficits. However, several studies provided evidence for the prognostic validity of SCD to predict the risk for subsequent dementia [2]. For example, the risk for AD was about 6.5-fold increased by SCD with worries over up to three years in an earlier study of the present cohort [6].

There is some evidence that the relationship between SCD and subsequent dementia differs in different demographic subpopulations. For example, a closer association between SCD and subsequent AD might exist in higher-educated individuals [7]. The cross-sectional association between frequency of SCD and sex is understudied. Available results are heterogeneous. Some studies found no association between SCD and sex [5 , 8–10], whereas others found SCD more frequently in males [11 –13] or in females [14 –17].

Divergent results might be explained by inconsistent methods used to examine the association. The study setting (population-based versus memory clinics), the operationalization of SCD, and the reported SCD severity might affect results [2, 18]. Although a recent meta-analysis found similar conversion rates of individuals with SCD to mild cognitive impairment (MCI) and dementia in community and non-community settings [1], this study did not address sex-specific associations. Moreover, the definition of cognitive impairment, the in-/exclusion of cognitively impaired participants, and the age of samples might affect the results. Some studies with younger or lifespan samples found more SCD in males [11, 13], whereas more SCD in males was less frequently [14 , 16–17], but also found in elderly samples [12]. Beyond the cross-sectional distribution of SCD in males and females, even less is known about the sex-specific longitudinal association between SCD and the subsequent risk for AD and dementia. To the best of our knowledge, the only exception is the study by Pérès et al. [19], in which subjective memory complaints nearly doubled the subsequent dementia risk in women, but not in men.

Table 1

Sample characteristics of participants without (modified) mild cognitive impairment and without dementia at baseline

	Overall sample: n = 2,422	Women: n = 1,556	Men: n = 866	χ² test or t-test: p
Sex, n (%)
Female	1556 (64.2)
Male	866 (35.8)
Age, M (SD)	79.63 (3.52)	79.71 (3.49)	79.47 (3.58)	0.105
Education, n (%)				<0.001
Low	1616 (66.7)	1098	518
medium	568 (23.5)	387	181
high	238 (9.8)	71	167
MMSE, M (SD)	27.94 (1.51)	27.92 (1.52)	27.98 (1.50)	0.363
GDS, M (SD)	2.13 (2.25)	2.30 (2.30)	1.84 (2.14)	<0.001
missing, n (%)	3 (0.1)
SCD (2 groups), n (%)				0.026
no SCD	1029 (42.5)	687	342
SCD	1393 (57.5)	869	524
SCD (3 groups), n (%)				<0.001
no SCD	1029 (42.5)	687	342
SCD without worries	1009 (41.7)	594	415
SCD with worries	384 (15.8)	275	109
Incident dementia until FUP9				0.003
no dementia until FUP9	1911 (78.9)	1198	713
dementia until FUP9	503 (20.8)	351	152
missing, n (%)	8 (0.3)
Incident AD until FUP9				0.001
no dementia until FUP9	1911 (78.9)	1198	713
dementia until FUP9	351 (14.5)	254	97
missing, n (%)	160 (6.6)

M, mean; SD, standard deviation; MMSE, Mini-Mental State Examination; GDS, Geriatric Depression Scale; SCD, subjective cognitive decline; FUP, follow-up; AD, Alzheimer’s disease.

As the current state of knowledge on the association between SCD, sex, and subsequent dementia must be described as insufficient and as heterogeneous, we aimed 1) to examine the cross-sectional distribution of SCD in males and females and 2) to investigate the longitudinal prediction of subsequent dementia and AD by SCD and sex. Based on the available literature regarding elderly samples, we expected 1) that SCD is more frequently reported by females and 2) that SCD better predicts the subsequent risk for dementia and AD in females. A higher sensitivity of women to detect subtle pathological changes was discussed to explain the sex-specific association between SCD and dementia [19]. To test this explanation, global cognition and memory scores at baseline in non-SCD and SCD will be compared separately for men and women. We expect that non-SCD women show better cognitive performance compared to SCD women, whereas no SCD-group differences are expected in men. The results of the study are expected to be valuable for the theoretical background of SCD and might also be useful for clinicians, regarding particularly the prognosis for SCD reports in males and females.

MATERIALS AND METHODS

Sample

We present data from a prospective multi-center study (German Study on Ageing, Cognition, and Dementia in Primary Care Patients (AgeCoDe) and Study on needs, health service use, costs, and health-related quality of life in a large sample of oldest-old primary care patients (85+) (AgeQualiDe), which is a continuation (follow-up 7–9) and extension of AgeCoDe). At baseline, participants at least 75 years of age were recruited via general practitioners (GP) in six German cities (Bonn, Düsseldorf, Hamburg, Leipzig, Mannheim, and Munich). Written informed consent was obtained from all participants. The study was conducted in accord with the Helsinki Declaration of 1975. The local ethics committees approved the study. The interviews were conducted in-person by trained research assistants. Follow-up assessments after baseline in 2003/2004 were conducted every 18 months. Follow-up assessments after follow-up 7 were conducted every 10 months. The present study included data of covariates and predictors assessed at baseline and information of dementia status until follow-up 9. Follow-up 9 was conducted up to 13 years after baseline. The sample initially consisted of 3,327 baseline participants. Participants with a diagnosis of dementia or those fulfilling modified criteria for MCI at baseline were excluded (resulting in n = 2,422). Sex was documented at baseline as female or male. Sample characteristics are given in Table 1.

Covariates

Age, education, global cognitive performance, and depressive symptoms were included as covariates in the regression models. Age at baseline was assessed in years. Education was classified as low, medium, or high according to the revised version of the international new CASMIN educational classification [20]. Global cognitive performance was assessed with the Mini-Mental State Examination (MMSE) [21]. Memory performance was assessed using the Consortium to Establish a Registry for Alzheimer's Disease (CERAD) delayed recall score that measures the number of words from the ten word list recalled after ten minutes. The sum score of the German short version of the Geriatric Depression Scale (GDS) [22, 23] was used to assess depressive symptoms. An alternative GDS score was computed by subtracting the memory-related item for additional analyses to control for a potential overlap with SCD (results not shown).

Subjective cognitive decline, modified mild cognitive impairment, and dementia diagnoses

SCD was assessed using the standardized answers to the question: “Do you feel like your memory is becoming worse?”. Participants could answer “no” or “yes”. If the question was affirmed, participants were additionally asked for the presence or absence of worries related to SCD and could state “yes, but this does not worry me” or “yes, this worries me”. SCD variables with two (no versus yes) or three levels (no; yes, without worries; yes, with worries) were built. The CERAD test battery and the structured interview for the diagnosis of dementia of the Alzheimer type, multi-infarct (or vascular) dementia and dementias of other etiology according to DSM-III-R, DSM-IV, and ICD-10 (SIDAM) [24] were used to assess cognitive performance. Criteria for modified MCI were: 1) absence of dementia, 2) preserved basic activities of daily living or only minimal impairment in complex instrumental functions, 3) evidence for objective cognitive impairment (= test performance more than one standard deviation (SD) below age-/education-specific norms) in at least one (amnestic or nonamnestic) SIDAM subscale (memory, orientation, intellectual abilities, verbal abilities, calculation, constructional abilities, language). The criteria differ from MCI consensus criteria [25] as SCD was not included as criterion for MCI. Dementia diagnoses were based on the SIDAM [24] implemented by trained research assistants. If SIDAM could not be assessed, a Global Deterioration Scale [26] score of at least 4 and/or a Blessed Dementia Rating Scale [27] score were used. Incident dementia cases (including their etiology) were validated by geriatric experts. AD was diagnosed according to DSM-IV criteria [28]. Outcome variables in the present study were all-cause dementia (i.e., AD, vascular dementia, mixed forms, specific types, and dementia not otherwise specified) and AD (also including mixed AD forms) diagnosed from follow-up 1 until follow-up 9.

Statistical analyses

Participants with dementia and those fulfilling modified criteria for MCI at baseline were excluded from the analyses to capture cases of pre-MCI SCD. In cross-sectional analyses, sex was used to predict SCD at baseline in two and three groups in binary and multinomial logistic regression models adjusted for the covariates age, education, MMSE, and depressive symptoms. Analyses of variance (ANOVA) were conducted to examine differences of non-SCD and SCD groups regarding the average delayed recall and the MMSE score using the Scheffé post hoc criterion for significance separately for women and men. Analyses of covariance (ANCOVA) were run additionally controlling for age, education, and depressive symptoms.

In longitudinal analyses, baseline variables were used to predict the transition to all-cause dementia or AD until follow-up 9 as outcome variable in Cox proportional hazards models adjusting for covariates in the following three steps. The covariates age, education, MMSE, and depressive symptoms were entered in step 1, sex (female = reference) and SCD (separate models for two and three groups) were entered in step 2, and the interaction term of SCD×sex was entered in step 3. Dummy variables were generated for the SCD variable in three groups. Only SCD without worries and SCD with worries and their respective interaction terms with sex were entered in these models. Longitudinal analyses stratified by sex were also conducted afterwards separately for women and men.

Level of significance was set to α< 0.05. The analyses were performed using SPSS version 24 for Windows.

Table 2

Prediction of subsequent all-cause dementia until follow-up 9 by covariates (age, education, global cognition, and depressive symptoms) and predictors (sex, subjective cognitive decline in two groups, interaction term sex×subjective cognitive decline in two groups) all assessed at baseline in Cox proportional hazards model

	HR	95% CI	p
Age	1.12	1.09–1.14	<0.001
Education			0.284
Education: medium	0.86	0.69–1.08	0.202
Education: high	1.11	0.81–1.52	0.525
MMSE	0.80	0.75–0.85	<0.001
GDS	1.07	1.03–1.11	0.001
Sex: male versus female	1.25	0.90–1.74	0.184
SCD (2 groups): yes versus no	1.76	1.40–2.21	<0.001
Sex×SCD (2 groups)	0.63	0.42–0.94	0.022

HR, hazard ratio; CI, confidence interval; MMSE, Mini-Mental State Examination; GDS, Geriatric Depression Scale; SCD, subjective cognitive decline; interaction term = sex (0 = female, 1 = male) multiplied by SCD in two groups (0 = no SCD, 1 = SCD).

RESULTS

1,029 participants at baseline reported no SCD (women: n = 687, men: n = 342). 1,393 participants reported SCD (women: n = 869, men: n = 524). Of these 1,393 participants, n = 1,009 reported SCD without worries (women: n = 594, men: n = 415) and n = 384 reported SCD with worries (women: n = 275, men: n = 109). In cross-sectional analyses at baseline, men had an increased risk to report dichotomized SCD in crude (OR = 1.21, 95% CI = 1.02–1.43, p = 0.026; n = 2,422) and (age-, education-, MMSE-, and GDS-) adjusted (OR = 1.22, 95% CI = 1.02–1.45, p = 0.029; n = 2,419) binary logistic regression models. When SCD was entered as three-level criterion variable in multinomial logistic regression models, women had a significantly lower risk to report SCD without worries (crude: OR = 0.71, 95% CI = 0.60–0.85, p < 0.001; adjusted: OR = 0.73, 95% CI = 0.60–0.87, p = 0.001) and tended to have a higher risk to report SCD with worries (crude: n = 2,422, OR = 1.26, 95% CI = 0.97–1.62, p = 0.082; adjusted: n = 2,419, OR = 1.23, 95% CI = 0.94–1.62, p = 0.131) compared to men.

ANOVA results showed that MMSE did not differ between the three SCD groups in women (F(2,1553) = 0.92, p = 0.398), but significant differences regarding delayed recall scores existed in women (F(2,1547) = 13.47, p < 0.001). Post hoc analyses using the Scheffé post hoc criterion for significance indicated that the mean delayed recall score was significantly higher in non-SCD females (M = 6.26, SD = 2.07) compared to both female SCD groups without (M = 5.85, SD = 2.15; p = 0.003) and with worries (M = 5.51, SD = 2.23; p < 0.001). As hypothesized, no significant group differences were found between the SCD groups in men regarding the scores of MMSE (F(2,863) = 0.46, p = 0.635) and delayed recall (F(2,857) = 1.28, p = 0.280). One-way ANCOVAs (controlling for age, education, and depressive symptoms) confirmed these results.

In longitudinal analyses, Cox proportional hazards models were used to predict transition to all-cause dementia and AD. After adjusting for covariates, the presence of SCD (in two groups: HR = 1.76, 95% CI = 1.40–2.21) was associated with an increased risk for subsequent all-cause dementia and the interaction term sex×SCD (HR = 0.63, 95% CI = 0.42–0.94) was associated with a decreased risk for subsequent all-cause dementia (see Table 2; n = 2,392). The interaction coefficient indicated that men with SCD had a lower risk for subsequent dementia compared to women with SCD. Although the direction of the results for AD was similar, the interaction term was no longer a significant predictor (sex×SCD interaction: HR = 0.70, 95% CI = 0.42–1.16, p = 0.170; n = 2,221).

Stratified analyses showed that the presence of SCD (in two groups) was associated with an increased risk for subsequent dementia in women (HR = 1.77, 95% CI = 1.40–2.23), but not in men (HR = 1.07, 95% CI = 0.77–1.51; see Table 3). Similar results were found for the presence of SCD and the risk for subsequent AD (women: n = 1,428, HR = 1.85, 95% CI = 1.41–2.44, p < 0.001; men: n = 786, HR = 1.31, 95% CI = 0.85–2.02, p = 0.222).

Table 3

Prediction of subsequent all-cause dementia until follow-up 9 by covariates (age, education, global cognition, and depressive symptoms) and subjective cognitive decline in two groups all assessed at baseline in Cox proportional hazards models separately for women and men

	Women (n = 1,536)		Men (n = 853)
	HR (95% CI)	p	HR (95% CI)	p
Age	1.12 (1.09–1.16)	<0.001	1.11 (1.06–1.16)	<0.001
Education		0.412		0.309
Education: medium	0.84 (0.64–1.09)	0.183	0.93 (0.60–1.43)	0.733
Education: high	0.95 (0.57–1.57)	0.828	1.33 (0.87–2.05)	0.188
MMSE	0.83 (0.77–0.89)	<0.001	0.74 (0.67–0.82)	<0.001
GDS	1.06 (1.01–1.11)	0.018	1.11 (1.03–1.19)	0.008
SCD (2 groups): yes versus no	1.77 (1.40–2.23)	<0.001	1.07 (0.77–1.51)	0.682

HR, hazard ratio; CI, confidence interval; MMSE, Mini-Mental State Examination; GDS, Geriatric Depression Scale; SCD, subjective cognitive decline.

Table 4

Prediction of subsequent all-cause dementia until follow-up 9 by covariates (age, education, global cognition, and depressive symptoms) and predictors (sex, subjective cognitive decline in three groups, interaction term sex×subjective cognitive decline in three groups) all assessed at baseline in Cox proportional hazards model

	HR	95% CI	p
Age	1.12	1.09–1.15	<0.001
Education			0.328
Education: medium	0.87	0.70–1.09	0.225
Education: high	1.10	0.80–1.51	0.562
MMSE	0.79	0.75–0.84	<0.001
GDS	1.05	1.01–1.09	0.020
Sex: male versus female	1.23	0.89–1.71	0.217
SCD 3 groups (dummy)
no SCD (dummy)	–	–	–
SCD without worries	1.48	1.15–1.91	0.002
SCD with worries	2.48	1.87–3.29	<0.001
Sex×SCD (dummy)
Sex×no SCD (dummy)	–	–	–
Sex×SCD without worries	0.68	0.44–1.04	0.078
Sex×SCD with worries	0.63	0.37–1.08	0.095

HR, hazard ratio; CI, confidence interval; MMSE, Mini-Mental State Examination; GDS, Geriatric Depression Scale; SCD, subjective cognitive decline; interaction terms: sex×SCD without worries = sex (0 = female, 1 = male) multiplied by SCD without worries (1 = SCD without worries, 0 = else), sex×SCD with worries = sex (0 = female, 1 = male) multiplied by SCD with worries (1 = SCD with worries, 0 = else).

The prediction of dementia by three levels of SCD was also examined. In the adjusted model, an association with an increased risk for all-cause dementia was found for SCD without worries (HR = 1.48, 95% CI = 1.15–1.91), SCD with worries (HR = 2.48, 95% CI = 1.87–3.29), and trends were found for the interaction terms sex×SCD without worries (HR = 0.68, 95% CI = 0.44–1.04, p = 0.078) and sex×SCD with worries (HR = 0.63, 95% CI = 0.37–1.08, p = 0.095, n = 2,392; see Table 4). Coefficients in this analysis also indicated that men with SCD without or with worries had a marginally reduced risk for subsequent dementia compared to women without or with SCD. Results for AD pointed towards the same direction, but the interaction terms did not approach significance in these models (n = 2,221; SCD without worries: HR = 1.63, 95% CI = 1.22–2.19, p = 0.001; SCD with worries: HR = 2.44, 95% CI = 1.73–3.44, p < 0.001; sex×SCD without worries interaction: HR = 0.72, 95% CI = 0.42–1.23, p = 0.232; sex×SCD with worries interaction: HR = 0.77, 95% CI = 0.39–1.50, p = 0.440).

Stratified analyses showed that the presence of SCD without and with worries was associated with an increased risk for subsequent dementia in women (SCD without worries: HR = 1.49, 95% CI = 1.16–1.92; SCD with worries: HR = 2.53, 95% CI = 1.90–3.36), but not in men (SCD without worries: HR = 0.98, 95% CI = 0.69–1.41; SCD with worries: HR = 1.46, 95% CI = 0.90–2.35; see Table 5). Similar results were found for the presence of SCD in three groups and the risk for subsequent AD, except that SCD with worries also predicted AD in men (women: n = 1,428; SCD without worries: HR = 1.65, 95% CI = 1.23–2.21, p = 0.001; SCD with worries: HR = 2.43, 95% CI = 1.72–3.44, p < 0.001; men: n = 786; SCD without worries: HR = 1.18, 95% CI = 0.75–1.87, p = 0.475; SCD with worries: HR = 1.89, 95% CI = 1.04–3.42, p = 0.037).

Table 5

Prediction of subsequent all-cause dementia until follow-up 9 by covariates (age, education, global cognition, and depressive symptoms) and subjective cognitive decline in three groups all assessed at baseline in Cox proportional hazards models separately for women and men

	Women (n = 1,536)		Men (n = 853)
	HR (95% CI)	p	HR (95% CI)	p
Age	1.12 (1.09–1.16)	<0.001	1.11 (1.06–1.16)	<0.001
Education		0.486		0.310
Education: medium	0.85 (0.66–1.11)	0.232	0.91 (0.59–1.40)	0.675
Education: high	0.92 (0.56–1.53)	0.756	1.32 (0.86–2.03)	0.204
MMSE	0.82 (0.76–0.88)	<0.001	0.75 (0.67–0.83)	<0.001
GDS	1.03 (0.99–1.08)	0.173	1.09 (1.01–1.18)	0.020
SCD (3 groups)		<0.001		0.208
SCD without worries	1.49 (1.16–1.92)	0.002	0.98 (0.69–1.41)	0.922
SCD with worries	2.53 (1.90–3.36)	<0.001	1.46 (0.90–2.35)	0.122

HR, hazard ratio; CI, confidence interval; MMSE, Mini-Mental State Examination; GDS, Geriatric Depression Scale; SCD, subjective cognitive decline.

DISCUSSION

Previous research with clinical and epidemiological samples has firmly established that SCD is a valid predictor of incident dementia. Whether sex moderates this association has rarely ever been addressed so far. Here we show, in an elderly (75+) GP-based cohort without objective cognitive impairment at baseline, that the report of women on their subjectively perceived cognitive decline is more strongly related to incident dementia than the report of men. SCD without and with worries in women both were associated with an about 1.5 to 2.5-fold increased risk for subsequent dementia, whereas in men only SCD with worries was significantly associated with the risk for subsequent AD, but not for subsequent all-cause dementia. However, the absence of an association of most SCD definitions with dementia in men might also be due to the reduced overall sample size and reduced incident cases in this gender group and could be detected in larger samples.

The stronger association between SCD and subsequent dementia in women might be explained by a better ability to perceive subtle pathological changes [19]. A stronger tendency of women to report cognitive problems was discussed by Sundermann et al. [29]. In line with the hypothesis of a higher sensitivity to perceive subtle pathological changes in females, women without SCD had significantly higher delayed recall scores than women with SCD, whereas delayed recall scores of the three SCD groups did not differ in men in the present study. No SCD group differences were found for the MMSE score. However, as the sample was restricted to cognitively healthy participants at baseline, it might be not surprising that female SCD group differences were found specifically for memory, but not for global cognition. Sundermann et al. [29] also found a stronger association between SCD and memory performance in women than in men in the ADNI study (mean age = 72.8 years), but the sex-specific results were driven by the MCI group and did not exist in the normal control group, opposite to our study results.

Other, not sex-specific, studies found decreasing predictive power of SCD with increasing cognitive impairment [30] that might be accompanied by reduced awareness or anosognosia regarding the own cognitive decline [31]. Therefore, our sex-specific results have theoretical implication for the pre-MCI SCD construct, and might also support clinical decision making in situations where evidence for the absence of objective mild cognitive impairment is available. In the absence of this information on MCI, the presence of SCD per se regardless of sex remains a risk-increasing factor. Future studies should investigate the sex-specific association between pre-MCI SCD and subsequent dementia also in clinical settings such as memory clinics, where MCI status can be identified. Additionally, more research on sex-specific associations in the context of SCD should be initiated, which might also affect the design of clinical trials or future clinical practice. Our results can provide a first contribution to the generation of sex-specific disease models for sex-sensitive prevention, diagnosis, prognosis, and personalized treatment in the future as recently proposed for AD by Ferretti et al. [32].

The cross-sectional finding that men had a higher risk for SCD at baseline was not expected. The result was driven by a higher risk of men to report SCD without worries, but tending to have a lower risk for SCD with worries compared to women, which again is in accordance with our hypothesis. Other authors found stronger associations between subjective and objective memory in men and discussed that higher levels of educational and occupational experiences might have provided more opportunities for inter-individual comparisons in males resulting in more appropriate cognitive self-assessments [33, 34]. As these results are not in accordance with our results regarding the delayed recall performance, different thresholds or response patterns might exist in women and men. Men might affirm SCD without worries earlier due to realistic self-evaluations, as cognitive and memory changes can be considered as normal to some extent in the oldest old without necessarily being an early sign of dementia, whereas SCD (also without worries) in women might capture pathological validity earlier. In accordance with this, men had a higher risk to report SCD in general and SCD without worries at baseline. A higher level of memory-specific cognitive reserve in women might exist [29], which possibly could also affect SCD thresholds and response patterns (e.g., earlier affirmation in men, but stronger pathological validity in women).

Our study has several strengths. Beside a large sample of very old participants, we provided data on a long follow-up period of up to 13 years. Our study fulfilled SCD plus features proposed by the SCD-I working group [2] such as assessing subjective decline in memory and the presence of associated worries. We used a rather strict definition of objective cognitive impairment (1 SD instead of 1.5 SD below norm). Additional analyses were conducted to support our findings (data not shown). The exclusion of the memory-related item of the GDS provided similar results. Due to the fact that we included a rather long follow-up period of up to 13 years, we also entered transition to all-cause dementia or AD until follow-up 3 (after 4.5 years) and until follow-up 6 (after 9 years) as criterion variables in additional analyses. The results for these shorter follow-ups regarding the interaction of sex and SCD (two or three groups) were similar or slightly more pronounced, indicating a robust association between SCD and subsequent dementia in women for short-, middle-, and long-term periods of follow-up as also found by Pérès et al. [19] in a younger and cognitively more impaired sample.

Beside these strengths, there are also limitations of our study. Our measure of SCD and the results that we obtained might be specific to memory. Therefore, results might be different if other cognitive domains of SCD would have been assessed. Biomarkers of dementia pathology were not part of the study design. Hence, although incident dementia diagnoses were validated by geriatric experts, some instances of etiological misclassification might have occurred. Although results for all-cause dementia and AD were rather similar, the lower number of AD cases might account for slight discrepancies. Some SCD plus features such as age of SCD onset were not included in the present study. The higher life expectancy of women might have affected our results and an earlier death before a clinical manifestation of dementia might have attenuated the association between SCD and dementia in men. However, an association between SCD with worries and AD was found in men.

To summarize, in a sample without initial cognitive impairment at baseline, men reported more SCD without worries than women. Over a longitudinal observation period of up to 13 years, baseline SCD in women was more strongly related to incident dementia than the report of men. SCD without and with worries predicted subsequent dementia and AD in women, whereas only SCD with worries predicted an increased risk for AD in men, but not for all-cause dementia. Our results might contribute to the assessment of dementia risk and to the clinical prognosis of SCD and dementia. For example, both SCD without and with worries in women appear to be relevant for the subsequent dementia risk, whereas only SCD with worries in men might be associated with an increased risk specifically for AD. Future studies should be conducted to examine sex-specific associations between SCD and dementia in clinical samples.

Footnotes

ACKNOWLEDGMENTS

We want to thank both all participating patients and their general practitioners for their good collaboration.

Members of the AgeCoDe & AgeQualiDe Study Group: Wolfgang Maier (Principal Investigator), Martin Scherer (Principal Investigator), Steffi G. Riedel-Heller (Principal Investigator), Heinz-Harald Abholz, Christian Brettschneider, Cadja Bachmann, Horst Bickel, Wolfgang Blank, Sandra Eifflaender-Gorfer, Marion Eisele, Annette Ernst, Angela Fuchs, André Hajek, Kathrin Heser, Frank Jessen, Hanna Kaduszkiewicz, Teresa Kaufeler, Mirjam Köhler, Hans-Helmut König, Alexander Koppara, Diana Lubisch, Tobias Luck, Dagmar Lühmann, Melanie Luppa, Tina Mallon, Manfred Mayer, Edelgard Mösch, Michael Pentzek, Jana Prokein, Alfredo Ramirez, Susanne Roehr, Anna Schumacher, Janine Stein, Susanne Steinmann, Franziska Tebarth, Hendrik van den Bussche (Principal Investigator 2002-2011), Carolin van der Leeden, Michael Wagner, Klaus Weckbecker, Dagmar Weeg, Jochen Werle, Siegfried Weyerer, Birgitt Wiese, Steffen Wolfsgruber, Thomas Zimmermann.

This publication is part of the German Research Network on Dementia (KND), the German Research Network on Degenerative Dementia (KNDD; German Study on Ageing, Cognition and Dementia in Primary Care Patients; AgeCoDe), and the Health Service Research Initiative (Study on Needs, health service use, costs and health-related quality of life in a large sample of oldest-old primary care patients (85+; AgeQualiDe)) and was funded by the German Federal Ministry of Education and Research (grants KND: 01GI0102, 01GI0420, 01GI0422, 01GI0423, 01GI0429, 01GI0431, 01GI0433, 01GI0434; grants KNDD: 01GI0710, 01GI0711, 01GI0712, 01GI0713, 01GI0714, 01GI0715, 01GI0716; grants Health Service Research Initiative: 01GY1322A, 01GY1322B, 01GY1322C, 01GY1322D, 01GY1322E, 01GY1322F, 01GY1322G).

The publication was also supported by the study “Healthy Aging: Gender specific trajectories into latest life” (AgeDifferent.De) that was funded by the German Federal Ministry of Education and Research (Funding program “Gesund – ein Leben lang”, grants 01GL1714A; 01GL1714B; 01GL1714C; 01GL1714D).

Authors’ disclosures available online ().

References

Mitchell

, Beaumont

, Ferguson

, Yadegarfar

, Stubbs

(2014) Risk of dementia and mild cognitive impairment in older people with subjective memory complaints: Meta-analysis. Acta Psychiatr Scand 130, 439–451.

Jessen

, Amariglio

, van Boxtel

, Breteler

, Cec-caldi

, Chetelat

, Dubois

, Dufouil

, Ellis

, van der Flier

, Glodzik

, van Harten

, de Leon

, McHugh

, Mielke

, Molinuevo

, Mosconi

, Osorio

, Perrotin

, Petersen

, Rabin

, Rami

, Reisberg

, Rentz

, Sachdev

, de la Sayette

, Saykin

, Scheltens

, Shulman

, Slavin

, Sperling

, Stewart

, Uspenskaya

, Vellas

, Visser

, Wagner

(2014) A conceptual framework for research on subjective cognitive decline in preclinical Alzheimer’s disease. Alzheimers Dement 10, 844–852.

Jack

Jr , Bennett

, Blennow

, Carrillo

, Dunn

, Haeberlein

, Holtzman

, Jagust

, Jessen

, Karlaw-ish

, Liu

, Molinuevo

, Montine

, Phelps

, Rankin

, Rowe

, Scheltens

, Siemers

, Snyder

, Sperling

(2018) NIA-AA Research Framework: Toward a biological definition of Alzheimer’s disease. Alzheimers Dement 14, 535–562.

Reid

, Maclullich

(2006) Subjective memory complaints and cognitive impairment in older people. Dement Geriatr Cogn Disord 22, 471–485.

Cooper

, Bebbington

, Lindesay

, Meltzer

, McManus

, Jenkins

, Livingston

(2011) The meaning of reporting forgetfulness: A cross-sectional study of adults in the English 2007 Adult Psychiatric Morbidity Survey. Age Ageing 40,711–717.

Jessen

, Wiese

, Bachmann

, Eifflaender-Gorfer

, Haller

, Kolsch

, Luck

, Mtjsch

, van den Bussche

, Wagner

, Wollny

, Zimmermann

, Pentzek

, Riedel-Heller

, Romberg

, Weyerer

, Kaduszkiewicz

, Maier

, Bickel

(2010) Prediction of dementia by subjective memory impairment: Effects of severity and temporalassociation with cognitive impairment. Arch Gen Psychiatry 67, 414–422.

Molinuevo

, Rabin

, Amariglio

, Buckley

, Dubois

, Ellis

, Ewers

, Hampel

, Kloppel

, Rami

, Reisberg

, Saykin

, Sikkes

, Smart

, Snitz

, Sperling

, van der Flier

, Wagner

, Jessen

(2017) Implementation of subjective cognitive decline criteria in research studies. Alzheimers Dement 13, 296–311.

Bassett

, Folstein

(1993) Memory complaint, memory performance, and psychiatric diagnosis: A community study. J Geriatr Psychiatry Neurol 6, 105–111.

Slavin

, Brodaty

, Kochan

, Crawford

, Trollor

, Draper

, Sachdev

(2010) Prevalence and predictors of “subjective cognitive complaints” in the Sydney Memory and Ageing Study. Am J Geriatr Psychiatry 18, 701–710.

10.

Kielb

, Rogalski

, Weintraub

, Rademaker

(2017) Objective features of subjective cognitive decline in a United States national database. Alzheimers Dement 13, 1337–1344.

11.

Paradise

, Glozier

, Naismith

, Davenport

, Hickie

(2011) Subjective memory complaints, vascular risk factors and psychological distress in the middle-aged: A cross-sectional study. BMC Psychiatry 11, 108.

12.

Wang

, van Belle

, Crane

, Kukull

, Bowen

, McCormick

, Larson

(2004) Subjective memory deterioration and future dementia in people aged 65 and older. J Am Geriatr Soc 52, 2045–2051.

13.

Holmen

, Langballe

, Midthjell

, Holmen

, Fik-seaunet

, Saltvedt

, Tambs

(2013) Gender differences in subjective memory impairment in a general population: The HUNT study, Norway. BMC Psychol 1, 19.

14.

Jonker

, Geerlings

, Schmand

(2000) Are memory complaints predictive for dementia? A review of clinical and population-based studies. Int J Geriatr Psychiatry 15, 983–991.

15.

Mol

, van Boxtel

, Willems

, Jolles

(2006) Do subjective memory complaints predict cognitive dysfunction over time? A six-year follow-up of the Maastricht Aging Study. Int J Geriatr Psychiatry 21, 432–441.

16.

Genziani

, Stewart

, Bejot

, Amieva

, Artero

, Ritchie

(2013) Subjective memory impairment, objective cognitive functioning and social activity in French older people: Findings from the Three Cities study. Geriatr Gerontol Int 13, 139–145.

17.

Tomita

, Sugawara

, Kaneda

, Okubo

, Iwane

, Taka-hashi

, Kaneko

, Yasui-Furukori

(2014) Sex-specific effects of subjective memory complaints with respect to cognitive impairment or depressive symptoms. Psychiatry Clin Neurosci 68, 176–181.

18.

Rabin

, Smart

, Crane

, Amariglio

, Berman

, Boada

, Buckley

, Chetelat

, Dubois

, Ellis

, Gifford

, Jefferson

, Jessen

, Katz

, Lip-ton

, Luck

, Maruff

, Mielke

, Molinuevo

, Naeem

, Perrotin

, Petersen

, Rami

, Reisberg

, Rentz

, Riedel-Heller

, Risacher

, Rodriguez

, Sachdev

, Saykin

, Slavin

, Snitz

, Sperling

, Tandetnik

, van der Flier

, Wagner

, Wolfsgruber

, Sikkes

(2015) Subjective cognitive decline in older adults: An overview of self-report measures used across 19 international research studies. J Alzheimers Dis 48(Suppl 1), S63–86.

19.

Pèrés

, Helmer

, Amieva

, Matharan

, Carcail-lon

, Jacqmin-Gadda

, Auriacombe

, Orgogozo

, Barberger-Gateau

, Dartigues

(2011) Gender differences in the prodromal signs of dementia: Memory complaint and IADL-restriction. A prospective population-based cohort. J Alzheimers Dis 27, 39–47.

20.

Brauns

, Steinmann

(1999) Educational reform in France, West-Germany and the United Kingdom. Updating the CASMIN educational classification. ZUMA-Nachrichten 44, 7–44.

21.

Folstein

, Folstein

, McHugh

(1975) “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12, 189–198.

22.

Sheikh

, Yesavage

(1986) Geriatric Depression Scale (GDS): Recent evidence and development of a shorter version. In Clinical Gerontology: A guide to Assessment and Intervention, Brink TL, ed. Haworth, New York, pp. 165–173.

23.

Gauggel

, Birkner

(1999) Validity and reliability of a German version of the Geriatric Depression Scale (GDS) [in German]. ZKlin Psychol Psychother 28, 18–27.

24.

Zaudig

, Hiller

(1996) SIDAM - A structured interview for the diagnosis of dementia of the Alzheimer type, multi-infarct (or vascular) dementia and dementias of other aetiology according to DSM-III-R, DSM-IV and ICD-10 (SIDAM-manual) [in German], Huber, Bern.

25.

Winblad

, Palmer

, Kivipelto

, Jelic

, Fratiglioni

, Wahlund

, Nordberg

, Backman

, Albert

, Almkvist

, Arai

, Basun

, Blennow

, de Leon

, DeCarli

, Erkinjuntti

, Giacobini

, Graff

, Hardy

, Jack

, Jorm

, Ritchie

, van Duijn

, Visser

, Petersen

(2004) Mild cognitive impairment - beyond controversies, towards a consensus: Report of the International Working Group on Mild Cognitive Impairment. J Intern Med 256, 240–246.

26.

Reisberg

, Ferris

, de Leon

, Crook

(1982) The Global Deterioration Scale for assessment of primary degenerative dementia. Am J Psychiatry 139, 1136–1139.

27.

Blessed

, Tomlinson

, Roth

(1968) The association between quantitative measures of dementia and of senile change in the cerebral grey matter of elderly subjects. Br J Psychiatry 114, 797–811.

28.

APA (1994) DSM-IV: Diagnostic and Statistical Manual of Mental Disorders, 4th edition, American Psychological Association, Washington, DC.

29.

Sundermann

, Edmonds

, Delano-Wood

, Galasko

, Salmon

, Rubin

, Bondi

(2018) Sex influences the accuracy of subjective memory complaint reporting in older adults. J Alzheimers Dis 61, 1163–1178.

30.

Wolfsgruber

, Wagner

, Schmidtke

, Frölich

, Kurz

, Schulz

, Hampel

, Heuser

, Peters

, Reischies

, Jahn

, Luckhaus

, Hull

, Gertz

, Schröder

, Pan-tel

, Rienhoff

, Ruuther

, Henn

, Wiltfang

, Maier

, Kornhuber

, Jessen

(2014) Memory concerns, memory performance and risk of dementia in patients with mild cognitive impairment. PLoS One 9, e100812.

31.

Roberts

, Clare

, Woods

(2009) Subjective memory complaints and awareness of memory functioning in mild cognitive impairment: A systematic review. Dement Geriatr Cogn Disord 28, 95–109.

32.

Ferretti

, Iulita

, Cavedo

, Chiesa

, Schumacher

Dimech A

, Santuccione

Chadha A

, Baracchi

, Girouard

, Misoch

, Giacobini

, Depypere

, Hampel

(2018) Sex differences in Alzheimer disease - the gateway to precision medicine. Nat Rev Neurol 14, 457–469.

33.

Hülür

, Hertzog

, Pearman

, Ram

, Gerstorf

(2014) Longitudinal associations of subjective memory with memory performance and depressive symptoms: Between-person and within-person perspectives. Psychol Aging 29, 814–827.

34.

Hülür

, Hertzog

, Pearman

, Gerstorf

(2015) Correlates and moderators of change in subjective memory and memory performance: Findings from the Health and Retirement Study. Gerontology 61, 232–240.