Abstract
Objectives:
This study investigates the impact of occupation-based motivational processes and social network variables on the incidence of dementia over 8 years.
Method:
Data were derived from the Leipzig Longitudinal Study of the Aged (LEILA75+), a population-based longitudinal study of individuals aged 75 years and older (n = 1692 at baseline). Motivational processes were estimated based on the main occupation using the Occupational Information Network database.
Results:
In a Cox proportional hazard model, motivational processes were not associated with the risk of dementia (hazard ratio [HR]: 0.93, 95% confidence interval [CI]: 0.74-1.16). Individuals with a higher frequency of social contact at baseline had a significantly lower risk of dementia (HR: 0.96, 95% CI: 0.91-0.99), while proximity of social contacts was not linked to the risk of dementia (HR: 1.03, 95% CI: 0.98-1.08). In individuals with low indices of motivational processes, the frequency of social contacts was associated with a lower risk of dementia (HR: 0.94, 95% CI: 0.88-1.00). On the other hand, proximity of social contacts was linked to a higher risk of dementia in individuals with high indices of motivational processes (HR: 1.09, 95% CI: 1.01-1.19).
Discussion:
Results indicate that the frequency and proximity of social contacts have a differential impact on the risk of dementia according to lower or higher indices of motivational processes, while the impact of motivational processes on risk of dementia could not be confirmed. Future studies should carefully disentangle different aspects of social interactions and their association with motivational processes.
Introduction
As predictors of cognitive impairment and dementia, motivational processes are rooted in the concept of motivational reserve (MR) developed by Forstmeier and Maercker, 1 which enables the brain to tolerate neuropathological age- and dementia-related changes without clinical manifestation. 1 The broader concept of brain reserve includes the concept of a cognitive reserve which has been proposed as an explanation for why Alzheimer disease (AD) diagnosis may be postponed in older adults with higher educational or occupational status. 2 In the Forstmeier and Maercker model, 1 MR and cognitive reserve are complementary concepts. The term “MR” refers to the contribution of lifetime motivational activities to general brain reserve 3,4 together with cognitive, 5 physical, 6 and social activities. 7 Motivational processes play a crucial role in the occupational context. In all occupations, motivational abilities are needed to a certain degree to achieve goals and successfully execute different tasks. 1,8 However, some occupations require more motivational processes, such as goal orientation and action planning, than other occupations. Research on job design and recruitment indicates that individuals with certain levels of (motivational) skills gravitate toward occupations needing these skills, and, conversely, working on a job with certain psychological demands affects one’s abilities. 9 Thus, using a particular skill (eg, linguistic, mathematical, social, or motivational) seems to result in improving the person’s skills in this area. Furthermore, motivation was shown to explain incremental variance in training outcomes beyond the effect of cognitive ability, even when considering the impact of cognitive attributes. 10 In a study exploring the link between occupational engagement and motivation of older adults with serious illnesses, occupational engagement was shown to foster motivation and persistence, 11 underlining the importance of motivation in the occupational context.
Two other studies 12,13 have already used estimations of midlife motivational abilities on the basis of the main occupation using the Occupational Information Network (O*NET) 14 database, providing detailed information on worker abilities. Motivational abilities were found to be associated with a reduced risk of mild cognitive impairment (MCI) and AD in apolipoprotein E ∊4 carriers in old age 12 and to moderate the relationship between cognitive impairment and the presence and progression of apathy. 13
Based on knowledge about motivational processes and the regulation of motivation, 15 motivational abilities can be described as skills that help individuals to effectively implement their goals, including goal orientation, action planning, decision regulation, activation regulation, motivation regulation, and self-efficacy. Higher developed motivational processes have been shown to act as protective factors against the manifestation of cognitive impairment. 1,12 Previous studies have primarily addressed the impact of self-efficacy on cognitive functioning, in general, 16 on academic performance 17 and on memory performance in old age. 18 Greater complexity of work may reduce the risk of AD, 19 –21 and education has been shown to protect against cognitive decline 22 and dementia. 23
Other relevant predictors related to the risk of developing dementia are social resources. People with a moderate or rich social network have a lower odds ratio of dementia than do those with a poor or limited social network. 3 Living alone without any close social ties increases the risk of developing dementia. 7 Likewise, Fratiglioni et al 24 have found that an active and socially integrated lifestyle may protect against dementia. Stimulating activities—be they mental, physical, or social—may thus protect against dementia, 25 and maintaining social relations in old age seems to be beneficial for maintaining cognitive functioning. However, the pathways through which these beneficial effects occur are not yet entirely clear.
Motivational Processes and Social Parameters
In a variety of studies, motivational variables were shown to remain significant factors even when considering the impact of social resources on health outcomes. 26 –28 Particularly, self-efficacy and related constructs such as perceived control 28,29 have been demonstrated to be crucial. According to the support-efficacy model from Antonucci and Jackson, 30 an individual’s general self-efficacy mediates the effects of social outcomes on health and well-being. There is evidence that self-efficacy is one of the psychosocial pathways through which social support operates. 26,27 Bisconti and Bergeman 28 suggested that increasing a person’s sense of control in social relationships may actually be the underlying factor that produces better psychological and physical outcomes in old age. Apart from self-efficacy and related concepts, motivational processes have not yet been considered in studies analyzing the impact of social network variables on health outcomes.
Few studies have examined the role of motivational abilities in cognitive functioning and dementia. 1,12 The impact of perceived control and social support on psychological well-being, 28,29 the impact of self-efficacy and social support on health 30 and major depression, 31 and the impact of social and motivational variables on adjustment disorder have been investigated. 32 However, to the best of our knowledge, no previous study has explored the impact of motivational and social resources on the risk of cognitive impairment and dementia.
Study Objectives and Goals
This study investigates the impact of occupation-based motivational processes and social network variables on the risk of dementia in the long term. Individuals with dementia were expected to have lower indices of motivational processes than healthy individuals, and occupation-based motivational processes and social network variables were expected to be associated with a lower risk of dementia. Based on the studies described earlier, we anticipated occupation-based motivational processes and social network variables to be associated. Occupation-based motivational processes were predicted to moderate the influence of social network variables on the risk of dementia, with higher indices of motivational processes leading to a less pronounced impact of social network variables on the risk of dementia.
Methods
Sample
This study is based on data from the Leipzig Longitudinal Study of the Aged (LEILA75+), an epidemiological study of dementia and MCI that was conducted from 1997 to 2005 in Leipzig, Germany, with a population-based sample of 1692 elderly participants at the baseline assessment. 33 Overall, the study covered 1 baseline and 5 follow-up assessments, on average, every 1.4 years. Study participants (n = 1500) were identified by systematic random sampling from an age-ordered list provided by the local registry office. In addition, institutionalized participants were included by proportion (n = 192).
Information on sample attrition is presented in Figure 1. Of the total sample of 1692 participants at baseline, clinical interviews, including neuropsychological assessment, could be conducted in 1265 (74.8%) participants. Of these 1265 participants, 220 were already classified as having dementia at baseline and thus excluded from the analyses. At follow-up 1, 70 of the remaining 1045 individuals included for analysis were only tested by proxy interviews, 66 refused to take part in the study, 108 died, 9 could not be located, and 20 had either incomplete assessments or did not participate because of other reasons. Thus, the remaining 772 participants included in further analysis were tested by neuropsychological assessment and clinical interview both at baseline and at follow-up 1. The mean age of the participants included in the analysis was 81.21 years (standard deviation [SD] = 4.67; range: 75.11-99.29; Table 1), and 74% of the participants were female. Based on the revised version of the international Comparative Analysis of Social Mobility in Industrial Nations (CASMIN) educational classification, 34 65% of the participants had a low education, 22% had a middle education, and 13% had a high education. The sample included in the analysis was younger participants compared to the overall sample (M = 82.56; SD = 5.37; range: 75.00-102.16); also, the overall sample included more female participants (75.80%), had less years of education, and lower Mini-Mental State Examination (MMSE) scores (see Table 1). When comparing the sample of participants by their motivational processes, a higher percentage of participants with high indices of motivational processes were male, were significantly younger, had a higher cognitive status indicated by the MMSE, were less depressed, had completed more years of education, and engaged in more frequent social interactions (Table 2). To ensure a reasonably large sample size, participants who participated indirectly by a fully structured proxy interview in follow-ups 2, 3, 4, and 5 were included in the analysis. This was the case in 29% of the participants. A significantly higher percentage of these individuals were male, were older, and had a higher proximity of their social contacts than the individuals participating directly in the study (Table 3); otherwise, there were no significant differences between the 2 groups.
Sample attrition and identification of cases with dementia.
Sample Descriptives Table.
Abbreviation: SD, standard deviation.
Sample Descriptives Table Comparing Individuals With Low Motivational Processes to Individuals With High Motivational Processes.
Abbreviations: SD, standard deviation; χ2, chi-square.
a P < .001.
b P < .05.
c P < .01.
Sample Descriptives Table Comparing Individuals Participating Directly (n = 549) With Individuals Participating by Proxy Interview.a
Abbreviations: SD, standard deviation; χ2, chi-square.
an = 223.
b P < .001.
c P < .05.
Procedure and Data Collection
Trained psychologists and physicians conducted structured clinical interviews and structured proxy interviews. The Structured Interview for the Diagnosis of Dementia of Alzheimer Type, Multi-Infarct Dementia, and Dementias of Other Etiology (SIDAM) 35 was used in the face-to-face interview during visits to participants’ homes. Short proxy interviews supplied additional information on subjective memory impairment and cognitive and psychosocial functioning, if necessary. Baseline interviews were conducted between January 1997 and June 1998. The 5 follow-up assessments were conducted between July 1998 and April 2005 (follow-up 1: 1998-2000; follow-up 2: 1999-2001; follow-up 3: 2000-2003; follow-up 4: 2001-2004; and follow-up 5: 2003-2005).
When it was not possible to directly involve the participant in the follow-up (eg, owing to death or severe weakness), a fully structured proxy interview, including the Clinical Dementia Rating (CDR) Scale, 36 assessing the cognitive functioning of the participant, was conducted.
Materials
Assessment of occupation-based motivational and cognitive processes
Motivational processes were estimated by reference to a sample of O*NET variables on the basis of each participant’s main occupation. The O*NET is the official occupational classification system of the US Department of Labor. 14 It consists of a hierarchically structured lexicon of some 1100 occupations and a large database of associated work and worker characteristics—the result of an ongoing large-scale research project conducted over recent decades. The database includes empirically collected data on the abilities and skills needed in each occupation. In the O*NET data collection program, questionnaires were used to assess samples of workers in each job. Each new version of the O*NET represents an update for these data. Version 12.0, which is used in this study, is based on samples of 20 to 70 incumbents per occupation. Most O*NET variables relating to skills, abilities, and work activities have been shown to have a high interrater reliability and to be valid (high correlations with expert ratings). 14
Participants in LEILA75+ were asked to report their main (longest) occupation. O*NET occupational codes were assigned independently by 2 coders; any coding differences were reconciled in discussion with the last author. Initial interrater agreement was 84% at the highest level of aggregation (2 digits), 79% at the second highest level (3 digits), and 67% at the lowest level of the detailed O*NET occupations (8 digits). Participants who had been housewives for most of their lives were classified on the basis of their second longest job; participants who had been housewives all of their lives were coded as “personal and home care aides.” Typical professions coded as requiring high motivational processes were engineer, teacher, nurse, manager, and lawyer; professions coded as requiring medium motivational processes were mechanician, clerk, and engine fitter; finally, professions coded as requiring low motivational processes were cashier, salesman, unskilled laborer, and waiter.
Two motivational and 4 cognitive O*NET variable values belonging to this O*NET occupation were assigned to the participants. The selection of these variables is described by Forstmeier and Maercker. 1 In short, variables were selected in a sample of nondemented elderly individuals on the basis of (a) their content validity and (b) their correlations with self-reported motivational processes and a measure of crystallized (verbal) intelligence. Two variables were highly significantly associated with self-reported motivational abilities but not with intelligence, that is (1) goal orientation (item 4.A.2.b.6) and (2) action planning (4.A.1.b.3). Four variables were highly significantly correlated with intelligence but not with self-reported motivational abilities, that is (1) selective attention (1.A.1.g.1), (2) recognizing problems (1.A.1.b.3), (3) assessing performance (2.A.2.d), and (4) social perceptiveness (2.B.1.a).
A composite for occupation-based motivational processes was constructed based on the z-standardized scores of goal orientation and action planning. The individual items were first standardized and then averaged. Likewise, a composite for midlife cognitive functioning was constructed based on the z-standardized scores of the 4 cognitive variables. Internal consistency (α) was 0.71 for the O*NET motivational processes total score and 0.78 for the O*NET cognitive functioning total score in the present study. The 2 total scores were used in the following analyses.
Assessment of characteristics of the social network
Participants’ social networks were assessed using the Practitioner Assessment of Network Type (PANT). 37,38 The PANT uses 8 precoded questions to identify different network types. Five support network types were identified in a study by Wenger, 39 based on the distance of local close kin (eg, “How far away, in distance, does your nearest child or other relative live?”), the frequency of interaction within networks (eg, “How often do you see any of your neighbors to have a chat with or do something with?”), and the level of interaction with the community and voluntary groups (eg, “Do you attend meetings of any community/neighborhood or social groups, such as old people’s clubs, lectures, or anything like that?”).
The PANT measure was originally constructed to assess different network types; however, in order to obtain dimensional scales, we calculated 2 scores representing the frequency and the proximity of social contacts. The proximity sum score comprised the items relating to the proximity of relatives, children, and siblings; the frequency sum score comprised the items relating to contact with relatives, neighbors, and friends and attendance of religious, community, and social meetings. The higher the frequency of interaction within the networks, the higher the frequency score; the higher the proximity of local close family, the higher the proximity score. Before calculating the sum scores, we recoded the values such that a high value indicated a richer social network.
Diagnosis of dementia and neuropsychological test battery
Clinical diagnosis of dementia was made according to Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition; DSM-IV) criteria (American Psychiatric Association). 40 The SIDAM diagnostic algorithms based on the DSM-IV diagnostic guidelines were used to discriminate between types of dementia, including dementia of the Alzheimer type, vascular dementia, and other, not further specified types of dementia. 35 Incident dementia cases could be diagnosed only at the predefined times of the follow-up assessments. Consensus conferences of physicians and psychologists were held for each participant. The cognitive criterion of dementia was based either on cognitive testing or, in the case of proxy interviews, on CDR data. 36 The CDR offers a global severity rating of cognitive impairment assessing 5 different stages, including (1) cognitively healthy, (2) questionable dementia, (3) mild dementia, (4) moderate dementia, and (5) severe dementia. 36 Depending on the closeness of the relationship between the caregiver and the participants of the study, information concerning the cognitive impairment may be more or less accurate and validity may differ. However, for those participants rated by their caregiver, a close relationship of proxy and study subjects can be assumed. 41 As the SIDAM-derived diagnosis of dementia (face-to-face interview) was assessed differently from the CDR-derived diagnosis (proxy interview), the agreement between the 2 instruments was investigated in a sample of 180 individuals showing a coefficient of 0.83, 41 indicating a very good agreement.
Neuropsychological functioning was assessed with the main instrument used in this study, the SIDAM. 35 The neuropsychological test battery of the SIDAM includes the following areas of neuropsychological functioning, orientation (assessment of orientation for time and place), memory (measured by delayed verbal recall of a word list and a fictitious name and address and delayed visual reproduction), intellectual abilities (assessed by items of abstract thinking and judgment), verbal abilities and calculation (assessed by calculating serial sevens, spelling backward, and digit span backward), constructional visuospatial abilities (assessed by copying figures) as well as aphasia and apraxia (assessed by naming objects, reading and obeying a sentence, writing a sentence, and performing a three-stage command). For each cognitive domain, age- and education-specific norms were considered in the evaluation of impairment in cognitive function. 42
Assessment of other covariates
Education was assessed by asking participants on their educational and vocational levels, and categorized as low, middle, or high based on the revised version of the international CASMIN educational classification. 34,43 Low education refers to individuals with inadequately completed general education, general elementary education, basic vocational qualification, or general elementary and vocational qualification. Middle education includes all individuals with an intermediate vocational qualification, an intermediate general qualification and vocational qualification, intermediate general qualification, a general maturity certificate, a vocational maturity certificate or a general maturity certificate, and vocational quality. Finally, high education refers to lower or higher tertiary education. 43 Cognitive status at baseline was assessed with the MMSE. 44 Complaints of subjective memory impairment were assessed before cognitive testing by asking participants about memory problems (answer “yes” or “no”). The capacity to perform instrumental activities of daily living (IADL) was assessed with 9 IADL items according to Schneekloth and Pothoff. 45 The 9 items, for example, cover ability to buy food, to use the telephone, or to handle finances. Participants with difficulties in at least one of the 9 IADL items were regarded as impaired.
Major depression was assessed with the Structured Clinical Interview for DSM (Third Edition Revised). 46 Participants were asked to rate their subjective health status (“How would you rate your current health status?”) on a 5-point scale (1 = very bad; 2 = bad; 3 = satisfactory; 4 = good; and 5 = very good). As described earlier, midlife cognitive functioning was estimated by reference to a sample of O*NET variables on the basis of each participant’s main occupation.
Statistical Analysis
Statistical analyses were performed using PASW for Windows, version 18. Group differences were analyzed using t tests and chi-square tests. To determine the impact of occupation-based motivational processes, the frequency of social contacts, and the proximity of social contacts, Cox proportional hazard models were computed controlling for all covariates. The Cox proportional hazard models included the exact date of diagnosis for each individual, determining whether individuals with lower indices of motivational processes were diagnosed earlier than individuals with high indices.
Due to unexpected results concerning the impact of the motivational processes on the risk of dementia with Cox proportional hazard models, additional post hoc analyses were conducted dividing the sample into individuals with high indices of motivational processes versus participants with low indices of motivational processes. In these 2 subsamples, the impact of frequency and proximity of social contacts was investigated with Cox proportional hazard models. The level of statistical significance was set at .05 for all analyses.
Results
Occupation-Based Motivational Processes and Incidence of Dementia
A total of 772 patients could be included for the calculation of incident dementia at the follow-up waves. Out of these 772 patients, 179 (23.2%) were classified as having dementia by the end of the study (Figure 1). Individuals diagnosed with dementia at one of the follow-up assessments (n = 179) had significantly lower indices of motivational processes (M = −0.07, standard error [SE] = 0.04, t(770) = 3.78, P < .001) than did those without dementia (n = 593, M = 0.20, SE = 0.06). To determine the impact of occupation-based motivational processes on risk of dementia, Cox regression analysis was used (Table 4). Occupation-based motivational processes were not found to have a significant impact on the risk of dementia (hazard ratio [HR]: 0.93, 95% confidence interval [CI]: 0.74-1.16). Reducing the control variables by excluding the MMSE scores and subjective memory complaints did not result in a significant change concerning the impact of motivational processes on risk of dementia (HR: 0.97, 95% CI: 0.64-1.45).
Cox Regression With Motivational Processes, Frequency, and Proximity of Social Contacts as Predictors and Risk of Dementia as Criterion.a
Abbreviations: CI, confidence interval; HR, hazard ratio.
aAll variables were included simultaneously in the model, together with age, gender, and education.
Social Network Variables and Risk of Dementia
Individuals diagnosed with dementia at a follow-up assessment did not have a significantly smaller frequency of social contacts (M = 9.71, SE = 0.32) or lower proximity of social contacts (M = 7.10, SE = 0.25) than individuals who were not diagnosed with dementia (frequency of social contacts: M = 9.95, SE = 0.17, t(770) = 0.67, P = .51; proximity of social contacts: M = 6.82, SE = 0.14, t(761) = −0.94, P = .35). However, when computing a Cox regression analysis, the frequency of social contacts proved to be a significant predictor (HR: 0.96, 95% CI: 0.91-0.99, P < .05) with a higher frequency of social contact associated with a lower risk of dementia. Proximity of social contacts did not have a significant impact on the risk of dementia (HR: 1.03, 95% CI: 0.98-1.08, P = .22).
Occupation-Based Motivational Processes and Social Network Parameters
In a next step, we examined whether occupation-based motivational processes and social network variables were associated. Spearman ρ correlations of the social network items revealed 2 items to be significantly related to occupation-based motivational processes (Table 5), “frequency of contact with friends” (r = .11, P < .01) and “attendance of community/social meetings” (r = .16, P < .01). Thus, the higher the participants’ occupation-based motivational processes, the higher their frequency of contact with friends and attendance of community or social meetings in old age. Occupation-based motivational processes correlated significantly with the frequency sum score (r = .09, P < .05) but not with the proximity sum score (r = −.02, P = .55).
Correlations Between Motivational Processes and Social Network Items.a
aThe values presented are Spearman ρ correlations.
b P < .01.
c P < .05.
Post Hoc Analyses
Due to the null findings concerning the impact of motivational processes on the risk of dementia in the Cox regression analysis, a post hoc analysis was computed separating individuals with high motivational indices from those with low motivational indices (Table 6).
Post Hoc Analysis: Cox Regression Analysis With Risk of Dementia as Criterion in Two Subsamples Separated by High Versus Low Indices of Motivational Processes.a
Abbreviations: CI, confidence interval; HR, hazard ratio.
aHazard ratios for social and motivational variables for individuals with low versus high indices of motivational processes. All analyses controlled for sociodemographic variables (age, sex, and education), MMSE (Mini-Mental State Evaluation) scores, midlife cognitive functioning, activities of daily life, subjective health status, subjective memory complaints, and major depression.
Subsample of individuals with low indices of occupation-based motivational processes
When only analyzing individuals with low motivational indices in a Cox regression model, the frequency of social contacts was linked to a lower risk of dementia (HR: 0.94, 95% CI: 0.88-1.00, P = .05), while proximity of social contacts (HR: 1.02, 95% CI: 0.95-1.10, P = .58) was not linked to the risk of dementia.
Subsample of individuals with high indices of occupation-based motivational processes
When analyzing the impact of social network variables on the risk of dementia in a Cox regression model including only individuals with high indices of motivational processes, the frequency of social contacts (HR: 0.99, 95% CI: 0.93-1.05, P = .72) was not linked to the risk of dementia, while proximity of social contacts was significantly linked to a higher risk of dementia (HR: 1.09, 95% CI: 1.01-1.19, P < .05).
Discussion
This study investigated the impact of occupation-based motivational processes and social network variables on the incidence of dementia over 8 years. Individuals diagnosed with dementia at least at one of the five follow-up assessments over an 8-year period had significantly lower indices of motivational processes than did those who were not diagnosed with dementia during the study period. However, when the impact of occupation-based motivational processes on risk of dementia was analyzed with a Cox proportional hazards model, occupation-based motivational processes did not emerge as a significant predictor.
Different reasons may account for this result. First, the Cox proportional hazards model considers time to event data (referring to time to incidence of dementia), which might have confounded the comparison of the means of motivational processes among the two groups—those who developed dementia and those who stayed cognitively healthy. Also, as the HR is below 1, indicating a (statistically not significant) lower risk of dementia for those with higher motivational processes, the limits of statistical power may have played a role. Other factors might have diminished the validity of the assessment of motivational processes. On one hand, motivational processes were assessed retrospectively. Although this offers the opportunity to explore associations between motivational processes and the risk of dementia over the life span from midlife to old age and data derived from the O*NET database have been shown to predict various outcomes, 47 the validity of these data might be reduced as they rely on the exact indication of the participants’ main occupation. On the other hand, motivational processes were estimated on the basis of each individual’s occupation by reference to a sample of an American classification system (O*NET), 14 which was applied to a German sample of older adults in this study. First, German occupations may not, in all cases, be comparable to American occupations; second, motivational processes were assessed indirectly based on the classification system O*NET. Further studies should therefore directly assess motivation-related variables. Third, a large percentage of women were working as housewives in this cohort, which was classified as an occupation requiring low motivational processes. It can be assumed that the occupation of these women was rather mirroring the society’s expectations and constraints than their motivational processes.
Social Network Parameters, Occupation-Based Motivational Processes, and Risk of Dementia
Although individuals who engaged in frequent social interactions had a significantly lower risk of dementia, proximity of social contacts did not have an impact on the risk of dementia. As similar results were found in previous studies concerning frequency of social contacts, 24,25,48 this effect was not surprising. Although we expected proximity of social contacts to have a beneficial effect, our results indicated no impact of the proximity of social contact on the risk of dementia. Different explanations can account for this lack of significance. Although some individuals simply enjoy having their friends and family in their neighborhood, others may be dependent on their family members’ help. Wenger 48 even reported that patients with dementia are more likely to live with others. Because the reason for the reported proximity of social contacts was not assessed, it is not possible to disentangle the two motives of the participants for living near their social contacts.
A post hoc analysis analyzing individuals with high indices of motivational processes separately from individuals with low indices of motivational processes shed some light on possible determinants of the unexpected null findings concerning the impact of motivational processes on the risk of dementia. Interestingly, when only including individuals with high indices of motivation in a Cox proportional hazards model, proximity of social contacts significantly predicted the risk of dementia, whereas the frequency of social contacts did not. Surprisingly, individuals with higher proximity of their social contacts had a significantly higher risk of dementia. Different explanations are conceivable. In line with Wenger, 48 individuals experiencing first signs of impairment in their daily activities may be more likely to live with their relatives to benefit from their support. As a consequence, it can be postulated that proximity of social contacts indicates first signs of cognitive decline in individuals with high indices of motivation. On the other hand, individuals experiencing more support from their social network may be more inclined to delegate difficult tasks to their family or friends. Hence, individuals living alone are more stimulated by difficult daily activities and tasks, which may postpone clinical signs of dementia.
In the group of individuals with low indices of motivational processes, results indicate that the frequency of social contacts was linked to a lower risk of dementia, confirming earlier results. 24,25,48 However, the frequency of social contact seems to postpone the diagnosis of dementia only in the group with lower indices of motivational processes, whereas no such tendency could be found in the other group. As we have stated earlier, motivational processes are rooted in the concept of MR developed by Forstmeier and Maercker, 1 referring to the contribution of lifetime motivational activities to general brain reserve 3,4 together with cognitive, 5 physical, 6 and social activities. 7 This particular result is interesting, as it indicates that deficits in one area (motivation-related aspects) can be compensated by resources in other areas (social resources), underlining not only to the importance of promotion of the acquisition of coping skills but also to strong family and social ties to develop reserves for old age. 49
Limitations, Strengths, and Outlook
The study has several shortcomings. As motivational variables were not assessed in midlife, they could only be estimated retrospectively on the basis of each participant’s main occupation by reference to a sample of O*NET variables. Due to these methodological issues, retrospectively assessed occupation-based motivational processes were analyzed together with social network parameters assessed at baseline. Thus, motivational and social variables refer to different points in time weakening the significance of the results of this study. To exclude the potential confounder of different points in time, future studies should assess motivational and social variables at the same time. Also, as social contact may be reduced due to dementia just prior to dementia onset, the effect regarding the frequency of social contacts may be biased. However, as we excluded individuals with a diagnosis of dementia at baseline and controlled for MMSE scores, social contact frequency seems to be a strong predictor of risk of dementia independent of cognitive impairment. Yet, a lower frequency of social contact may also be a consequence of early signs of cognitive impairment not measurable by the MMSE scores.
Furthermore, not all relevant risk factors for dementia could be taken into account, for example, cognitive and physical activities, vascular risk factors, or the ApoE ε4 genotype. Further studies on this topic should include these variables to determine the relationship between motivational processes and incidence of dementia. Also, a validation of the PANT in a German-speaking population was beyond the scope of this study, raising the question of the comparability between English- and German-speaking populations. The strengths of the present study include its longitudinal design covering a period of 8 years. Including motivational processes based on the participants’ main occupation even offered the analysis of data over the life span from midlife to old age. The present study offers insight into the impact of several predictor variables on the risk of dementia in a large sample of 1692 individuals at baseline, considering different aspects of social interactions, motivational processes in midlife, and their association with the risk of dementia in old age. Additional strengths of the study include the procedures for diagnosing dementia according to international standards, which involved consensus conferences of physicians and psychologists.
Summary
Results of this study indicate that the frequency of social contacts contribute to the postponement of clinical signs leading to a diagnosis of dementia in old age. The impact of frequency and proximity of social contacts differs in individuals with high versus individuals with low indices of motivational processes. Individuals with low indices of motivational processes benefit from higher frequency of social contacts. Thus, deficits in one area (motivation-related aspects) seem to be compensated by resources in other areas (social resources).
Footnotes
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 field work of LEILA75+ was supported by the Interdisciplinary Centre for Clinical Research Leipzig at the Faculty of Medicine, University of Leipzig (project C07), Germany. Sonja Fankhauser was supported by the Velux Foundation, Zurich, Switzerland and was a fellow of the International Max Planck Research School of the Life Course LIFE. Tobias Luck was supported by LIFE – Leipzig Research Center for Civilization Diseases, University of Leipzig, Germany. LIFE is financed by means of the European Union, by the European Regional Development Fund (ERDF) and by means of the Free State of Saxony within the framework of the excellence initiative.