Disorientation in Time and Place in Old Age: Longitudinal Evidence from Three Old Age Cohorts in Germany (AgeDifferent.de Platform)

Abstract

Background:

Only little evidence is available on disorientation, one of the most challenging symptoms of Alzheimer’s disease and related dementias.

Objectives:

The aim of this study was to investigate the prevalence of disorientation in older age in association with the level of cognitive status, personal characteristics, and life events.

Methods:

Three longitudinal population-based cohort studies on cognitive health of elderly adults were harmonized (LEILA 75 + , AgeCoDe/AgeQualiDe, AgeMooDe). Participants who completed a baseline and at least one follow-up assessment of cognitive functioning and who did not have stroke, Parkinson’s disease, atherosclerosis, kidney disease, and/or alcoholism were included in the analysis (n = 2135, 72.6% female, mean age 80.2 years). Data was collected in standardized interviews and questionnaires with the participant, a proxy informant, and the participant’s general practitioner.

Results:

Making three errors in the MMSE other than in the questions on orientation (MMSE_wo) came with a probability of 7.8% for disorientation, making ten errors with a probability of 88.9%. A lower MMSE_wo score (HR 0.75, CI 95 0.71–0.79, p < 0.001), older age (HR 1.11, CI 95 1.08–1.14, p < 0.001), and living in a nursing home (HR 1.64, CI 95 1.02–2.64, p = 0.042) were associated with incident disorientation. Impairments in walking (OR 2.41, CI 95 1.16–4.99, p = 0.018) were associated with a greater probability for prevalent disorientation. None of the life events were significant.

Conclusion:

Our findings suggest that disorientation is primarily associated with cognitive status. Regular walking activities might possibly reduce the risk for disorientation but further research is necessary.

Keywords

Cognitive functioning cognitive status dementia longitudinal cohort orientation symptoms

INTRODUCTION

Disorientation—the loss of the knowledge of temporal and spatial information about one’s person [1]—is a common symptom in aging, especially in the context of Alzheimer’s disease and related dementias (ADRD) [2]. With the aging of societies worldwide, the number of older people who experience disorientation is increasing. Therefore, it is necessary to understand this phenomenon and to develop ways of dealing with it.

People of older age might feel disoriented in unfamiliar environments (e.g., moving to a new home, going to a new place) [3] but it might also happen in their daily environment. Wandering and other phenomena of disorientation often appear during moderate stages of ADRD [4], but there is insufficient research explaining why and when it occurs [5]. Knowledge on disorientation is very limited. Studies have shown that cognitive status is a predictor for disorientation [6]. Loss of gray matter [7] and degeneration of brain areas such as the parietal lobe and the hippocampus [8] seem to increase the likelihood for temporal or spatial disorientation. It is not clear at what level of cognitive functioning disorientation is most likely to occur. Yet, this is important to know for caregivers and carers in nursing homes as well as for other people such as pharmacists, drivers, and hairdressers. For predicting disorientation in daily life and for defining purposeful strategies to deal with it among the elderly, a better understanding of when symptoms occur is essential.

For successful orientation, spatial, temporal, and social information must be processed, which depends mental maps in the person’s brain. These are reflected in a sequential posterior–anterior pattern of activity in the brain [9]. Spatial reasoning is a combination of knowledge of the properties of space as well as the ability to abstract appropriate references based on the available information [10]. With dementia, these abilities get lost. People with dementia, for instance, show spatial rotation deficits which do not occur in normal aging and seems to be disease-specific [11]. Evidence indicates that orientation is related to performance in episodic memory, attention, and processing speed and is predicted by the thickness of the entorhinal cortex, the hippocampus, and the inferior temporal cortex as well as the glucose metabolism of the middle-inferior temporal cortex and the hippocampus [8]. In fact, a study demonstrated that disorientation is associated with the deterioration of the pathway linking the hippocampus to the superior parietal and posterior cingulate cortex in the right atmosphere [12], brain areas affected very early in Alzheimer’s disease [13, 14]. It is therefore likely that disorientation is a symptom that is closely connected to the pathology of dementia. However, it is unclear whether it is only disease-driven or whether it is susceptible to personal characteristics and life events.

The aim of this study was therefore to investigate the prevalence of disorientation in the course of cognitive aging. Specifically, the study takes an explorative approach to examine the association between disorientation and the level of cognitive status (first aim) as well as personal characteristics (secondary aim). Moreover, life events such as change of residence may trigger disorientation [15]. Hence, the third aim of the study was to explore the relevance of the events ‘Living alone’ after being used to live with someone, ‘Becoming widowed’, ‘Hospital stay’, ‘Moving’, and ‘Losing someone close’ with respect to disorientation.

METHODS

Study design and sample

Data from three longitudinal cohort studies, the “Leipzig Longitudinal Study of the Aged” (LEILA 75 + , n = 1,692, 1997–2013), the “German Study on Ageing, Cognition, and Dementia in Primary Care Patients” (AgeCoDe/AgeQualiDe, n = 3,327, 2003–2016) and the study on “Late-life depression in primary care: needs, health care utilization and costs” (AgeMooDe, n = 1,451, 2012–2014) were pooled to one harmonized dataset, the AgeDifferent.de dataset (N = 6,470). The LEILA 75 + is a representative population-based study of people aged 75 years and older in Leipzig, Germany (for details, see [16]). Participants were recruited via random sampling from an age-ordered list provided by the local registry office and were followed up in up to six assessments at an interval of 18 months and a long-term assessment after 13 years. The AgeCoDe/AgeQualiDe study of people aged 75 years and older is a multicenter prospective cohort study of general practitioner (GP) patients in six different German cities. Community-dwelling individuals without a fatal disease were invited to participate in up to ten assessments at an interval of 18 months (for details, see [17]). The AgeMooDe study is a multicenter cohort study in six different German cities. Participants aged 75 years and older without fatal disease were recruited from GP practices and completed up to two assessments that were conducted at an interval of 12 months (for details, see [18]). The datasets were merged to increase the sample size and the stability of the models. All participants from the three original cohort studies provided written informed consent prior to study participation. The ethics committees of all the participating centers separately approved each of the original studies. All studies were conducted in accord with the Helsinki Declaration of 1975 on Medical Research Involving Human Subjects.

Of the pooled sample, 649 (10.0%) participants did not complete the baseline assessment and 1,128 (17.4%) participants did not complete a follow-up assessment. We excluded 92 (1.4%) participants from analysis because of incomplete cognitive assessment and n = 2,466 (38.1%) participants because they had a medical condition (in any assessment wave) that can cause disorientation (i.e., stroke, Parkinson’s disease, atherosclerosis, kidney infection/disease, alcoholism). The medical conditions were assessed at each assessment wave via a questionnaire that was completed by the patients’ GP. As the prevalence of these medical conditions is generally high in the older population, this is the major source of loss of participants. The final sample comprised 2,135 participants; a flow chart is depicted in Supplementary Figure 1. Excluded participants were more likely to have a lower level of education (χ² = 7.634, p = 0.022), to be male (χ² = 33.82, p < 0.001), older (81.2 versus 80.6, Kruskal-Wallis χ² = 26.92, p < 0.001), living with someone instead of living alone (χ² = 15.92, p < 0.001), have diabetes (χ² = 55.37, p < 0.001), heart disease (χ² = 145.64, p < 0.001), difficulties walking (χ² = 111.369, p > 0.001), difficulties seeing (χ² = 17.06, p = 0.001), difficulties hearing (χ² = 19.26, p < 0.001), or a moderate/severe dementia stage (χ² = 69.39, p < 0.001).

Cognitive status and ADRD

Assessment of cognitive status was performed in the participants’ homes by trained physicians and psychologists. In all three studies, the “Structured Interview for Diagnosis of Dementia of Alzheimer type, Multi-infarct Dementia, and Dementia of other Etiology according to the Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition, Revised (DSM-III-R), Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV), and International Classification of Diseases, 10th Revision (ICD-10)” (SIDAM) was used, which included sections on eight cognitive domains (orientation, memory, abstract reasoning, verbal ability, calculation, constructional ability, aphasia, apraxia), a section for clinical diagnostics, and the Mini-Mental Status Examination (MMSE). The MMSE was used as an indicator for cognitive status after subtracting the questions on orientation (MMSE_wo).

Disorientation

To investigate disorientation, we used the section on orientation in the SIDAM comprising ten questions (year, season, date, weekday, month, address, floor, city, state, country). Factor analysis was used to evaluate whether all forms of disorientation occur simultaneously or whether there are different clusters of disorientation symptoms. We used principal component analysis based on the positive-semidefinite correlation matrix of the ten orientation items. Model fit and screeplot suggest either a one or two factor solution, but as the loading on the first factor was 827.96 (proportion 0.972) and the loading on the second factor was only 0.29 (proportion 0.034) and the uniqueness of items was higher for the one factor solution (Supplementary Table 1), it seemed appropriate to work with the one factor solution. The latent variable for disorientation was then calculated via generalized structural equation modelling (correlation matrix is shown in Supplementary Table 2). This latent construct score was reversed so that a higher score indicates more impairments in orientation. Receiver operating characteristics (ROC) were used to determine a cutoff for disorientation. We run ROC curves for 1) disorientation (at least one symptom of disorientation), 2) ADL (more than two points), and 3) IADL (more than two points). As we were interested in correctly classifying those without disorientation, we set the specificity to be >90% and the percentage of correctly classified people to be >75%. The optimal cutoff was 7.163 with a specificity of 100%, a correct classification of 83.6%, and a sensitivity of 39.6% for disorientation, a specificity of 90.3%, a correct classification of 78.3%, and a sensitivity of 36.0% for activities of daily living (ADL), and a specificity of 91.2%, a correct classification of 89.4%, and a sensitivity of 25.2% for IADL. According to this cutoff-point, a binary variable for disorientation was created with 0 representing full orientation and 1 representing disorientation.

Confounders

Information on confounding variables were obtained in a standardized interview asking structured questions on education, gender, age, marital status, living conditions, and others. Information on health status including difficulties in hearing, seeing, walking, etc., were obtained in structured interviews with proxy informants and via standardized questionnaires completed by the participants’ GPs. Deficits in ADL were assessed using the Barthel-Index and the Lawton Instrumental Activities of Daily Living Scale (IADL). Both scales consist of items on the ability to manage daily life (e.g., shopping, food preparation) that were each rated on three levels (0 - independent, 1 - some assistance, 2 - completely dependent). The sum of the items made up the ADL and IADL score.

Life events

Life events were determined based on the information that the participants provided for his or her living situation and marital status as well as based on the questions “Have you been admitted to the hospital since the last interview?”, “Did you move since the last interview?”, and “Did you lose a family member due to death?”. By comparing the participants’/their proxies’ answers to the answers from the previous assessment, it was determined whether the participant had experienced the following events between two assessments: Event ‘Living alone’, for participants who used to live with someone and then lived alone; Event ‘Widowed’, for participants who used to be married and then were widowed; Event ‘Hospital’, for participants who reported a hospital stay since the last interview; Event ‘Moving’, for participants who reported moving since the last interview with the codes ‘no’ (0), ‘yes, but not nursing home’ (1), and ‘yes, in nursing home’ (2); Event ‘Loss’, for participants who reported the loss of a family member since the last interview with the codes ‘no’ (0), ‘yes, but not a spouse’ (1), and ‘yes, a spouse’ (2).

Statistical analyses

All statistical analyses employed an alpha level for statistical significance of 0.05 (two-tailed) and were performed using Stata 15. Cases with missing values were excluded from the analysis.

Characteristics between participants who had disorientation over the study and those who did not as well as between participants who were lost early in the study versus those that were followed-up longer were analyzed via Pearson’s chi square test for categorical variables and via Kruskall-Wallis test for continuous variables.

Associations between disorientation and cognitive status were first analyzed on an observational level. All participants together provided a total of i = 16,731 observations throughout the entire assessment period. To estimate how well cognitive status predicts disorientation, we calculated Cohen’s kappa statistics for each item of the SIDAM as well as ROC for the MMSE score without disorientation items (MMSE_wo).

Associations between disorientation and characteristics of participants (age, gender, education, marital status, living condition, difficulties walking, seeing, hearing, diabetes, heart disease at baseline) were analyzed via Pearson’s chi square test and logistic regression. To estimate to what extent these participant characteristics predict the development of disorientation over the follow-up period, we ran a Fine and Gray competing risk model (disorientation versus mortality) with time in days as time variable, excluding every participant who had disorientation at baseline. A Fine and Gray model is used instead of a Cox proportional hazard model if there is a competing event to the outcome of interest. In this case, the event is ‘death’, which may bias the results on the outcome ‘disorientation’. The model incorporates the hypothetical risk for disorientation of those who died [19]. We adjusted the analyses for the number of follow-ups that the participant completed.

The sample loss over the study period was the following: In FU2, n = 1,386 (64.9%) completed the assessment, n = 1,012 (47.4%) in FU3, n = 788 (36.9%) in FU4, n = 644 (30.2%) in FU5, n = 370 (17.3) in FU6, n = 255 (11.9%) in FU7, n = 229 (10.7%) in FU8, and n = 205 (9.6%) in FU9. Overall, n = 921 (43.1%) died over the study (average time until death 8.65 years (SD 3.27)). Those lost early in the study (only two assessments) were more likely to have a higher level of education (χ² = 20.46, p < 0.001), to be male (χ² = 12.96, p < 0.001), married (χ² = 17.14, p = 0.001), older (81.04 versus 80.34, Kruskal-Wallis Test χ² = 9.043, p = 0.003), live with someone or in nursing home (χ² = 9.07, p = 0.011), have diabetes (χ² = 7.03, p = 0.008), heart disease (χ² = 43.46), difficulties walking (χ² = 16.34, p = 0.001), seeing (χ² = 20.64, p < 0.001), or hearing (χ² = 23.61, p < 0.001). There was no difference in the MMSE scores at baseline (27.0 versus 27.1, Kruskal-Wallis Test χ² = 2.479, p = 0.115) but those lost early had a younger age of ADRD onset (84.2 versus 87.6, Kruskal-Wallis Test χ² = 4.61, p = 0.032).

In addition, multilevel mixed-effects regression was conducted allowing all the predictors to vary across time. The model included time (years since age 68), a squared term for time, the number of follow-ups, and the interaction with time of each variable as well as random effects for participants and a random slope.

The impact of life events on the probability of disorientation was analyzed using the event and the interaction of the event with time as additional predictor in the multilevel mixed-effects regression described above. Analyses were rerun for each event separately.

For sensitivity analysis, we repeated all the analyses but separately for disorientation in time and disorientation in space with disorientation being defined as one error in the orientation questions.

RESULTS

At baseline, n = 155 (7.3%) participants had disorientation. Another n = 377 (19.0%) developed disorientation over the study period (n = 101 in FU1, n = 64 in FU2, n = 61 in FU3, n = 38 in FU4, n = 30 in FU5, n = 43 in FU6, n = 10 in FU7, n = 12 in FU8, n = 18 in FU9). The mean number of FUs was 4.4 (standard deviation, SD 2.6) with a mean study time of 4.5 years (SD 3.7 years). The mean study time until disorientation was 4.57 years (SD 3.74 years). A total of n = 825 (38.6%) participants who did not have disorientation at baseline died over the study period without developing disorientation. Of those who died, n = 197 (23.9%) had developed disorientation during the course of the study before their death. Characteristics of the participants at baseline are shown in Supplementary Table 3.

Level of cognitive status and disorientation

On an observation level, in cognitive testing, the question that came with the highest agreement with disorientation was “remembering the word CENT” (Cohen’s kappa of 0.176, 46.2% agreement), followed by “remembering a street address” (Cohen’s kappa of 0.175, 40.4% agreement), and “remembering the word TABLE” (Cohen’s kappa of 0.159, 28.8% agreement; details in Supplementary Table 4). Using the criteria of at least one error in the remembering-a-name-and-address task came with a slightly better kappa of 0.189 (48.9% agreement).

The ROC curve for the MMSE score without the orientation questions (MMSE_wo) suggests that, at 19 points, the chance to be correctly classified as not having any disorientation is 53.4% (specificity), while the chance of being correctly classified as having disorientation is 4.9% (sensitivity; details in Supplementary Table 5, Supplementary Figure 2). At 10 points, the rate of disorientation is 78.3% (sensitivity); however, clearly not everyone with this score had disorientation (specificity 0.08%, Supplementary Table 5). Logistic regression analyses suggest that the probability of having disorientation is 7.8% or less for people with a MMSE_wo score of 17 or higher. With a MMSE_wo score of 10, the probability is 88.9% (details in Supplementary Table 6, Supplementary Figure 3).

Personal characteristics and disorientation

Results from logistic regression analysis suggest that the likelihood of having disorientation at baseline was significantly higher for those who were older, who had a lower MMSE_wo score, who had impairments walking, difficulties hearing, or were living in a nursing home (Table 1). The likelihood of incident disorientation over the follow-up period was significantly higher for those who were older, who had a lower MMSE_wo score (Fig. 1), and who were living in a nursing home (Table 2). We rerun the analyses for institutionalized and community-dwelling individuals separately. For institutionalized individuals, impairments walking (OR 2.43, CI95 0.31–18.78, p = 0.396) and age (OR 0.98, CI95 0.85–1.14, p = 0.831) lost significance as predictor for disorientation at baseline, but age remained a significant predictor for incident disorientation (HR 1.37, CI95 1.17–1.61, p < 0.001). Moreover, for institutionalized individuals, diabetes (HR 3.12, CI95 1.12–8.69, p = 0.029) increased and being married (HR 0.16, CI95 0.03–0.72, p = 0.018) decreased the risk for incident disorientation.

Table 1

Estimates from logistic regression analysis of the association between disorientation and characteristics of the participants

		OR	SE	p	95% CI
Age		1.072	0.026	0.004	1.023	1.123
Gender	Female	Ref.
	Male	0.780	0.219	0.378	0.449	1.356
Education	High school or less	Ref.
	(Some) college	0.919	0.209	0.710	0.588	1.436
	Graduate/university	0.396	0.208	0.078	0.142	1.109
Walking	No problems	Ref.
	Difficulties	1.413	0.317	0.123	0.910	2.193
	Impairments	2.410	0.896	0.018	1.163	4.996
Seeing	No problems	Ref.
	Difficulties	1.119	0.293	0.666	0.671	1.865
	Impairments	0.806	0.364	0.632	0.333	1.951
Hearing	No problems	Ref.
	Difficulties	0.515	0.128	0.008	0.316	0.839
	Impairments	2.087	0.966	0.112	0.843	5.172
Diabetes	Yes	1.281	0.304	0.297	0.804	2.039
Heart disease	Yes	0.664	0.160	0.090	0.414	1.067
Marital status	Single	Ref.
	Married	1.341	0.755	0.603	0.445	4.044
	Divorced	1.117	0.705	0.861	0.324	3.846
	Widowed	1.418	0.674	0.463	0.558	3.599
Living situation	Alone	Ref.
	With someone	1.324	0.454	0.413	0.676	2.594
	Nursing Home	2.318	0.779	0.012	1.199	4.480
MMSE_wo		0.666	0.029	<0.001	0.612	0.724

95% CI, 95% confidence interval; MMSE_wo, Mini-Mental Status Exam score without orientation items; OR, odds ratio; p, level of significance; Ref., reference category; SE, standard error.

Fig.1

Cumulative incidence of disorientation by MMSE_wo score as predicted by age, gender, education, marital status, living condition, difficulties walking, seeing, and hearing, diabetes, and heart disease. MMSE_wo, Mini-Mental Status Exam score without orientation items.

Table 2

Estimates from Fine and Gray modelling predicting incident disorientation based on characteristics of the participants

		HR	SE	p	95% CI
Age		1.109	0.016	<0.001	1.078	1.142
Number of FUs		0.728	0.030	<0.001	0.671	0.790
Gender	Female	Ref.
	Male	0.844	0.130	0.273	0.624	1.143
Education	High school or less	Ref.
	(Some) college	1.092	0.153	0.530	0.829	1.437
	Graduate/university	1.169	0.265	0.493	0.749	1.824
Walking	No problems	Ref.
	Difficulties	0.977	0.122	0.852	0.765	1.248
	Impairments	0.929	0.272	0.802	0.524	1.649
Seeing	No problems	Ref.
	Difficulties	1.243	0.186	0.147	0.927	1.668
	Impairments	0.746	0.205	0.286	0.436	1.278
Hearing	No problems	Ref.
	Difficulties	1.022	0.123	0.855	0.808	1.293
	Impairments	1.204	0.444	0.615	0.584	2.480
Diabetes		0.886	0.132	0.416	0.661	1.187
Heart disease		1.059	0.129	0.637	0.834	1.345
Marital status	Single	Ref.
	Married	0.825	0.215	0.459	0.495	1.374
	Divorced	0.738	0.199	0.261	0.434	1.253
	Widowed	0.974	0.193	0.895	0.660	1.437
Living situation	Alone	Ref.
	With someone	1.199	0.222	0.326	0.834	1.725
	Nursing Home	1.638	0.397	0.042	1.019	2.635
MMSE_wo		0.754	0.019	<0.001	0.716	0.794

95% CI, 95% confidence interval; FUs, follow-ups; HR, hazards ratio; MMSE_wo, Mini-Mental Status Exam score without orientation items; p, level of significance; Ref., reference category; SE, standard error.

Multilevel mixed-effects regression analyses were conducted to assess the general influence of personal characteristics on disorientation. Results confirm the importance of the MMSE_wo score, older age, walking, and living in a nursing home (Table 3). For institutionalized individuals, the MMSE_wo score remained the only predictor for incident disorientation.

Table 3

Estimates from multilevel mixed-effects logistic regression analysis predicting disorientation based on characteristics of the participants, adjusted for number of follow-ups and its interaction with age (time variable)¹

		b	SE	p	95% CI
Age		0.235	0.103	0.022	0.034	0.435
Age^*age		0.003	0.002	0.122	–0.001	0.007
Number of FUs		–0.261	0.066	<0.001	–0.389	–0.132
Number of FUs^*age		0.013	0.004	0.001	0.006	0.020
MMSE_wo		–0.471	0.076	<0.001	–0.619	–0.323
MMSE_wo^*age		–0.012	0.004	0.009	–0.020	–0.003
Gender	Female	Ref.
	Male	–0.300	0.456	0.510	–1.193	0.593
Gender^*age	Female	Ref.
	Male	–0.008	0.027	0.757	–0.062	0.045
Education	High school or less	Ref.
	(Some) college	–0.085	0.421	0.840	–0.911	0.741
	Graduate/university	–0.452	0.706	0.941	–1.436	1.331
Education^*age	High school or less	Ref.
	(Some) college	0.002	0.025	0.949	–0.047	0.050
	Graduate/university	0.005	0.042	0.906	–0.077	0.087
Walking	No problems	Ref.
	Difficulties	0.868	0.379	0.022	0.125	1.610
	Impairments	1.033	0.622	0.097	–0.186	2.253
Walking^*age	No problems	Ref.
	Difficulties	–0.051	0.023	0.024	–0.095	–0.007
	Impairments	–0.037	0.034	0.265	–0.103	0.038
Seeing	No problems	Ref.
	Difficulties	0.147	0.458	0.749	–0.751	1.044
	Impairments	0.316	0.659	0.632	–0.975	1.607
Seeing^*age	No problems	Ref.
	Difficulties	–0.013	0.026	0.602	–0.064	0.037
	Impairments	–0.020	0.034	0.555	–0.087	0.047
Hearing	No problems	Ref.
	Difficulties	–0.013	0.383	0.974	–0.763	0.738
	Impairments	0.367	0.845	0.664	–1.289	2.025
Hearing^*age	No problems	Ref.
	Difficulties	–0.021	0.022	0.340	–0.064	0.022
	Impairments	0.006	0.043	0.880	–0.077	0.090
Diabetes	Yes	0.139	0.407	0.731	–0.658	0.938
Diabetes^*age	Yes	–0.018	0.024	0.467	–0.065	0.029
Heart disease	Yes	–0.104	0.365	0.776	–0.819	0.612
Heart disease^*age	Yes	0.002	0.021	0.912	–0.039	–0.044
Marital status	Single	Ref.
	Married	0.134	0.939	0.886	–1.706	1.974
	Divorced	0.644	1.021	0.529	–1.348	2.645
	Widowed	0.497	0.785	0.527	–1.043	2.036
Marital status^*age	Single	Ref.
	Married	0.018	0.056	0.749	–0.091	0.127
	Divorced	–0.018	0.059	0.757	–0.133	0.097
	Widowed	0.000	0.045	1.000	–0.089	0.089
Living situation	Alone	Ref.
	With someone	1.274	0.566	0.024	0.164	2.384
	Nursing Home	2.491	0.654	<0.001	1.208	3.773
Living situation^*age	Alone	Ref.
	With someone	–0.051	0.032	0.104	–0.113	0.011
	Nursing Home	–0.075	0.034	0.029	–0.142	–0.008

¹estimates for number of follow-ups and its interaction with age not reported; b, coefficient; 95% CI, 95% confidence interval; FUs, follow-ups; MMSE_wo, Mini-Mental Status Exam score without orientation items; p, level of significance; Ref., reference category; SE, standard error.

Life events and disorientation

The impact of life events on disorientation, analyzed via multilevel mixed-effects regression analysis, yielded no statistically significant effects (Table 4). The results were no different for institutionalized and community-dwelling individuals.

Table 4

Estimates from multilevel mixed-effects logistic regression analysis predicting disorientation based on characteristics of the participants and life events, as estimated in separate analyses for each event adjusted for and its interaction with age, age (time variable), number of follow-ups, gender, education, marital status, living condition, difficulties walking, seeing, and hearing, diabetes, heart disease, and MMSEwo at baseline

		b	SE	p	95% CI
‘Living alone’		–0.514	1.627	0.752	–3.702	2.675
‘Living alone’^*age		–0.020	0.091	0.825	–0.198	0.158
‘Widowed’		–0.815	1.626	0.616	–4.003	2.372
‘Widowed’^*age		0.038	0.088	0.669	–0.135	0.210
‘Hospital’		0.013	0.607	0.983	–1.178	1.203
‘Hospital’^*age		0.000	0.036	0.995	–0.071	0.071
‘Moving’	Yes, not nurs.	–0.284	2.144	0.895	–4.486	3.919
	Yes, not nurs.^*age	–0.003	0.137	0.984	–0.270	0.265
	Yes, nurs.	1.444	2.256	0.522	–2.978	5.866
	Yes, nurs.^*age	–0.097	0.132	0.462	–0.355	0.162
‘Loss’	Yes, not spouse	0.671	0.820	0.413	–0.937	2.279
	Yes, not spouse^*age	–0.060	0.048	0.212	–0.155	0.034
	Yes, spouse	–1.027	1.389	0.459	–3.749	1.694
	Yes, spouse^*age	0.072	0.083	0.383	–0.090	0.235

b, coefficient; 95% CI, 95% confidence interval; MMSE_wo, Mini-Mental Status Exam score without orientation items; nurs., nursing home; p, level of significance; SE, standard error.

Sensitivity analyses

To analyze differences for disorientation in time versus disorientation in space, we repeated the analyses for time and space separately. Results (Supplementary Tables 7–12) suggest that, for disorientation at baseline, the predictors for disorientation in time were similar to those of the general construct of disorientation, except that living in a nursing home was not a strong predictor. For disorientation in space at baseline, men, higher educated individuals, and those without seeing impairments were significantly less likely to have symptoms of disorientation, while the MMSEwo score was non-significant. For incident disorientation in space, only male gender remained a significant predictor, the MMSEwo score became significant, and living in a nursing home was non-significant (results from multilevel mixed-effects regression analysis confirm this, see Supplementary Tables 11 and 12). For incident disorientation in time, higher education and no seeing impairments were significantly less likely to experience incident disorientation; however these factors did not retain significance in the multilevel mixed-effects regression analysis (Supplementary Tables 8 and 9). None of the life events were significant predictors for either type of disorientation (results not shown).

DISCUSSION

The aim of this study was to investigate the prevalence of disorientation in aging. We examined the association between disorientation and the level of cognitive status as well as personal characteristics. Results suggest that about every second person who makes seven or more errors in the MMSE (other than on the questions on orientation) has disorientation. The likelihood of disorientation is higher for people who score lower on the MMSE as well as for older people, for those living in a nursing home, and to some extent for those with impairments walking. Among institutionalized individuals, diabetes increased the risk for incident disorientation and being married decreased the risk. While diabetes might exacerbate the neurodegenerative processes that lead to disorientation, married institutionalized individuals might benefit from a spouse that provides the necessary stimulation to maintain orientation for a slightly longer time period. Further results from sensitivity analyses suggest that orientation in space is more preserved in men, in those higher educated, and in those without seeing impairments. It is important to note that this observation only holds for prevalent and not for incident disorientation. However, it demonstrates that, as men are known to have a better spatial reasoning than women, it seems that this ability is also better retained in old age. Higher educated individuals might be able to compensate for early symptoms of neurodegeneration better than lower educated people (i.e., cognitive reserve) so that they can maintain a better orientation in space. Nonetheless, education does not seem to protect against disorientation in general, as our findings suggest.

Research on orientation is extremely rare. The few studies that exist show that, in the brain, time disorientation seems to be associated with a reduced cerebral blood flow in the left posterior cingulate cortex (PCC) [20] and a disrupted connection between the PCC and the right ventral attention network [21]. One study indicated that the severity of disorientation seems to be related to the gray matter volume in the right inferior parietal lobule (IPL) [22] so that disorientation may be a result of neurodegenerative processes [23]. Space orientation seems to be associated with volume and glucose metabolism of the inferior temporal cortex and the hippocampus [8] and the pathway linking the hippocampus to the superior parietal and posterior cingulate cortex in the right atmosphere [12], brain areas affected very early in Alzheimer’s disease [13, 14]. If orientation does indeed depend on brain areas, which deteriorate early in dementia, it is not surprising that we observed a strong association between disorientation and the MMSE score in our study.

The main predictors for disorientation in our analysis are living in a nursing home and impairments walking. They might be so relevant because a lack of mobility is also a lack of stimulation. The more an individual is moving independently in his or her environment, the more brain resources are stimulated which are involved in the process of orientation. According to the ‘use it or lose it’ theory, a lack of stimulation leads to the deterioration of brain areas [24]. A lack of walking activities could therefore expedite the deterioration of brain areas needed for orienting in space and time and enhance the likelihood of disorientation.

We also examined life events as predictors for disorientation. None of them were significant. Evidence on events or situations that might enhance disorientation is extremely rare. A previous study demonstrated that a repeated change of residence might increase the likelihood of disorientation [15]. We could not confirm the relevance of change of residence in our study. However, our standard errors were large so that we cannot exclude the possibility entirely. Further research is necessary to investigate phenomena of disorientation in daily life.

The study comes with some limitations. First, we used data from longitudinal cohort studies and did not observe events of disorientation in daily life. Second, our study investigated cognitive status, not dementia. Cognitive status was measured by the MMSE, a good indicator for dementia-related cognitive decline. Yet, the MMSE has ceiling effects for those who have still a relatively good cognitive functioning so that it is unclear what the association is in very early stages. Further, disorientation might also differ by dementia type as a study has shown that disorientation is a good marker for frontotemporal dementia. Third, we excluded participants with medical conditions that are known to cause disorientation (i.e., stroke, Parkinson’s disease, atherosclerosis, kidney infection/disease, alcoholism) with the purpose of avoiding biased results due to these conditions. Hence, we cannot draw any conclusion on these types of patients. It is important to note that excluding participants with medical conditions led to a large loss of participants for our studies so that it is possible we lost variance at the lower end and underestimated the true effect. Fourth, we did not take into account potential effects due to delirium caused by acute or other chronic diseases. We hope that the great number of participants in our analysis canceled out a potential bias. Fifth, the data on life events was obtained from self-reported instruments and we did not validate them. Finally, disorientation might be related to deficits in specific attentional processes. Our study did not assess cognitive abilities in enough detail to test such associations.

Disorientation is a frustrating symptom for people affected by it and their caregivers. By today, knowledge on disorientation is very limited. Our study aimed at identifying when disorientation is most likely to occur. The findings confirm that cognitive status is the major predictor for disorientation, with living in a nursing home and having impairments walking coming with an increased risk. Future studies may investigate whether an intervention that comprises walking activities with orientation abilities may decrease the risk for disorientation and act as a stimulation for brain areas relevant for successful orientation. Further research is needed to determine causal factors and useful strategies for preventing and delaying disorientation.

Footnotes

ACKNOWLEDGMENTS

This publication is part of the study “Healthy Aging: Gender specific trajectories into latest life” (AgeDifferent.de). We thank the members of the AgeDifferent.de Study Group: Steffi G. Riedel-Heller (Principal Investigator), Franziska Förster, Johannes Golchert, André Hajek, Kathrin Heser, Hans-Helmut König, Wolfgang Maier, Alexander Pabst, Elzbieta Buczak-Stec, Michael Wagner, Birgitt Wiese.

The study was supported by the German Federal Ministry of Education and Research (Funding program “Gesund – ein Leben lang”, grants 01GL1714A; 01GL1714B; 01GL1714C; 01GL1714D), the German Research Foundation (DFG, grant # TH2137/3-1), and the Hans und Ilse Breuer Foundation.

Authors’ disclosures available online ()

The supplementary material is available in the electronic version of this article: .

References

Cossa

, Sala

, Spinnler

(1995) Alzheimer patients know their date of birth but not their age: A study on disorientation. Int J Geriatr Psychiatry 10, 99–106.

Liu

, Gauthier

(1991) Spatial disorientation in persons with early senile dementia of the Alzheimer type. Am J Occup Ther 45, 67–74.

Monacelli

, Cushman

, Kavcic

, Duffy

(2003) Spatial disorientation in Alzheimer’s disease: The remembrance of things passed. Neurology 61, 1491–1497.

Hope

, Keene

, McShane

, Fairburn

, Gedling

, Jacoby

(2001) Wandering in dementia: A longitudinal study. Int Psychogeriatr 13, 137–147.

Lai

, Arthur

(2003) Wandering behaviour in people with dementia. J Adv Nurs 44, 173–182.

Joray

, Herrmann

, Mulligan

, Schnider

(2004) Mechanism of disorientation in Alzheimer’s disease. Eur Neurol 52, 193–197.

Lim

, Iaria

, Moon

(2010) Topographical disorientation in mild cognitive impairment: A voxel-based morphometry study. J Clin Neurol 6, 204–211.

Sousa

, Gomar

, Goldberg

(2015) Neural and behavioral substrates of disorientation in mild cognitive impairment and Alzheimer’s disease. Alzheimers Dement (N Y) 1, 37–45.

Peer

, Salomon

, Goldberg

, Blanke

, Arzy

(2015) Brain system for mental orientation in space, time, and person. Proc Natl Acad Sci U S A 112, 11072–11077.

10.

Freksa

(1992) Using orientation information for qualitative spatial reasoning. In Theories and methods of spatio-temporal reasoning in geographic space. Springer, pp. 162–178.

11.

Armstrong

, Cloud

(1998) The emergence of spatial rotation deficits in dementia and normal aging. Neuropsychology 12, 208.

12.

Giannakopoulos

, Gold

, Duc

, Michel

J-P

, Hof

, Bouras

(2000) Neural substrates of spatial and temporal disorientation in Alzheimer’s disease. Acta Neuropathol 100, 189–195.

13.

Convit

, de Leon

, Golomb

, George

, Tarshish

, Bobinski

, Tsui

, De Santi

, Wegiel

, Wisniewski

(1993) Hippocampal atrophy in early Alzheimer’s disease: Anatomic specificity and validation. Psychiatr Q 64, 371–387.

14.

Zhou

, Dougherty

Jr. , Hubner

, Bai

, Cannon

, Hutson

(2008) Abnormal connectivity in the posterior cingulate and hippocampus in early Alzheimer’s disease and mild cognitive impairment. Alzheimers Dement 4, 265–270.

15.

Pai

, Jacobs

(2004) Topographical disorientation in community-residing patients with Alzheimer’s disease. Int J Geriatr Psychiatry 19, 250–255.

16.

Riedel-Heller

, Busse

, Aurich

, Matschinger

, Angermeyer

(2001) Prevalence of dementia according to DSM-III-R and ICD-10: Results of the Leipzig Longitudinal Study of the Aged (LEILA75+) Part 1. Br J Psychiatry 179, 250–254.

17.

Heser

, Tebarth

, Wiese

, Eisele

, Bickel

, Kohler

, Mosch

, Weyerer

, Werle

, Konig

, Leicht

, Pentzek

, Fuchs

, Riedel-Heller

, Luppa

, Prokein

, Scherer

, Maier

, Wagner

, Age CoDe Study Group (2013) Age of major depression onset, depressive symptoms, and risk for subsequent dementia: Results of the German study on Ageing, Cognition, and Dementia in Primary Care Patients (AgeCoDe). Psychol Med 43, 1597–1610.

18.

Stein

, Pabst

, Weyerer

, Werle

, Maier

, Miebach

, Scherer

, Stark

, Kaduszkiewicz

, Wiese

, Moor

, Bock

, Konig

, Riedel-Heller

(2016) Unmet care needs of the oldest old with late-life depression: A comparison of patient, caring relative and general practitioner perceptions - Results of the AgeMooDe study. J Affect Disord 205, 182–189.

19.

Fine

, Gray

(1999) A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc 94, 496–509.

20.

Yamashita

K-I

, Taniwaki

, Utsunomiya

, Taniwaki

(2014) Cerebral blood flow reduction associated with orientation for time in amnesic mild cognitive impairment and Alzheimer disease patients. J Neuroimaging 24, 590–594.

21.

Yamashita

K-I

, Uehara

, Prawiroharjo

, Yamashita

, Togao

, Hiwatashi

, Taniwaki

, Utsunomiya

, Matsushita

, Yamasaki

, Kira

J-I

(2019) Functional connectivity change between posterior cingulate cortex and ventral attention network relates to the impairment of orientation for time in Alzheimer’s disease patients. Brain Imaging Behav 13, 154–161.

22.

Oishi

, Yamasaki

, Tsuru

, Minohara

, Tobimatsu

(2018) Decreased gray matter volume of right inferior parietal lobule is associated with severity of mental disorientation in patients with mild cognitive impairment. Front Neurol 9, 1086–1086.

23.

Dafni-Merom

, Peters-Founshtein

, Kahana-Merhavi

, Arzy

(2019) A unified brain system of orientation and its disruption in Alzheimer’s disease. Ann Clin Transl Neurol 6, 2468–2478.

24.

Salthouse

(2006) Mental exercise and mental aging evaluating the validity of the “use it or lose it” hypothesis. Perspect Psychol Sci 1, 68–87.