Association Between Olfactory Test Data with Multiple Levels of Odor Intensity and Suspected Cognitive Impairment: A Cross-Sectional Study

Abstract

Background:

Olfactory function decline has recently been reported to be associated with a risk of cognitive impairment. Few population-based studies have included younger adults when examining the association between olfactory test data with multiple odor intensities and suspected cognitive impairment.

Objective:

We investigated the association between high-resolution olfactory test data with fewer odors and suspected cognitive impairments. We also examined the differences between older and younger adults in this association.

Methods:

The Japanese version of the Montreal Cognitive Assessment (MoCA-J) was administered to 1,450 participants, with three odor-intensity-level olfactometry using six different odors. Logistic regressions to discriminate suspected cognitive impairment were conducted to examine the association, adjusted for age, sex, education duration, and smoking history. Data were collected from the Program by Tohoku University Tohoku Medical Megabank Organization, with an additional olfactory test conducted between 2019 and 2021.

Results:

We generally observed that the lower the limit of distinguishable odor intensity was, the higher the MoCA-J score was. The combination of spearmint and stuffy socks contributed most to the distinction between suspected and unsuspected cognitive impairment. Furthermore, the association was significant in women aged 60–74 years (adjusted odds ratio 0.881, 95% confidence interval [0.790, 0.983], p = 0.024).

Conclusions:

The results indicate an association between the limit of distinguishable odor intensity and cognitive function. The olfactory test with multiple odor intensity levels using fewer odors may be applicable for the early detection of mild cognitive impairment, especially in older women aged 60–74 years.

Keywords

Alzheimer’s disease cross-sectional studies dementia tests logistic models mental status odds ratio olfactory disorders risk smell

INTRODUCTION

As of 2023, 50 million people worldwide have dementia, and most of them are older adults. The number of people with dementia may reach 152 million by 2050 [1]. Therefore, identifying possible risk factors and predictors is important to improve the early detection of persons at a high risk of developing dementia. For the early detection of dementia, some fundamental studies have used various data, such as blood [2], retinal tomography [3], walking speed [4], and facial images [5].

In recent years, several researchers have shown that impaired olfactory function is associated with a risk of developing cognitive impairment [6 –15] and Alzheimer’s disease (AD) [16 –22]. Specifically, scholars have shown that combining olfactory measures for assessing the risk of conversion to mild cognitive impairment (MCI) or AD is valuable [23] and that olfactory identification is more profoundly impaired in AD patients than in MCI patients [24]. Evidence also exists to indicate the relationship between olfactory impairment and neurodegeneration in the brain [14] and the association between rapid olfactory decline and impaired cognition [25]. Smells are detected by olfactory receptors, which transmit the information to the olfactory cortex via the olfactory bulb. Then, the olfactory cortex processes the information and conveys it to the amygdala and hippocampus, where it is linked to judgments of likes and dislikes as well as memory. In AD, tau, an abnormal protein, accumulates in nerve cells, causing brain damage; however, tau damage spreads over time from the olfactory cortex to the hippocampus, causing olfactory dysfunction before memory impairment appears [26].

Considering these descriptions, when assessing suspected cases of cognitive impairment in clinical settings, conducting olfactory tests that use fewer odors to allow for the high-resolution quantification of cognitive dysfunction could be beneficial; this could reduce the cost of equipment, odor inventory, and burden of odor switching. However, for the olfactory test kits that have been commonly used in previous studies—such as the University of Pennsylvania Smell Identification Test (UPSIT) [27]—a reduction in the number of odors would lead to lower-resolution olfactory test data and, consequently, a lower resolution of the suspected estimation of cognitive dysfunction. One way to achieve a higher resolution with fewer odors is to perform olfactory tests using multiple odor intensities.

Regarding age, few researchers have conducted population-based studies that examine the association between olfactory tests and suspected cognitive impairment in older adults compared to younger adults. Thus, to the best of our knowledge, this is the first population-based study aimed at showing the association between high-resolution olfactory test data with fewer odors and suspected cognitive impairment in the study population (participants aged 31–91 years). This study also examined the differences regarding this association between older and younger adults to identify age groups that may contribute to the early detection of MCI.

MATERIALS AND METHODS

Participants

Since July 2014, in Japan, the Tohoku University Tohoku Medical Megabank Organization has been actively collecting data on cognitive function test, age, sex, education duration, and smoking history from participants of the ongoing program named Health Surveillance of the Brain and Psychological State Program. Alongside the data collection procedures of this program, we conducted olfactory tests from August 27, 2019, to March 30, 2021. To collect olfactory test data, we used our own olfactory testing equipment and the 4-item Japanese pocket smell test, which is the UPSIT’s shorter version A (hereinafter referred to as UPSIT-A) [27]. Participants were men and women aged 20 years or older who were in good physical condition on the day of the health survey. Although the possibility of mild respiratory infections and nasal conditions, such as sinusitis and allergies, in some participants could not be ruled out, we did not consider it a major problem. The exclusion criteria were pregnant women, women suspected of being pregnant, and those with metallic implants who did not have a doctor’s certificate for magnetic resonance imaging safety.

Of the 1,855 people invited to participate in an olfactory test, 1,788 provided consent; however, only 1,450 participants completed the olfactory and cognitive tests and responded to a questionnaire on age, sex, education duration, and smoking history, as shown in Fig. 1. Age, sex, education duration, and smoking history are currently considered to be related to cognition and olfaction.

Fig. 1

Participants’ inclusion flowchart.

Procedures and data collection

In the test using our olfactory testing device, four kinds of pleasant odors (vanillin, cyclotene, γ-undecalactone, and (–)-carvone) and two kinds of unpleasant odors (n-valeric acid and 4-methyl-3-hexenoic acid) were selected among the living odors of stable single compounds. This was done with reference to the T&T Olfactometer (Daiichi Yakuhin Sangyo, Japan), which is the only olfactory test covered by insurance in Japan. For each odor, three odor intensity levels were prepared using triethyl citrate as a dilution solvent, as follows:

Vanilla: vanillin (2, 3, 4)

Wet rag: 4-methyl-3-hexenoic acid (3, 4, 5)

Caramel: cyclotene (2, 3, 4)

Spearmint: (–)-carvone (2, 3, 4)

Stuffy socks: n-valeric acid (3, 4, 5)

Yellow peach: γ-undecalactone (3, 4, 5)

The numbers in parentheses above indicate the odor intensity. Two panelists judged the intensity using a six-point scale ranging from 0 to 5, according to standards from the Japanese Ministry of the Environment and as described herein: 0 = none, 1 = very weak, 2 = weak, 3 = moderate, 4 = strong, and 5 = very strong.

The testing device used to release the odor was the Aroma Shooter^®, produced by Aromajoin Corporation. The Aroma Shooter is a tennis ball-sized device that can be loaded with up to six triangular cartridges, each containing a solid-state scent. An olfactory test system was constructed by connecting six Aroma Shooters to a single-board computer (Raspberry Pi). Aroma cartridges for each odor were obtained from Aromajoin Corporation, and one Aroma Shooter was filled with the same odor at three different intensities. After the lowest-intensity odor was released from the device, participants were instructed to select one of 12 pictures on a computer screen that they felt best matched the sensed odor. Each odor was tested at all three intensities. In addition to the 12 pictures, the option “I do not know” was provided on the screen. In an olfactory test where the correct answer is chosen from a list of alternatives, participants may choose it by chance, even without identifying the odor. In particular, when using multiple odor intensities, scoring a correct answer at a low concentration without considering the results at a higher concentration could sometimes be problematic. For example, when using three levels of odor intensity, if a participant chooses the correct answer at low or medium concentration by chance, a large score is assigned (3 points if hit at a low concentration, 2 points if hit at medium concentration, and so on). To minimize as much as possible the negative impacts of choosing the right answer by accident, we administered tests at all odor concentrations and used the scoring method as demonstrated in Fig. 2. In this method, if a correct answer is given at a certain concentration but an incorrect answer is given at a higher concentration, the correct answer is considered a fluke, and no points are added. T-Score-6 denotes the total score of all six odors.

Fig. 2

Olfactory data scoring in our own testing device. A check mark was given when the picture indicating the released odor was correctly selected, whereas no mark was given in the case of an incorrect answer.

In the UPSIT-A, the four “microencapsulated odorants” (strawberry, chocolate, mint, and smoke) were applied to the separate mounts and then lightly rubbed with cotton swabs by the operator to volatilize them. The participants identified the corresponding odor from four given choices for each mount, and the number of correct answers was used as the UPSIT-A score (0–4 points).

For cognitive function, we used the Japanese version of the Montreal Cognitive Assessment (MoCA-J), which is one of the most commonly used scales for detecting cognitive impairment. We introduced the new variable MoCA-Cognitive Impairment (MoCA-CI), which is denoted as 0 or 1, with 1 indicating a MoCA-J score <26, a clinical cut-off value for suspected MCI [28].

Statistical analyses

The participants were divided, as equally as possible, into four groups in ascending order according to their T-Score-6. To determine the adjustment variables for the association analysis between the olfactory test and suspected cognitive impairment, we conducted the Jonckheere–Terpstra and Cochran–Armitage trend tests on the distribution of continuous and categorical variables, respectively, in the four groups. We tested age, sex, education duration, and smoking history as candidate adjustment variables.

To further characterize the olfactory data obtained with our device, we show a box plot depicting the distribution of T-Score-6 using the UPSIT-A score. We also illustrate a box plot graph expressing the distribution of MoCA-J scores by an olfactory score for each odor.

To identify the combination of odors that contributes most to the distinction between suspected and unsuspected cognitive impairment, the total scores for the respective combinations were calculated for all combinations from the six odors, and the logistic regressions of MoCA-CI were conducted with the items specified in the aforementioned trend tests as adjustment variables. In these analyses, sex was set to 0 and 1 for women and men, respectively, and smoking history was set to 0 and 1 for more than or less than 100 cigarettes smoked in the lifetime, respectively. Using this combination, the association between olfactory test data and suspected cognitive impairment was compared separately for each sex using logistic regressions.

In addition, we compared the association among people aged 45–59, 60–74, and 75–89 years for each sex. To test whether the interaction of age groups and olfactory data is statistically significant for MoCA-CI, we added cross-products of age group and olfactory data in the logistic regression.

All statistical analyses, except for the Jonckheere–Terpstra test, were performed using statistical packages, including Statsmodels and Scipy.stats, in Python 3.7.4. The Jonckheere–Terpstra test was performed using the Clinfun package in R software since it is not provided in Python. In this study, statistical significance was set at a p-value<0.05.

Ethical considerations

The ethical review board of the Tohoku University Tohoku Medical Megabank Organization approved this study, which conforms to the principles of the Declaration of Helsinki [29]. All the participants provided written informed consent for the use of their data for research purposes.

RESULTS

Participants’ characteristics

Table 1 presents the characteristics of the 1,450 participants. Data are expressed as mean±standard deviation (SD) for continuous variables and as numbers and percentages for categorical variables. The mean MoCA-J score was 25.6 (SD = 2.6), indicating that 44.0% of the participants were suspected to have cognitive impairment (MoCA-CI = 1).

Table 1

Characteristics of participants by T-Score-6 quartiles

	ALL	Q1	Q2	Q3	Q4	p
Range of T-Score-6	0–18	0–4	5–7	8–10	11–18
Number of samples	1,450	378	369	394	309
Age, y	63.3±9.8	69.6±8.0	65.1±8.8	60.5±9.0	57.2±8.7	<0.001
Men, n (%)	524 (36.1)	211 (55.8)	156 (42.3)	102 (25.9)	55 (17.8)	<0.001
Education, y	13.5±1.8	13.3±1.9	13.4±1.7	13.6±1.8	13.6±1.7	0.003
Smoking ever, n (%)	544 (37.5)	169 (44.7)	152 (41.2)	128 (32.5)	95 (30.7)	<0.001
MoCA-J	25.6±2.6	24.6±2.7	25.4±2.5	26.0±2.4	26.6±2.1	<0.001
MoCA-CI = 1, n (%)	638(44.0)	220 (58.2)	183 (49.6)	148 (37.6)	87 (28.2)	<0.001

Continuous data are described as mean±standard deviation. Categorical data are described as number (percentage). MoCA-J denotes the Japanese version of the Montreal Cognitive Assessment. MoCA-CI (Cognitive Impairment) takes a value of 1 for a MoCA-J score <26 and 0 for a MoCA-J score ≥26. p-values are derived by Jonckheere–Terpstra and Cochran–Armitage tests for trend on the distribution of continuous and categorical variables, respectively.

The rightmost column in Table 1 presents the p-value for the trend across groups from Q1 to Q4. Age, sex, education duration, smoking history, MoCA-J score, and MoCA-CI significantly increased or decreased. Those who were younger, women, had more years of education, and had no history of smoking were more likely to maintain their olfactory function. Considering these results, we decided to adjust for age, sex, educational duration, and smoking history in the following analysis.

Characteristics of olfactory test data obtained with our devices

Figure 3 shows a box plot of the distribution of T-Score-6 by UPSIT-A score. A monotonically increasing trend was observed. Figure 4 shows a box plot graph depicting the distribution of the MoCA-J score by the olfactory score for each odor. In general, the higher the olfactory score was, the higher the MoCA-J score was. In particular, the first quartile showed a monotonically increasing trend for all odors. For each odor, the maximum MoCA-J score in the olfactory score was 30. The center of the MoCA-J score did not differ significantly among the values in the olfactory score of the wet rag. Stuffy socks had a smaller difference between the first and third quartiles in the MoCA-J score than the otherodors.

Fig. 3

Box plot depicting the distribution of T-Score-6 by UPSIT-A score.

Fig. 4

Box plot depicting the distribution of MoCA-J score by olfactory test score for each odor.

Odor combination that contributes most to the distinction between suspected and unsuspected cognitive impairment

Table 2 presents the top 10 odor combinations with the highest area under the curves (AUCs) in the logistic regressions of MoCA-CI among all combinations from the six odors, listed in increasing order. The combination of spearmint and stuffy socks had the highest AUC (0.7191). The total score of spearmint and stuffy socks was used as the olfactory data in the following results. Hereinafter, this score is referred to as T-Score-2.

Table 2

Top 10 odor combinations with the highest AUCs in logistic regressions of MoCA-CI among all combinations from the six odors

Odor combination	AUC
4,5	0.7191
4,5,6	0.7185
5	0.7177
1,4,5,6	0.7176
3,4,5,6	0.7175
1,4,5	0.7175
3,4,5	0.7174
5,6	0.7172
4,6	0.7171
4	0.7170

The combinations are listed in the order of increasing value of the area under the curve (AUC). Age, sex, education duration, and smoking history were used as adjustment variables. The odor combination numbers correspond to each odor as follows: 1: Vanilla, 2: Wet rag, 3: Caramel, 4: Spearmint, 5: Stuffy socks, and 6: Yellow peaches. MoCA-CI (Cognitive Impairment) has a value of 1 for a MoCA-J score <26 and 0 for a MoCA-J score ≥26. MoCA-J denotes the Japanese version of the Montreal Cognitive Assessment.

Table 3 presents the detailed results of the logistic regression using T-Score-2 compared to the results using UPSIT-A and T-Score-6. T-Score-2 and UPSIT-A were significantly associated with MoCA-CI, especially with a p-value <0.01 for T-Score-2 (T-Score-2: AOR 0.898, 95% confidence interval [0.838, 0.963], p = 0.002; UPSIT-A: AOR 0.806, 95% confidence interval [0.652, 0.995], p = 0.044). T-Score-6 showed no significant relationship with MoCA-CI (AOR 0.967, 95% confidence interval [0.933, 1.001], p = 0.060).

Table 3

Logistic regression results of MoCA-CI with T-Score-2, UPSIT-A, and T-Score-6, respectively

Olfactory score	Explanatory and adjusted variables	AOR
		Estimate	[95% CI]	p
T-Score-2	Olfactory score	0.898	[0.838, 0.963]	0.002
	Age	1.059	[1.045, 1.074]	<0.001
	Sex	2.101	[1.581, 2.790]	<0.001
	Education duration	0.958	[0.899, 1.022]	0.193
	Smoking history	0.888	[0.674, 1.170]	0.399
UPSIT-A	Olfactory score	0.806	[0.652, 0.995]	0.044
	Age	1.063	[1.049, 1.077]	<0.001
	Sex	2.166	[1.634, 2.872]	<0.001
	Education duration	0.953	[0.894, 1.016]	0.141
	Smoking history	0.898	[0.682, 1.182]	0.443
T-Score-6	Olfactory score	0.967	[0.933, 1.001]	0.060
	Age	1.060	[1.046, 1.075]	<0.001
	Sex	2.078	[1.557, 2.773]	<0.001
	Education duration	0.957	[0.898, 1.020]	0.180
	Smoking history	0.905	[0.688, 1.192]	0.479

MoCA-CI (Cognitive Impairment) takes a value of 1 for a MoCA-J score <26 and 0 for a MoCA-J score ≥26. MoCA-J denotes the Japanese version of the Montreal Cognitive Assessment. T-Score-2 indicates the total score of spearmint and stuffy socks. T-Score-6 indicates the total score of vanilla, wet rag, caramel, spearmint, stuffy socks, and yellow peach. UPSIT-A is the shorter version A of the UPSIT. UPSIT, University of Pennsylvania Smell Identification Test; AOR, Adjusted Odds Ratio; 95% CI, 95% confidence interval. The AOR of 0.898 for T-Score-2 means the prevalence ratio of suspected cognitive impairment per one-point increase of T-Score-2.

Table 4 presents the logistic regression results of the MoCA-CI in men and women. The olfactory test data were significantly related to MoCA-CI in women (AOR 0.899, 95% confidence interval [0.825, 0.978], p = 0.014) but not in men (AOR 0.899, 95% confidence interval [0.799, 1.012], p = 0.077).

Table 4

Logistic regression results of MoCA-CI for each sex

Explanatory and adjusted variables	AOR
	Men			Women
	Estimate	[95% CI]	p	Estimate	[95% CI]	p
T-Score-2	0.899	[0.799, 1.012]	0.077	0.899	[0.825, 0.978]	0.014
Age	1.058	[1.037, 1.080]	<0.001	1.065	[1.046, 1.083]	<0.001
Education duration	0.918	[0.839, 1.005]	0.063	1.006	[0.918, 1.103]	0.902
Smoking history	1.294	[0.851, 1.966]	0.228	0.646	[0.446, 0.937]	0.021

MoCA-CI (Cognitive Impairment) takes a value of 1 for a MoCA-J score <26 and 0 for a MoCA-J score ≥26. MoCA-J stands for the Japanese version of the Montreal Cognitive Assessment. Numbers of samples: MoCA-CI = 0 and 1 are 211 and 313, respectively, for men and 601 and 325, respectively, for women. T-Score-2 indicates the total score of spearmint and stuffy socks. AOR, Adjusted Odds Ratio; 95% CI, 95% confidence interval. The AOR value of 0.899 for T-Score-2 (Men) means the prevalence ratio of suspected cognitive impairment per one-point increase of T-Score-2.

Age groups for which olfactory tests might be more reliably associated with suspected cognitive impairment

Table 5 presents the results of the logistic regression using MoCA-CI in men and women aged 45–59, 60–74, and 75–89 years. The olfactory test data were significantly related to MoCA-CI in women aged 60–74 years (AOR 0.881, 95% confidence interval [0.790, 0.983], p = 0.024) but not in women aged 45–59 (AOR 0.901, 95% confidence interval [0.769, 1.055], p = 0.196) and 75–89 years (AOR 0.966, 95% confidence interval [0.656, 1.423], p = 0.863). The olfactory test data were not significantly related to MoCA-CI in men of all age groups. No significant interactions between age groups and olfactory data for MOCA-CI were observed as shown inTable 6.

Table 5

Logistic regression results of the comparison between people of different age groups by sex

Age group	Explanatory and adjusted variables	AOR
		Men			Women
		Estimate	[95% CI]	p	Estimate	[95% CI]	p
45≤age < 60 years	T-Score-2	0.910	[0.720, 1.149]	0.428	0.901	[0.769, 1.055]	0.196
(Numbers of samples: MoCA-CI = 0 and 1 are 73 and 44 for men, and 268 and 81 for women, respectively.)	Age	1.102	[1.007, 1.204]	0.034	1.076	[1.016, 1.140]	0.012
	Education duration	1.185	[0.968, 1.451]	0.101	1.033	[0.864, 1.235]	0.724
	Smoking history	1.291	[0.564, 2.956]	0.545	0.647	[0.363, 1.153]	0.139
60≤age < 75 years	T-Score-2	0.885	[0.748, 1.048]	0.156	0.881	[0.790, 0.983]	0.024
(Numbers of samples: MoCA-CI = 0 and 1 are 98 and 197 for men, and 300 and 203 for women, respectively.)	Age	1.116	[1.042, 1.196]	0.002	1.078	[1.028, 1.130]	0.002
	Education duration	0.768	[0.674, 0.875]	<0.001	0.953	[0.848, 1.071]	0.417
	Smoking history	1.578	[0.837, 2.974]	0.158	0.586	[0.348, 0.987]	0.044
75≤age < 90 years	T-Score-2	0.951	[0.702, 1.289]	0.747	0.966	[0.656, 1.423]	0.863
(Numbers of samples: MoCA-CI = 0 and 1 are 31 and 69 for men, and 23 and 38 for women, respectively.)	Age	0.895	[0.759, 1.054]	0.183	1.240	[0.977, 1.573]	0.077
	Education duration	1.022	[0.833, 1.252]	0.838	1.643	[0.996, 2.707]	0.052
	Smoking history	1.123	[0.430, 2.936]	0.813	0.375	[0.034, 4.141]	0.423

MoCA-CI (Cognitive Impairment) takes a value of 1 for a MoCA-J score <26 and 0 for a MoCA-J score ≥26. MoCA-J stands for the Japanese version of the Montreal Cognitive Assessment. T-Score-2 indicates the total score of spearmint and stuffy socks. AOR, Adjusted Odds Ratio; 95% CI, 95% confidence interval. The AOR value of 0.881 for T-Score-2 (women aged 60–75) means the prevalence ratio of suspected cognitive impairment per one-point increase of T-Score-2.

Table 6

Analysis of age group differences in the association between age groups and olfactory test (Logistic regression of MoCA-CI with interaction terms of age groups and olfactory test)

Explanatory and adjusted variables	AOR
	Men			Women
	Estimate	[95% CI]	p	Estimate	[95% CI]	p
T-Score-2	0.719	[0.510, 1.013]	0.059	0.869	[0.662, 1.140]	0.310
Age group	1.382	[0.833, 2.291]	<0.001	2.121	[1.330, 3.380]	<0.001
Age group * T-Score-2	1.106	[0.929, 1.315]	0.258	1.000	[0.865, 1.157]	0.998
Education duration	0.916	[0.839, 1.002]	0.055	0.976	[0.890, 1.069]	0.605
Smoking history	1.273	[0.839, 1.931]	0.257	0.673	[0.466, 0.972]	0.035

MoCA-CI (Cognitive Impairment) takes a value of 1 for a MoCA-J score of <26 and 0 for a MoCA-J score of ≥26. MoCA-J stands for the Japanese version of the Montreal Cognitive Assessment. T-Score-2 indicates the total score of spearmint and stuffy socks. AOR, Adjusted Odds Ratio; 95% CI, 95% confidence interval. The AOR value of 0.719 for T-Score-2 (Men) means the prevalence ratio of suspected cognitive impairment per one-point increase of T-Score-2. Age group takes a value of 1, 2, and 3 for people aged 45–59, 60–74, and 75–89 years, respectively. Age Group * T-Score-2 indicates age group value multiplied by the T-Score-2 value.

DISCUSSION

This study’s results are consistent with those of existing studies showing that an association between olfactory test data and cognitive function data could exist [6 –15]. The study from Yahiaoui-Doktor et al. [11], which reported the above association in a population including young and older people, especially supports our results, although multiple odor intensities were not used in the cited study. The scores on the horizontal axis in Fig. 4 indicate the limit of distinguishable odor intensity. High scores indicate low olfactory limits. The lower the odor intensity limit, the higher the MoCA-J scores, suggesting an association between the limit of distinguishable odor intensity and MoCA-J scores.

Furthermore, a previous study has shown that the association between olfactory function and cognitive performance was greater for women than for men [30]. This finding roughly supports our result in which the olfactory test data were significantly related to the MoCA-CI in women but not in men.

Previous research has shown that olfactory impairments occur more than 15 years before the onset of cognitive dysfunction [14]. Although there were no statistically significant differences among age groups, the olfactory data was significantly related to MoCA-CI in women aged 60–74 years but not in those aged 45–59 years. One possible reason for these findings is that the latter group, compared to the former group, included more people whose olfactory function could have only recently started to decline, not giving enough time for this to reflect in their cognition. The reason for the insignificance of the olfactory data in women aged 75–89 years could be that this group had few people with high olfactory test scores (percentage of people whose olfactory scores were 5 or 6:33.2% for those aged 45–59, 19.7% for 60–74, 3.3% for 75–89 years). In the Heinz Nixdorf Recall Study, an association between olfactory function and cognitive performance was found in participants aged 65–74 years but not in those aged 55–64 and 75–86 years [30], which supports our results.

As shown in Fig. 4, the maximum MoCA-J score was 30 for every olfactory score, and the first quartile of the MoCA-J monotonically increased with the olfactory score. In other words, the worse the olfactory score, the greater the difference between the maximum and first quartiles. This phenomenon could be because of the aforementioned idea that it takes more than several years for olfactory score decline to reflect any cognitive decline. This means that participants whose olfactory function worsened just before their olfactory tests would show little or no cognitive decline (as evaluated by the MoCA-J), while the participants whose olfaction had declined long ago could display a greater cognitive decline in line with the degree of olfactory decline. Unfortunately, this could not be verified in this study as we did not assess when each participant’s olfactory function had declined.

The smell of spearmint can stimulate both the trigeminal and the olfactory nerves in the nasal cavity. A previous study [31] has shown that damage to the trigeminal nerve affects cognitive function. The results obtained in this study, using olfactory test data containing spearmint, may encompass a reduction in MoCA-J scores due to trigeminal nervereduction.

In all logistic regression results, the AORs of sex were >1.0 (Table 3). These results indicate that men are more susceptible to suspected cognitive impairment than women. Roberts et al. reported that the incidence of MCI was generally higher in men than in women [32]. Our results were consistent with this finding.

This study had some limitations. In recent years, many types of genetic olfactory receptors have been activated in various patterns by odorants, and their variations have resulted in individual differences in odor sensitivity. For example, the olfactory receptor OR5A1 is associated with changes in amino acid sequence and odor perception [33]. Owing to genetic differences, some of our participants might have found that some of the odors were either easy or difficult to identify. In addition, our sample might have included patients with non-amnestic MCI. Previous studies have reported no relationship between non-amnestic MCI and olfactory tests [17]. Moreover, some viruses, such as the common cold and influenza viruses, can damage olfactory cells, and in some cases, people may not recover their smell for months or even a year or more after having such viruses [34, 35]. Our study might also have included participants with temporary olfactory impairment or organic diseases, such as sinusitis and nasal allergies. However, these conditions were not diagnosed, so whether they affect the distinction is not clear. The factors mentioned above might have negatively impacted the results of the analyses on the association between the olfactory test and the suspected cognitive impairment. While the combination of spearmint and socks showed a strong association with suspected cognitive dysfunction, further research is needed to determine the medical reasons for this association.

In summary, this study is the first to show an association between high-resolution olfactory test data with fewer odors and suspected cognitive impairment and examine differences in this association between older and younger adults. The results imply the existence of an association between the limit of distinguishable odor intensity and cognitive function. This study also demonstrates that the introduction of multiple odor intensities provides us with olfactory data that is more significantly associated with suspected cognitive impairment and has higher resolution with fewer odors, compared to the UPSIT-A. In addition, the findings showcase that the association between olfactory data and suspected cognitive impairment in the MoCA-J criteria was statistically significant in women aged 60–74 years and not significant in women aged 45–59 and 75–89 years. Thus, the simple olfactory test used in this study could be effectively applied for the early detection of MCI in the clinical field, especially among younger older women aged 60–74 years with a lower prevalence of dementia compared to their older counterparts.

Based on the MoCA-J criteria, more than half of the participants in this study were cognitively normal. Existing studies have shown that among participants with average cognition, those with poor odor identification are more likely to develop MCI than those with good odor identification [22]. Correspondingly, we plan to confirm the association between olfactory test data and multiple levels of odor intensity as a precursor to cognitive decline by conducting cognitive function tests on the same participants several years later. We also intend to consider the risk prediction method of cognitive impairment by using olfactory data together with genotyping and magnetic resonance imaging of the brain in future research.

Footnotes

ACKNOWLEDGMENTS

We would like to thank and acknowledge the members of Tohoku University Tohoku Medical Megabank Organization for supporting our study and allowing us to use their equipment. We would like to thank Naoko Tsutsumi for performing the olfactory tests.

FUNDING

Toyota Group Companies provided research funds to Toyota Central R&D Labs., Inc. These funders provided support, in the form of salaries, to authors SS, TI, MT, and KH. Except for olfactory test data, the data in this study were collected as part of the Health Surveillance of the Brain and Psychological State Program by Tohoku University Tohoku Medical Megabank Organization, which was subsidized by the government [Grant number: JP21tm0124005]. Toyota Central R&D Labs., Inc. also provided a collaborative study grant to Tohoku University.

CONFLICT OF INTEREST

Beyond their role as funders (see FUNDING), Toyota Group Companies and Toyota Central R&D Labs., Inc., did not play any additional role in the design and conduct of the study; in data collection, analysis, and interpretation; or in the preparation of the manuscript. The olfactory test data were collected through a collaborative study between Toyota Central R&D Labs., Inc., and Tohoku University. All other data in this study were collected as part of the Health Surveillance of the Brain and Psychological State Program by Tohoku University Tohoku Medical Megabank Organization, which was subsidized by the government. Toyota Central R&D Labs., Inc and Tohoku University have a pending Japanese patent (Japanese Patent Application No. 2022-113409).

DATA AVAILABILITY

The data that support the findings of this study are available from Tohoku Medical Megabank Organization, but restrictions apply to the availability of these data, which were used under permission for the current study and are not publicly available. Data are also available from the corresponding author upon reasonable request and the permission of Tohoku Medical Megabank Organization.

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

References

Livingston

, Huntley

, Sommerlad

, Ames

, Ballard

, Banerjee

, Brayne

, Burns

, Cohen

, Cooper

, Costafreda

, Dias

, Fox

, Gitlin

, Howard

, Kales

, Kivimaki

, Larson

, Ogunniyi

, Orgeta

, Ritchie

, Rockwood

, Sampson

, Samus

, Schneider

, Selbæk

, Teri

, Mukadam

(2020) Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. Lancet 396, 413–446.

Ohara

, Hata

, Tanaka

, Honda

, Yamakage

, Yoshida

, Inoue

, Hirakawa

, Kusakabe

, Shibata

, Teraoka

, Kitazono

, Kanba

, Satoh

, Ninomiya

(2019) Serum, soluble triggering receptor expressed on myeloid cells 2 as a biomarker for incident dementia: The Hisayama Study. Ann Neurol 85, 47–58.

Ito

, Sasaki

, Takahashi

, Nozaki

, Matsuguma

, Motomura

, Ui

, Shikimoto

, Kawasaki

, Yuki

, Sawada

, Mimura

, Tsubota

, Tsugane

(2020) Quantitative assessment of the retina using OCT and associations with cognitive function. Ophthalmology 127, 107–118.

Lipnicki

, Crawford

, Kochan

, Trollor

, Draper

, Reppermund

, Maston

, Mather

, Brodaty

, Sachdev

(2017) Risk factors for mild cognitive impairment, dementia and mortality: The Sydney memory and ageing study. J Am Med Dir Assoc 18, 388–395.

Umeda

, Kameyama

, Kojima

, Ishii

, Kidana

, Yakabe

, Ishii

, Urano

, Ogawa

, Akishita

(2020) Cognitive function has a stronger correlation with perceived age than with chronological age. Geriatr Gerontol Int 20, 779–784.

Wilson

, Schneider

, Arnold

, Tang

, Boyle

, Bennett

(2007) Olfactory identification and incidence of mild cognitive impairment in older age. Arch Gen Psychiatry 64, 802–888.

Schubert

, Carmichael

, Murphy

, Klein

, Cruickshanks

(2008) Olfaction and the 5-year incidence of cognitive impairment in an epidemiological study of older adults. J Am Geriatr Soc 56, 1517–1521.

Baba

, Kikuchi

, Hirayama

, Nishio

, Hosokai

, Kanno

, Hasegawa

, Sugeno

, Konno

, Suzuki

, Takahashi

, Fukuda

, Aoki

, Itoyama

, Mori

, Takeda

(2012) Severe olfactory dysfunction is a prodromal symptom of dementia associated with Parkinson’s disease: A 3-year longitudinal study. Brain 135, 161–169.

Park

, Kim

, Cha

(2021) The association between olfactory function and cognitive impairment in older persons with cognitive impairments: A cross-sectional study. Healthcare (Basel) 9, 399.

10.

Cha

, Kim

, Son

(2021) Associations between cognitive function, depression, and olfactory function in elderly people with dementia in Korea. Front Aging Neurosci 13, 799897.

11.

Yahiaoui-Doktor

, Luck

, Riedel-Heller

, Loeffler

, Wirkner

, Engel

(2019) Olfactory function is associated with cognitive performance: Results from the population-based LIFE-Adult-Study. Alzheimers Res Ther 11, 43.

12.

Sanke

, Mita

, Yoshii

, Someya

, Yamashiro

, Shimizu

, Ohmura

, Onuma

, Watada

(2021) Olfactory dysfunction predicts the development of dementia in older patients with type 2 diabetes. Diabetes Res Clin Pract 174, 108740.

13.

Growdon

, Schultz

, Dagley

, Amariglio

, Hedden

, Rentz

, Johnson

, Sperling

, Albers

, Marshall

(2015) Odor identification and Alzheimer disease biomarkers in clinically normal elderly. Neurology 84, 2153–2160.

14.

Dintica

, Marseglia

, Rizzuto

, Wang

, Seubert

, Arfanakis

, Bennett

, Xu

(2019) Impaired olfaction is associated with cognitive decline and neurodegeneration in the brain. Neurology 92, 700–709.

15.

Suzuki

, Sugiura

, Nakashima

, Teranishi

, Shimono

, Murotani

, Sakurai

, Uchida

, Saji

(2022) Cognitive impairment is correlated with olfactory identification deficits in older Japanese adults: A cross-sectional study using objective and subjective olfactory measures. Geriatr Gerontol Int 22, 924–929.

16.

Tabert

, Liu

, Doty

, Serby

, Zamora

, Pelton

, Marder

, Albers

, Stern

, Devanand

(2005) A 10-item, smell identification scale related to risk for Alzheimer’s disease. Ann Neurol 58, 155–160.

17.

Roberts

, Christianson

, Kremers

, Mielke

, Machulda

, Vassilaki

, Alhurani

, Geda

, Knopman

, Petersen

(2016) Association between olfactory dysfunction and amnestic mild cognitive impairment and Alzheimer disease dementia. JAMA Neurol 73, 93–101.

18.

Devanand

, Liu

, Tabert

, Pradhaban

, Cuasay

, Bell

, Leon

MJD

, Doty

, Stern

, Pelton

(2008) Combining early markers strongly predicts conversion from mild cognitive impairment to Alzheimer’s disease. Biol Psychiatry 64, 871–879.

19.

Jimbo

, Inoue

, Taniguchi

, Urakami

(2011) Specific feature of olfactory dysfunction with Alzheimer’s disease inspected by the odor stick identification test. Psychogeriatrics 11, 196–204.

20.

Mihoko

, Hyuma

, Takehiko

, Uemura

, Tsutsumimoto

, Miyaguchi

, Shimada

(2013) Olfactory identification and cognitive performance in community-dwelling older adults with mild cognitive impairment. Chem Senses 39, 39–46.

21.

Olofsson

, Ekström

, Larsson

, Nordin

(2021) Olfaction and aging: A review of the current state of research and future directions. i-Perception 12, 20416695211020331.

22.

Murphy

(2019) Olfactory and other sensory impairments in Alzheimer disease. Nat Rev Neurol 15, 11–24.

23.

Wheeler

, Murphy

(2021) Olfactory measures as predictors of conversion to mild cognitive impairment and Alzheimer’s disease. Brain Sci 11, 1391.

24.

Jung

, Shin

, Lee

(2019) Olfactory function in mild cognitive impairment and Alzheimer’s disease: A meta-analysis. Laryngoscope 129, 362–369.

25.

Pacyna

, Han

, Wroblewski

, McClintock

, Pinto

(2023) Rapid olfactory decline during aging predicts dementia and GMV loss in AD brain regions. Alzheimers Dement 19, 1479–1490.

26.

Braak

, Thal

, Ghebremedhin

, Del Tredici

(2011) Stages of the pathologic process in Alzheimer disease: Age categories from 1 to 100 years. J Neuropathol Exp Neurol 70, 960–969.

27.

Doty

, Shaman

, Kimmelman

, Dann

(1984) University of Pennsylvania smell identification test: A rapid quantitative olfactory function test for the clinic. Laryngoscope 94, 176–178.

28.

Fujiwara

, Suzuki

, Yasunaga

, Sugiyama

, Ijuin

, Sakuma

, Inagaki

, Iwasa

, Ura

, Yatomi

, Ishii

, Tokumaru

, Homma

, Nasreddine

, Shinkai

(2010) Brief screening tool for mild cognitive impairment in older Japanese: Validation of the Japanese version of the Montreal Cognitive Assessment. Geriatr Gerontol Int 10, 225–232.

29.

World Medical Association (2013) WMA Declaration of Helsinki – ethical principles for medical research involving human subjects. JAMA 310, 2191–2194.

30.

Tebrügge

, Winkler

, Gerards

, Weimar

, Moebus

, Jöckel

, Erbel

, Jokisch

, Heinz Nixdorf Recall Study Investigative Group (2018) Olfactory function is associated with cognitive performance: Results of the Heinz Nixdorf Recall Study. J Alzheimers Dis 63, 319–329.

31.

Cicco

, Fantozzi

MPT

, Cataldo

, Barresi

, Bruschini

, Faraguna

, Manzoni

(2018) Trigeminal, visceral and vestibular inputs may improve cognitive functions by acting through the locus coeruleus and the ascending reticular activating system: A new hypothesis. Front Neuroanat 11, 130.

32.

Roberts

, Geda

, Knopman

, Cha

, Pankratz

, Boeve

, Tangalos

, Ivnik

, Rocca

, Petersen

(2012) The incidence of MCI differs by subtype and is higher in men: The Mayo clinic study of aging. Neurology 78, 342–351.

33.

Jaeger

, McRae

, Bava

, Beresford

, Hunter

, Jia

, Chheang

, Jin

, Peng

, Gamble

, Atkinson

, Axten

, Paisley

, Tooman

, Pineau

, Rouse

, Newcomb

(2013) A Mendelian trait for olfactory sensitivity affects odor experience and food selection. Curr Biol 23, 1601–1605.

34.

de Haro-Licer

, Roura-Moreno

, Vizitiu

, González-Fernández

, González-Ares

(2013) Long term serious olfactory loss in colds and/or flu. Acta Otorrinolaringol Esp 64, 331–338.

35.

Welge

, Wolfensberger

(2006) Olfactory disorders following upper respiratory tract infections. Adv Oto-Rhino-Laryngol 63, 1.