Mastication and Swallowing Times Associated with Eating Satisfaction Among Community-Dwelling Older People

Abstract

Background: The time from mastication to swallowing is used as an indicator of masticatory and swallowing functions. However, there have been no reports on reasonable eating times associated with eating satisfaction. Clarifying the reasonable time for eating and drinking to achieve eating satisfaction will indicate the level of mastication and swallowing functions that contributes to maintaining an individual’s quality of life. Objective: This study aimed to determine the time from food intake to the end of swallowing that is associated with eating satisfaction. Methods: A cross-sectional survey of 437 community-dwelling older people was conducted. Mastication and swallowing times (SST-MST) were measured using the Saku-Saku Test with a 2 g rice cracker. Food intake difficulty and eating satisfaction were evaluated using a questionnaire. The association between the SST-MST and the difficulty in eating food or eating satisfaction was assessed by sensitivity, specificity, Youden index, sensitivity–specificity ratio, positive likelihood ratio, negative likelihood ratio, odds ratio, and 95% confidence interval. Results: Most indices indicated that a cutoff point of 25 s on the SST-MST was associated with not having difficulty in eating food items. Moreover, when the SST-MST cutoff was set to 25 s, all indices showed a favorable association with eating satisfaction. Conclusions: In community-dwelling older people in their 70s and 80s, an SST-MST of about 25 s for 2 g of rice crackers was modestly associated with eating satisfaction.

1. Introduction

In general, with aging, there is a reduction in masticatory muscle strength and increase in tooth loss, resulting in a decline in masticatory ability and an extended masticatory time [1]. In addition, weak bolus propulsion due to decreased tongue muscle strength and delayed swallowing reflex responses are associated with dysphagia [2]. These age-related phenomena interfere with nutritional intake in older people [3] and contribute to physical decline, such as sarcopenia and frailty. Decreased masticatory and swallowing functions lead to not only decreased physical function but also decreased psychological satisfaction [4]. Furthermore, these functional declines are associated with narrower food choices [5,6].

Although mastication and swallowing together constitute the eating function, separate evaluation methods have been devised to assess the two functions. Masticatory function has been evaluated using sieving methods, focusing on finely pulverizing food [7]. Swallowing function is analyzed using imaging evaluations such as video-fluorographic examinations [8]. Observing the sequence of eating functions separately for mastication and swallowing has contributed to a more detailed understanding of each function. Therefore, these elaborate observations have led to an advanced understanding that mastication and swallowing proceed in a complex manner [9].

Several studies have attempted to evaluate the multiple complex functions from mastication to swallowing [10,11,12,13,14]. In these studies, one of the indices used is the time from food intake to the end of swallowing. However, there are no external criteria for evaluating these time indicators. Although representative values for the population have been identified, desirable values have not been demonstrated. Identifying an evidence-based reference range for the time elapsing between food intake and the completion of swallowing will provide a practical indicator of masticatory and swallowing function that supports quality of life. Accordingly, this study aimed to determine the duration from food intake to end of swallowing, which is one of the indicators that is associated with eating satisfaction in community-dwelling older people.

2. Materials and Methods

2.1. Study Participants

The study included 437 community-dwelling individuals (212 men, 225 women; age range 70–99 years, mean age 80.1 ± 4.9 years) who visited one of the forty-one dental clinics included in the study from the Chubu region of Japan (

Table 1. Characteristics of the study participants.

		n	%
Total		437	100.0
Age (years)	70–79	228	52.2
	80–89	198	45.3
	≥90	11	2.5
Sex	Male	212	48.5
	Female	225	51.5
Number of natural teeth	0	16	3.7
	1–9	54	12.4
	10–19	102	23.3
	≥20	265	60.6
Number of natural and prosthetic teeth	10–19	4	0.9
	≥20	433	99.1
Use of dentures	Using	266	60.9
	Having but not using	12	2.7
	Not having	159	36.4
Satisfaction with eating	1: Satisfied	198	45.3
	2	106	24.3
	3	92	21.1
	4	24	5.5
	5	15	3.4
	6: Dissatisfied	2	0.5

). The inclusion criteria were as follows: community-dwelling, independent, age ≥ 70 years and those who provided informed consent to participate in the study. The exclusion criteria were as follows: hearing or language impairment that could affect the interview or physical or mental disability that could interfere with the oral examination. Written informed consent was obtained from all participants.

2.2. Data Collection

Mastication and swallowing times were measured using rice crackers (HappyTurn, Kameda Seika Co., Ltd., Niigata, Japan). Participants, while seated, were instructed to eat half a piece of rice cracker (2 g) at their usual pace and swallow when ready. This method, known as the Saku-Saku Test (SST), was developed for evaluating masticatory function in patients with dysphagia [15]. The mastication and swallowing times (SST-MST) were measured visually by a dentist. The measurement started when the rice cracker was placed in the mouth. The dentist counted the number of seconds on a digital stopwatch until movement of the thyroid cartilage accompanying the swallowing motion was no longer observed.

Food intake difficulty and eating satisfaction were evaluated using a questionnaire. Participants were shown a list of 16 food items and asked which items they found difficult to eat. Subjective eating satisfaction was measured using a 6-point face scale scored from satisfied to dissatisfied. The number of natural and prosthetic teeth and the use of dentures were examined by a dentist.

These questions and measures were included in an oral health survey of community-dwelling older people conducted by the local government and dental associations. This study was a secondary analysis of the survey data and was approved by the Ethics Committee of the Japanese Red Cross Toyota College of Nursing (project no. 1910). Clinical examinations/questionnaires were collected between August 2020 and March 2021; this secondary analysis was performed after anonymization.

2.3. Statistical Analysis

Participants were classified into four groups based on the SST-MST and food items that they found difficult to eat. The SST-MST was divided into two groups using a set cutoff value in seconds. The number of food items participants found difficult to eat was categorized as 0–1 vs. ≥2.

Sensitivity, specificity, Youden index [16], sensitivity–specificity ratio (SSR), positive likelihood ratio, and negative likelihood ratio were calculated based on the number of participants divided into four groups using two variables. Any threshold that produced a negative Youden index (J < 0) was classified as having no diagnostic value. The SSR was obtained by dividing the lower value of sensitivity and specificity by the higher value (SSR = min(Sens, Spec)/max(Sens, Spec)). An SSR value close to 1 (0 ≤ SSR ≤ 1) indicates a good balance between sensitivity and specificity. Additionally, the odds ratios (diagnostic odds ratio: DOR) and 95% confidence intervals (CIs) for having difficulty eating ≤1 food items were calculated when the SST-MST was below the cutoff value. This study prioritized the Youden index and SSR to balance sensitivity and specificity because its purpose was to propose a pragmatic reference rather than to develop a diagnostic test. A cutoff was regarded as acceptable when Youden index ≥ 0.10 and SSR ≥ 0.70 (i.e., the lower of sensitivity or specificity was at least 70% of the higher). These thresholds were chosen a priori to obtain a reasonably balanced diagnostic performance while keeping the procedure simple for clinical use.

The variables of subjective eating satisfaction were evaluated on a 6-grade scale. Grades 1 or 2 indicated that participants were satisfied, whereas grades 3 to 6 indicated that they were dissatisfied. The same analysis was performed among the variables of satisfaction and SST-MST. The SST-MST cutoff was considered in increments of 5 s, from 10 to 60 s, to determine the best cutoff value associated with difficulty in eating foods and eating satisfaction.

Analyses were performed using SPSS Statistics 29 (IBM, Armonk, New York, USA).

3. Results

3.1. Characteristics of the Study Participants

Of the total number of participants, approximately 50% were 70–79 years old and the remaining were over 80 years old (Table 1). Of the participants, 60.6% had more than 20 natural teeth and 99.1% had more than 20 teeth, including natural and prosthetic teeth. Dentures were used by 60.9% of the participants.

Using a six-level eating-satisfaction scale, 69.6% of participants (grades 1–2) were classified as satisfied and 30.4% (grades 3–6) as dissatisfied.

3.2. Difficulty in Eating Foods

According to the participants, vinegared octopus (25.6%) was the most difficult food to eat (

Table 2. List of 16 food items with corresponding percentage of participants who had difficulty in eating them.

Food Item	n	%
Dried potato	46	10.5
Pickled radish	95	21.7
Vinegared octopus	112	25.6
Rice cake	60	13.7
Lotus root	42	9.6
Grilled pork	24	5.5
Chinese cabbage pickles	28	6.4
Apple	32	7.3
Minced fish	6	1.4
Bread	10	2.3
Konjac food	31	7.1
Banana	4	0.9
Wafer	15	3.4
Boiled egg	15	3.4
Dried laver seaweed	29	6.6
Undaria pinnatifida	17	3.9

), followed by pickled radish (21.7%) and rice cake (13.7%). Of the participants, 59.3% answered that they could eat all 16 foods without difficulty (

Table 3. Distribution of participants who had difficulty in eating according to number of foods among the 16 food items.

Number of Food Items Difficult to Eat	n	%
0	259	59.3
1	53	12.1
2	36	8.2
3	40	9.2
4	12	2.7
5	11	2.5
6	6	1.4
7	4	0.9
8	5	1.1
9	2	0.5
10	6	1.4
11	2	0.5
12	0	0.0
13	0	0.0
14	1	0.2
15	0	0.0
16	0	0.0

), whereas 28.6% had difficulty eating ≥2 of the 16 listed food items.

3.3. Association Between SST-MST and Difficulty in Eating Foods

The Youden index, SSR, and positive likelihood ratio indicated that an SST-MST of ≤25 s was associated with difficulty in eating ≤1 food item (

Table 4. Association between SST-MST and difficulty in eating foods.

	Number of Participants with Difficulty in Eating Food Item(s)		Sens (%)	Spec (%)				95% CI
SST-MST (s)	≤1	≥2			YI	SSR	OR	LL	UL		+LR	−LR
≤10	19	10	6.1	92.0	ndv	0.07	0.7	0.3	1.7	ns	0.8	1.0
>10	293	115
≤15	53	24	17.0	80.8	ndv	0.21	0.9	0.5	1.5	ns	0.9	1.0
>15	259	101
≤20	126	38	40.4	69.6	0.10	0.58	1.6	1.0	2.4	ns	1.3	0.9
>20	186	87
≤25	206	63	66.0	49.6	0.16	0.75	1.9	1.3	2.9	*	1.3	0.7
>25	106	62
≤30	246	84	78.8	32.8	0.12	0.42	1.8	1.1	2.9	*	1.2	0.6
>30	66	41
≤35	276	95	88.5	24.0	0.12	0.27	2.4	1.4	4.1	*	1.2	0.5
>35	36	30
≤40	290	106	92.9	15.2	0.08	0.16	2.4	1.2	4.5	*	1.1	0.5
>40	22	19
≤45	301	114	96.5	8.8	0.05	0.09	2.6	1.1	6.3	*	1.1	0.4
>45	11	11
≤50	303	116	97.1	7.2	0.04	0.07	2.6	1.0	6.7	*	1.0	0.4
>50	9	9
≤55	304	121	97.4	3.2	0.01	0.03	1.3	0.4	4.2	ns	1.0	0.8
>55	8	4
≤60	307	122	98.4	2.4	0.01	0.02	1.5	0.4	6.4	ns	1.0	0.7
>60	5	3

ndv: no diagnostic value, * p <0.05, ns: not significant. SST-MST: Saku-Saku Test—mastication and swallowing times; Sens: sensitivity; Spec: specificity; YI: Youden index; SSR: sensitivity specificity ratio (SSR = min(Sens, Spec)/max(Sens, Spec)); OR: odds ratio (=diagnostic odds ratio [DOR]); CI: confidence interval; LL: lower limit; UL: upper limit; +LR: positive likelihood ratio; −LR: negative likelihood ratio.

). The Youden index was 0.16, the SSR was 0.75, and the positive likelihood ratio was 1.3. In contrast, the negative likelihood ratio and odds ratio (=DOR) supported a cutoff value of 45 s. The negative likelihood ratio was 0.4, and the odds ratio was 2.6, with a 95% CI: 1.1–6.3.

Although the 45 s threshold yielded very high sensitivity (96.5%) and a moderately low negative likelihood ratio (0.4), its specificity was only 8.8%, and the SSR fell to 0.1, indicating a pronounced imbalance. According to the predefined criteria, 25 s was therefore adopted as the pragmatic cutoff.

3.4. Association Between SST-MST and Eating Satisfaction

All indices showed a modest association with eating satisfaction when the SST-MST was set at 25 s as the cutoff (

Table 5. Association between SST-MST and eating satisfaction.

	Number of Participants Satisfaction with Eating		Sens (%)	Spec (%)				95% CI
SST-MST (s)	Satisfied	Dissatisfied			YI	SSR	OR	LL	UL		+LR	−LR
≤10	17	12	5.6	91.0	ndv	0.06	0.6	0.3	1.3	ns	0.6	1.0
>10	287	121
≤15	51	26	16.8	80.5	ndv	0.21	0.8	0.5	1.4	ns	0.9	1.0
>15	253	107
≤20	115	49	37.8	63.2	0.01	0.60	1.0	0.7	1.6	ns	1.0	1.0
>20	189	84
≤25	200	69	65.8	48.1	0.14	0.73	1.8	1.2	2.7	*	1.3	0.7
>25	104	64
≤30	237	93	78.0	30.1	0.08	0.39	1.5	1.0	2.4	ns	1.1	0.7
>30	67	40
≤35	260	111	85.5	16.5	0.02	0.19	1.2	0.7	2.0	ns	1.0	0.9
>35	44	22
≤40	276	120	90.8	9.8	0.01	0.11	1.1	0.5	2.1	ns	1.0	0.9
>40	28	13
≤45	288	127	94.7	4.5	ndv	0.05	0.9	0.3	2.2	ns	1.0	1.2
>45	16	6
≤50	291	128	95.7	3.8	ndv	0.04	0.9	0.3	2.5	ns	1.0	1.1
>50	13	5
≤55	295	130	97.0	2.3	ndv	0.02	0.8	0.2	2.8	ns	1.0	1.3
>55	9	3
≤60	298	131	98.0	1.5	0.00	0.02	0.8	0.2	3.8	ns	1.0	1.3
>60	6	2

ndv: no diagnostic value, * p <0.05, ns: not significant. SST-MST: Saku-Saku Test—mastication and swallowing times; Sens: sensitivity; Spec: specificity; YI: Youden index; SSR: sensitivity specificity ratio (SSR = min(Sens, Spec)/max(Sens, Spec)); OR: odds ratio (=diagnostic odds ratio [DOR]); CI: confidence interval; LL: lower limit; UL: upper limit; +LR: positive likelihood ratio; −LR: negative likelihood ratio.

). The Youden index was 0.14 and SSR was 0.73 with a sensitivity of 65.8, a specificity of 48.1, a positive likelihood ratio of 1.3, a negative likelihood ratio of 0.7 and an odds ratio of 1.8 with 95% CI: 1.2–2.7.

4. Discussion

The SST-MST attempts to evaluate eating satisfaction in older people from the perspective of masticatory and swallowing functions. Several reports have shown that decreased masticating and swallowing function leads to longer meal times [3,11,12]. On the other hand, when the food oro-sensory exposure duration is short, satisfaction with the meal decreases [17]. A meal that enhances eating satisfaction may have a moderate eating pace, neither too fast nor too slow. Various cognitive, motor, and sensory functions generally decline with age. In addition, functions related to mastication and swallowing tend to decline, leading to decreased eating satisfaction. However, previous reports have not established the eating pace to achieve a sense of eating satisfaction. The results of this study have revealed that maintaining the ability to eat 2 g of rice crackers in approximately 25 s may lead to greater eating satisfaction. This standard indicates that one of the clear goals of maintaining oral function in older people is to ensure a comfortable diet in daily life. In the present study, oral function denotes a composite performance that encompasses not only intrinsic oro-motor capacity but also the quality and fit of dentures and other prosthetic restorations.

Tongue weakness, dysphagia, prolonged meal time outcomes and malnutrition are present in residents of long-term care facilities [3]. Slow swallowing process, which frequently occurs in patients with dementia, results in malnutrition [18]. The community-dwelling older people in this study were somewhat in better health status than residents of long-term care facilities or those with dementia. However, even in independently living older people, dysphagia is a risk factor for malnutrition [19]. Therefore, extended meal times may be related to the nutritional status in this population. A decline in nutritional status may be caused by the inability to eat a sufficient amount of food and avoidance of foods that are likely to cause aspiration. Dysphagia due to stroke causes limited food and liquid intake [18]. Mild impairment of mastication and swallowing function manifests as prolonged meal times in community-dwelling older people. It also affects eating satisfaction by restricting food and water intake.

However, the rice crackers used in the SST-MST are not an absolute test food, and the 25 s duration observed here is specific to this protocol–food combination. Previous studies employed protocols different from the SST-MST and used biscuits or shortbread as test foods, rather than the rice crackers used in our study. In a study of healthy individuals, Hiiemae et al. reported that the feeding eating time from bite to terminal swallow of 2.5 g biscuits was 21.5 s (range: 14.0–32.4 s) [20]. Furthermore, the eating time from ingestion to terminal swallow measured by Hiiemae and Palmer using 8 g of shortbread was 23.6 s [9]. In addition, Palmer et al. used 8 g of shortbread and confirmed that the eating time was 19.6 s [21]. Huckabee et al. reported that there were significant differences in the time required for mastication and swallowing depending on the type of cracker [22]. The results of these previous studies suggest that the cutoff values determined in this study are no longer valid for different test foods. However, the results of this study indicate that eating satisfaction may involve meal times that are neither short nor long. It is possible to obtain criteria for evaluating mastication and swallowing functions from various perspectives by determining an eating satisfaction time for similar foods and foods with different textures.

The rice crackers used in this study are widely consumed and a popular snack in Japan. It is important that the test food is accepted by the majority of the target population, as this improves the acceptability of the subjects and facilitates the conduct of the test. Furthermore, it has been reported that the physical properties of the rice crackers used in this study are good as test foods for mastication [15]. It has a middle hardness among commercial confectioneries such as rice crackers and biscuits [23]. The texture of the rice crackers used in this study were crispy and light, and hard enough to need mastication [15]. Therefore, this test food cannot be swallowed without mastication. Rice crackers easily mix with saliva to form a food bolus after mastication and an appropriate amount of oil prevents stickiness during chewing. This physical property lowers the risk of aspiration and increases the safety of test foods.

The present study created and used a measure called the sensitivity–specificity ratio (SSR), which is similar to the DOR and Youden index [24]. DOR is calculated as the ratio of the positive likelihood ratio to the negative likelihood ratio, and the Youden index is defined as sensitivity + specificity − 1. For screening tests that obtain continuous measures, cutoff values should be selected, and it is recommended that cutoff values be based on maximized diagnostic odds ratios and the Youden index [16,25]. It is well known that there are noticeable trade-offs between sensitivity and specificity, as well as between positive and negative predictive values [26]. So if the sensitivity or specificity is extremely high, the other will be extremely low. In such cases, the DOR and Youden indices are useful. In the early phases of screening test development, the preference for choosing the optimal cutoff value is often a criterion that weighs both sensitivity and specificity equally [24]. Using the SSR value as a reference when deciding the cutoff value of a screening test contributes to considering the equivalence of sensitivity and specificity. SSR is derived from sensitivity and specificity, making it independent of prevalence. Furthermore, it is an index that more succinctly and clearly demonstrates the balance between sensitivity and specificity than commonly used indices.

Two candidate cutoff values, 25 s and 45 s, emerged from the diagnostic metrics. While the 45 s threshold maximized sensitivity and minimized the negative likelihood ratio, its extremely low specificity rendered it unsuitable for routine screening. In contrast, the 25 s threshold showed the highest Youden index (0.16) together with an SSR of 0.75, thereby satisfying the prespecified balance criterion. Consequently, 25 s is proposed as a clinically reasonable reference, provided it is applied in conjunction with complementary oral-function measures rather than as a stand-alone diagnostic boundary. Clinically, a 25 s chewing duration on the SST-MST may serve as a simple chairside screening indicator to flag potentially reduced masticatory efficiency and prompt further evaluation in dental settings, while not functioning as a stand-alone diagnostic criterion. Although not examined in this study, this benchmark could inform the development of chewing habit-retraining protocols for older adults and the design of trials to evaluate their effectiveness.

Our study has some limitations. First, we did not separate the reference values for the SST-MST measure by sex. Hägglund’s study showed that males and females take different amounts of time to masticate and swallow the same food [27]. However, since the SST-MST was assumed to be a clinical measure, we aimed to make it a simple criterion; therefore, we did not evaluate for males and females separately in this study. Second, the diagnostic performance of the SST-MST was only moderate; its sensitivity and specificity were modest, and the corresponding Youden indices did not meet commonly accepted thresholds. Therefore, the SST-MST may be unsuitable as a stand-alone test. Rather, we envisage its use in combination with other oral-function measures, functioning as a single item within a broader scale for evaluating eating satisfaction.

Third, it has been pointed out that mastication and swallowing times vary with age [11,14,22]. Therefore, the reference values obtained in this study apply to community-dwelling older people aged 70–99 years (mean 80.1 years); further research is needed to confirm their applicability to younger and more advanced-age populations. Additionally, future research should incorporate external validation, objective motion-based measures, validated psychometric scales for patient-centered outcomes, and standardized assessments of denture quality/fit and of cognitive/mood status as potential confounders.

Fourth, the 16 food items used for the difficulty in eating test were selected by a managerial dietician taking into consideration the hardness and texture of the food; however, food selection was not sufficiently objective. Fifth, the SST-MST was measured by a dentist until the movement of the thyroid cartilage accompanying the swallowing motion was no longer observed. However, examination of the mouth or pharynx was not performed, and it was unclear whether all the food was swallowed.

Finally, this result is for Kameda Seika’s Happy Turn 2 g; other foods may have different comfortable mastication and swallowing times.

5. Conclusions

In conclusion, in this specific cohort of community-dwelling older people (70–99 years) and with this specific test food (half a 2 g rice cracker), an SST-MST cutoff of approximately 25 s showed a weak, although statistically significant, association with an insufficient indicator of eating satisfaction. Accordingly, the SST-MST should be regarded as a simple screening tool rather than a diagnostic test and used in conjunction with other clinical measures; the 25 s value is a pragmatic, context-dependent reference.