Do Temporal Changes in Complete Tooth Loss Point to Cohort Effects?

ABSTRACT

Objectives

Complete tooth loss is more common among older adults; however, its prevalence has declined significantly over time. It is currently unknown whether age and period effects act independently or whether they interact to produce differential effects by birth cohorts. This study aimed to examine temporal changes in the prevalence of complete tooth loss in Australia, examine the effects of age, period and cohort and assess life course changes within birth cohorts.

Methods

The study used data from three repeated cross‐sectional waves of Australia's National Dental Telephone Interview Survey (2005, 2013 and 2021), each comprising independent samples. The prevalence of edentulism among adults aged 41 years and older was described. The independent effects of age and period and examined cohort effects were estimated using an Age‐Period‐Cohort Interaction (APC‐I) model.

Results

The prevalence of complete tooth loss among adults aged 41 years and older declined from 12% in 2005 to 6% in 2021. The analysis shows that while edentulism increased with age, it decreased significantly over time across most age groups. Despite the low prevalence of edentulism, younger generations showed higher‐than‐expected levels of complete tooth loss. Significant cohort effects were observed, with older cohorts showing high tooth loss and younger cohorts exhibiting concerning trends.

Conclusion

Despite overall progress in reducing edentulism, the increase in complete tooth loss among younger generations poses a threat to long‐term oral health outcomes in Australia. Ongoing surveillance and targeted public health interventions are critical in preventing a resurgence in tooth loss, particularly among younger populations. Comprehensive oral health programs and equitable access to care are essential to maintaining Australia's progress in oral health and in addressing emerging challenges.

1. Introduction

A lifelong history of oral disease and trauma, alongside the challenges of accessing oral health care services, contributes to tooth loss [1]. Complete tooth loss (edentulism) can lead to significant functional impairments and have a profound long‐term impact on overall health and wellbeing [2].

During midlife, adults face increasing vulnerability to physical and mental changes that impact their health [3]. Staying healthy at this life stage is vitally important as it predicts health in older age [3]. Given the associations between tooth loss and systemic health [4], examining differences in cohorts during midlife can enable the early identification of oral health problems that affect health and wellbeing later in life.

With the rising number of older adults globally [5] and in Australia [6], the consequences of tooth loss are becoming an ever‐growing concern. Temporal changes reflect shifts in the collective experiences of populations arising from the combined effects of aging, social influences, and environmental factors [7].

Historical events, such as the introduction of new dental technologies, changes in dental care practice, and public health policies, have also played a key role in shaping oral health outcomes [8]. These changes impact different birth cohorts as they age, with varying effects across age groups, leading to cohort‐specific differences in oral health outcomes. Moreover, cohort effects can evolve throughout life [7], depending on their exposure to such policies and interventions. Understanding these population‐level shifts is vital for designing public health strategies that address the needs of various generations and reduce the burden of tooth loss.

In Australia, the prevalence of edentulism among adults over the age of 15 years has declined significantly; from 14.4% in (1987–1988) to 4.0% in 2017–18 [9]. The latter equates to approximately 798 489 individuals [10]. Understanding the temporal dynamics of tooth loss remains crucial for developing public health strategies to reduce the associated burden on individuals' wellbeing. Although the global count of prevalent cases of edentulism continues to rise [11], Australia's progress highlights the potential for improvement and the necessity of focusing on preventive and restorative oral health measures.

The aims of this study were to (1) examine trends in edentulism, (2) explore the extent to which edentulism varied as a function of three dimensions of time: age, period and cohort and (3) determine whether the observed differences accumulated, remained constant or diminished during a cohort's life‐course.

2. Methods

2.1. Data Source

Australia's National Dental Telephone Interview Survey (NDTIS) is conducted, on average, on a three‐yearly cycle since 1994 and is part of a strategic plan to monitor oral health and use of dental services in Australia. NDTIS participants are a randomly selected representative sample of Australian adults aged 18 years and older. The survey collects data through a telephone interview, with an option of completing the questionnaire online from 2017. A core set of questions collected at every cycle ensures the monitoring of trends in key oral health and dental visiting indicators.

The data for this study is from three repeated cross‐sectional waves of the NDTIS (2005, 2013 and 2021), each comprising independent samples of Australian adults. NDTIS 2005 was part of the National Study of Adult Oral Health (NSAOH 2004–2006), which additionally included oral epidemiological clinical examinations. The University of Adelaide Human Research Ethics Committee provided ethics approval for NSAOH 2004–2006/NDTIS 2005 (H‐001‐2004), NDTIS 2013 (HS‐2013‐014) and NDTIS 2021 (H‐2020‐153). Informed consent was obtained from the participants in all three time points of the survey.

The three NDTIS waves created two balanced 8‐year intervals in which age groups and time periods are of equal length. Midlife‐ and older‐adult NDTIS participants (aged 41 years and over) were included in the study sample. Participants were categorised into six age groups for each survey period, with each group spanning approximately 8 years, reflecting the interval between survey periods: 41–48 years, 49–56 years, 57–64 years, 65–72 years, 73–80 years and 81+ years. The synthetic cohorts roughly spanned across four generations: ‘the Greatest Generation’ (1917–1924 and 1925–1932), ‘the Silent Generation’ (1933–1940 and 1941–1948), ‘Baby Boomers’ (1949–1956 and 1957–1964) and ‘Generation X’ (1965–1972 and 1973–1980). The synthetic cohort approach assumes that migration and mortality do not have a differential impact on the outcome across cohorts. Please refer to Table 2 for a visual representation of the age groups, periods and cohorts included in the analysis.

TABLE 2.

Population prevalence of complete tooth loss (95% CI) by age, period and birth cohort in Australian adults, 2005–2021.

Survey periods	Age groups						Birth cohorts
	41–48	49–56	57–64	65–72	73–80	81+
						39.9 (35.3, 44.8)	1917–1924
					31.0 (27.6, 34.6)	34.9 (28.1, 42.3)	1925–1932
				19.6 (17.4, 22.1)	19.7 (15.1, 25.3)	17.2 (10.3, 27.2)	1933–1940
			11.4 (9.9, 13.2)	11.9 (9.1, 15.3)	9.5 (6.4, 13.7)		1941–1948
		3.9 (3.1, 4.9)	5.3 (3.8, 7.2)	6.8 (4.1, 10.8)			1949–1956
2005	1.3 (0.8, 2.0)	2.8 (1.7, 4.7)	5.3 (3.1, 8.9)				1957–1964
2013	0.3 (0.1, 0.9)	2.6 (1.0, 6.7)					1965–1972
2021	1.3 (0.5, 3.1)						1973–1980

Note: The significance for grey shaded cells is denote age–period combinations contributing to each birth cohort trajectory used in the APC‐I analysis. Cohorts are represented diagonally across survey periods.

2.2. Data Analysis

2.2.1. Outcome Variable

Complete tooth loss was a dichotomous variable (yes/no) ascertained from the question ‘Do you have any of your own natural teeth?’

2.2.2. Weighting and Descriptive Analysis

Analyses were weighted to account for NDTIS sampling methodology. After descriptive analyses, a detailed examination of cohort effects utilised APC interaction (APC‐I) models following the approach proposed by Luo and Hodges [7].

2.2.3. Age Period Cohort Interaction (APC‐I) Models

Age Period Cohort (APC) models traditionally estimate independent age, period and cohort effects; however, because the three variables are perfectly collinear, the model is not identifiable [7, 12]. To address this, age and period were treated as main effects, while cohort was modelled as their interaction [7]. Formally, the APC‐I model specifies the log‐odds of edentulism for age group a in period p as:

$$\text{logit}\left\{\mathrm{Pr}\left(Y_{\mathrm{ap}}=1\right)\right\}=\mu +{\alpha }_a+{\pi }_p+{\delta }_{\mathrm{ap}},$$

where ${\alpha }_a$ and ${\pi }_p$ are age and period main effects, and ${\delta }_{\mathit{ap}}$ is the age‐by‐period interaction term representing the cohort deviation [7, 12].

The APC‐I approach recognises that period effects vary across age groups, thereby defining a cohort effect [7, 13]. When age effects are constant across periods, or period effects are constant across age groups, ${\delta }_{\mathit{ap}}=0$ and no cohort effect exists [7].

An APC‐I model for edentulism including age and period main effects and their interactions were fitted for each cohort (full model). Cohort effects were evaluated using a global deviance test comparing the full model with a baseline model containing only age and period main effects [12]. Following a significant test, inter‐cohort and intra‐cohort analyses were conducted [12, 14].

Inter‐cohort deviations quantify how observed edentulism differs from the values expected under the absence of cohort effects, isolating cohort‐specific influences beyond age and period [7]. Intra‐cohort analyses examined linear trends in deviations from the age and period main effects. This approach evaluated whether a cohort's advantage/disadvantage persisted, increased, or decreased with age. The trend of the cohort's average deviation, along with the direction of the slope, shows whether cumulative advantage or disadvantage is emerging or if the cohort's deviation remains stable over time [7].

If a cohort's average deviation is significant but the linear trend is not, this supports the constant effects hypothesis [7]. This hypothesis suggests that the observed cohort effect remains consistent over time, without significant changes as the cohort ages. In contrast, if both the average deviation and the linear trend have the same positive or negative sign, it indicates a cumulative advantage or disadvantage [7]. If the average deviation and linear trend have opposite signs, or if the slope is significant but the average deviation is not, it suggests that the initial advantage or disadvantage diminishes over time, supporting the age‐as‐leveller hypothesis [7]. If neither are significant, there is no clear pattern in cohort variation [7].

The models used effect coding (sum‐to‐zero coding) [12]. Age‐by‐period interactions were visualised through cohort deviation heat maps [12].

SAS software version 9.4 (SAS Institute Inc.) was used for data cleaning, editing and development of the dataset and Stata Statistical Software release 17 (StataCorp) for descriptive analyses. The APC‐I models were fitted using the R Statistical Software v4.1.2 (R Core Team 2021) via the APCI package, developed by Xu and Luo [12].

3. Results

The sample consisted of 9258 participants aged 41+ years in 2005, 4597 in 2013, and 3508 in 2021 (Table 1). Younger age groups had higher participation and slightly more women. Approximately one‐third were cardholders (an indicator of eligibility to receive public dental care), and similarly, nearly one‐third reported avoiding dental visits due to cost. In 2005, more participants reported being uninsured, but this reversed in 2013, with insurance coverage increasing further by 2021.

TABLE 1.

Sample description for the Australian National Dental Telephone Interview Survey for the years 2005, 2013 and 2021, adult population aged 41 years and over.

	NDTIS 2005 (n = 9258)		NDTIS 2013 (n = 4597)		NDTIS 2021 (n = 3508)
	n	Weighted % (95% CI)	n	Weighted % (95% CI)	n	Weighted % (95% CI)
Edentulism
Yes a	1255	11.8 (11.0, 12.6)	360	7.8 (6.8, 8.9)	173	5.5 (4.3, 6.8)
No	8003	88.2 (87.4, 89.0)	4237	92.2 (91.1, 93.2)	3335	94.5 (93.2, 95.7)
Age
41–48	2217	27.2 (25.9, 28.6)	1067	25.3 (23.6, 27.2)	680	21.4 (19.1, 24.0)
49–56	2064	23.5 (22.3, 24.6)	1054	24.0 (22.3, 25.8)	703	21.0 (18.6, 23.6)
57–64	1965	19.2 (18.2, 20.4)	994	19.2 (17.7, 20.7)	780	21.4 (19.0, 23.9)
65–72	1411	13.2 (12.4, 14.1)	746	15.2 (13.9, 16.7)	693	16.3 (14.4, 18.3)
73–80	1074	11.5 (10.6, 12.4)	449	9.7 (8.7, 10.9)	476	13.4 (11.7, 15.3)
81+	527	5.4 (4.8, 6.0)	287	6.5 (5.6, 7.5)	176	6.6 (5.3, 8.1)
Sex
Male	3722	48.3 (47.1, 49.6)	1942	48.4 (46.4, 50.4)	1597	49.2 (46.3, 52.1)
Female	5536	51.7 (50.4, 52.9)	2655	51.6 (49.6, 53.6)	1911	50.8 (47.9, 53.7)
Cardholder (eligibility to receive public dental care)
Yes	3458	34.0 (32.5, 35.5)	1604	35.8 (34.0, 37.8)	1074	31.1 (28.6, 33.8)
No	5792	66.0 (64.5, 67.5)	2982	64.2 (62.2, 66.0)	2337	68.9 (66.2, 71.4)
Insurance status
Insured	4365	48.1 (46.4, 49.8)	2626	52.7 (50.7, 54.7)	2068	54.7 (51.8, 57.7)
Uninsured	4867	51.9 (50.2, 53.6)	1928	47.3 (45.3, 49.3)	1387	45.3 (42.3, 48.2)
Avoided dental visit due to cost
Yes	2569	27.2 (26.0, 28.5)	1346	32.9 (31.0, 34.8)	915	28.3 (25.7, 31.0)
No	6685	72.8 (71.5, 74.0)	3250	67.1 (65.2, 69.0)	2562	71.7 (69.0, 74.3)

^a The assessment of trends also estimated the age‐standardised prevalence of complete tooth loss, using the 2001 census of the Australian population as the referent. The population age‐standardised estimates were very similar: 12% in 2005, 8% in 2013, and a slightly lower population estimate of 5% in 2021.

3.1. Descriptive Analysis (Aim 1 and Aim 2)

The proportion of Australian adults aged 41+ years with no natural teeth declined significantly from 12% in 2005 to 8% in 2013, and further to 6% in 2021 (Table 1).

Table 2 presents estimates by age (columns), period (diagonals) and cohorts (rows). The 1917–1924 and the 1973–1980 birth cohorts had only one data point, while the 1925–1932 and 1965–1972 birth cohorts had data points for two periods.

Figure 1 (age‐by‐period interactions) and Table 2 show important reductions in the prevalence of edentulism, with the pattern of decline varying across age groups and survey periods.

FIGURE 1.

Period‐specific age patterns (A) and age‐specific period trends (B) for complete tooth loss (%) among Australian adults, 2005–2021.

Age‐specific period trends (Figure 1A; Table 2) show declines across survey periods for both older (65+ years) and middle‐aged adults (57 to 65 years), with 3‐ and 2‐fold reductions, respectively. In contrast, the two younger age groups showed steady low prevalence estimates.

Within survey periods, absolute differences were notable among older age groups, while relative differences were notable in younger age groups, reflecting the higher prevalence of edentulism among the older adults. Period‐specific age patterns (Figure 1B) show that edentulism prevalence consistently increased with increasing age across all survey years. Among older age groups, stark differences in prevalence were seen between survey periods; however, these differences were less pronounced in younger age groups. The slopes show points of inflection, where increases in edentulism shift from lower to higher rates, with this point moving to older age groups over time, indicating a threshold for edentulism that is being delayed across survey periods. For example, in 2005, a steep increase occurred at ages 49–56 years, shifting to ages 57–64 years in 2013. In 2021, a slower, steadier increase was seen up to ages 73–80 years, where the 2005 slope showed less growth than expected, while the 2013 slope exhibited a steep rise in edentulism prevalence.

Results for the within‐ and between‐cohort analyses are presented in Table 2. Around one‐third of the 1925–1932 cohort experienced complete tooth loss, with prevalence remaining stable across survey periods. Similarly, cohorts born before 1957 had relatively stable estimates. In contrast, cohorts born after 1956 showed greater relative changes; for example, the 1965–1972 cohort showed an almost 9‐fold increase in edentulism between 2013 and 2021.

At ages 73–80 years and 81+ years, the prevalence of edentulism markedly decreased with successive birth cohorts. The differences were smaller at ages 65–72 years and even less pronounced at ages 41–56 years and 57–64 years.

3.2. APC‐I Models (Aim 2 and Aim 3)

The global deviance test for edentulism indicated cohort effects. Table 3 presents estimates for age and period main effects, inter‐cohort average deviations and intra‐cohort life‐course trends. The intercept represents the global mean for the outcome, with age and period main effects as deviations from this mean, and cohort effects as deviations from age and period effects [12].

TABLE 3.

APC‐I model for complete tooth loss.

	Complete tooth loss
	Coefficient	SE	P > |t|	% Estimate a
Intercept	−2.534	0.06	0.00	7.9
Age effect
41–48 years	−2.243	0.21	0.00	89.4
49–56 years	−0.920	0.17	0.00	60.2
57–64 years	−0.071	0.11	0.52	6.9
65–72 years	0.521	0.11	0.00	68.4
73–80 years	1.047	0.10	0.00	184.9
81+ years	1.666	0.12	0.00	429.1
Period effect
2005	0.499	0.07	0.00	64.7
2013	−0.152	0.09	0.09	14.1
2021	−0.347	0.10	0.00	29.3
Inter‐cohort average deviation
1917–1924	−0.038	0.13	0.76	3.7
1925–1932	0.292	0.09	0.00	33.9
1933–1940	−0.006	0.09	0.95	0.6
1941–1948	−0.069	0.08	0.39	6.7
1949–1956	−0.214	0.11	0.04	19.3
1957–1964	0.024	0.14	0.86	2.4
1965–1972	−0.270	0.24	0.27	23.7
1973–1980	0.790	0.31	0.01	120.3
Intra‐cohort life dynamics (slope)
1917–1924	NA	NA	NA	NA
1925–1932	0.15	0.14	0.28	NA
1933–1940	−0.33	0.16	0.04	NA
1941–1948	−0.34	0.14	0.02	NA
1949–1956	−0.01	0.16	0.96	NA
1957–1964	0.09	0.20	0.66	NA
1965–1972	0.65	0.27	0.02	NA
1973–1980	NA	NA	NA	NA

Note: The intercept was calculated as (exp (coefficient) × 100). The percent deviations were calculated as |(1 − exp.(coefficient))| × 100 This was applied consistently for both positive and negative coefficients.

^a % Estimates for the intercept and percent deviations were calculated using the approach by Lu and Luo¹⁴.

3.2.1. Age and Period Main Effects

The global estimate for edentulism was 7.9%. The main effects of age show a steady increase as the population ages. The main effects of period show a sharp decline from 2005 to 2013, but less so from 2013 to 2021. Significant age and period effects were identified overall, with only two exceptions: the 57–64 years age group and the 2013 survey period, which showed negative deviations of 6.9% and 14.1%, respectively. Generally, negative deviations were observed among younger age groups and more recent periods, suggesting a lower‐than‐average outcome, while positive deviations were noted among older age groups and the earliest period, suggesting a higher‐than‐average outcome.

The 41–48 years age group exhibited the highest negative deviation, with its coefficient estimate indicating an 89.4% lower outcome compared to the global mean. Conversely, the oldest age group showed the highest positive deviation, with the estimated age coefficient suggesting an increased prevalence of edentulism by 429.1% compared to the global mean. For period main effects, the percent change from the global mean for the 2005 period was 64.7% increased prevalence, whereas the 2021 period presented a 29.3% decrease in the prevalence of edentulism.

3.2.2. Cohort Effects

Figure 2 presents the estimated interaction terms from the APC‐I model. Each square of the heatmap represents an age‐by‐period interaction term, interpreted as the deviation from the main effects of age and period for a cohort at that age and in that period. The intensity of the tone reflects the magnitude of the deviation effect size, with darker shades corresponding to larger and lighter shades corresponding to smaller effects (or no effect). Estimates with negative signs indicate a lower‐than‐expected prevalence of the outcome whereas positive signs indicate a higher‐than‐expected prevalence. Each diagonal running from the lower left to the upper right represents one cohort. The dotted lines on the heatmap represent a significant average inter‐cohort deviation (i.e., the average cohort effect differs from the main effects of age and period), but a non‐significant intra‐cohort slope (i.e., no change within the cohort) [12]. These results can also be seen in Table 3.

FIGURE 2.

Heatmap plot of estimated cohort effects for complete tooth loss by age group and period, Australian adults ages 41 years and older, 2005–2021.

A higher‐than‐expected outcome and positive deviation from the main effects of age and period for the 1925–32 cohort was found. The 1933–1940 and the 1941–1948 cohorts showed small or no effects in the first two periods but a greater, negative magnitude of the deviation in 2021, indicating lower than expected prevalence of edentulism. The 1949–1956 cohort consistently showed a lower‐than‐expected outcome and small effect size deviation. The 1957–1964 cohort consistently showed smaller effect sizes as indicated by the lighter tones of pink. A large, positive effect size of the deviation for the 1973–1980 cohort was found, indicating a higher‐than‐expected prevalence. The average cohort deviation for this group was 0.79, equivalent to a 120.3% higher than expected prevalence (Table 3).

The intra‐cohort slope could not be assessed for the 1917–1920 and 1973–1980 cohorts as only a single time‐point was available, preventing an assessment of life‐course dynamics for these two cohorts. Table 3 indicates the intra‐cohort slopes were significant for three cohorts: 1933–40, 1941–48 and 1965–72 (positive for the latter and negative for the others). The intra‐cohort slopes should, however, be interpreted together with the inter‐cohort deviations [7].

The constant effect hypothesis was supported in the 1925–1932 cohort, which showed higher‐than‐expected tooth loss throughout their life‐course with a higher proportion with edentulism compared to other cohorts at the same age. The 1949–1956 birth cohort also presents constant effects across the life‐course, but with lower‐than‐expected outcome.

The age‐as‐leveller hypothesis was supported in the 1933–1940, 1941–1948 and 1965–1972 birth cohorts. These results likely indicate that compared to other cohorts, the 1933–1940 and the 1941–1948 cohorts became edentulous at younger ages, but eventually, as other cohorts aged, they presented similar estimates. The 1965–1972 birth cohort presented a lower‐than‐expected outcome when young, but higher than expected when older.

4. Discussion

This study is based on repeated nationally representative samples of the Australian adult population, which allows the use of synthetic cohorts to assess population‐wide trends in outcomes across multiple birth cohorts and for extended periods [15]. The study demonstrated significant changes in the prevalence of edentulism from 2005 to 2021, by age and birth cohort. Edentulism increased with age and decreased with period; moreover, age and period interacted differentially across cohorts.

By examining cohort effects as the interaction between age and period effects through the APC‐I approach, the methodological and conceptual limitations associated with attempting to estimate the effect of each dimension separately were addressed [7]. Age effects are explained by the cumulative nature of dental disease; as the population ages, the proportion of people with missing teeth increases as seen by increasing patterns of edentulism across the different survey periods. This is of particular relevance where the population is shifting towards older age groups, such as in Australia. However, the latter periods show greater retention of teeth in older adults, highlighting the need to maintain good oral health and facilitate access to oral health care services among older adults.

Declines in the prevalence of edentulism are seen with cohort substitution—older cohorts are replaced by younger cohorts with less disease [8]. Despite this finding, younger cohorts seem to be experiencing higher‐than‐expected complete tooth loss. Although the estimates remain low, this trend raises concerns. If younger cohorts are indeed having more extractions at younger ages, the decline in edentulism seen over recent decades may stall or reverse. This would increase unfavourable oral health outcomes with potential detriments to general health and wellbeing. Given the non‐reversible nature of tooth loss, it will be essential to continue oral health surveillance activities to monitor trends, identify and address the causes of this unexpected rise in younger cohorts.

The effects of period events cannot be underestimated. Reductions in edentulism among different birth generations have been linked to the transition in dental treatment philosophies and reductions in dental caries rates due to the wide‐spread introduction of water fluoridation and use of fluoridated toothpaste [8, 16]. Members of the three younger cohorts in our study would have been eligible to receive dental care through the Australian School Dental Scheme (SDS) established in 1973 [17]. The reductions noted in our study could at least be partially explained by the implementation of the SDS, in which all Australian children had access to dental services. However, changes to the SDS in funding and child eligibility since 1980 may have reduced the population‐wide positive effects seen early in its implementation. This is a plausible explanation for the worsening of outcomes in younger generations and warrants further investigation. There is no equivalent public dental service upon turning 18 years of age, unless eligible for public dental care as an adult or covered under parental private dental insurance. Young adults having to pay for dental services out‐of‐pocket could translate into declines in visiting and potentially explain the worsening of outcomes in younger cohorts.

A study strength is the national representativeness of the samples, which increases the generalisability of the findings. However, a limitation is the inclusion of only three periods. Although this was done to balance the analysis (i.e., equal year intervals for age and period groupings), it allowed for only one time point for the youngest and oldest birth cohort, which impeded assessing their intra‐cohort slopes. Similarly, only two time points were available for another two birth cohorts, for which life course dynamics should be interpreted with caution [7]. Besides the implementation of different government policies, synthetic cohorts are subject to the effects of international migration [15] and this analysis did not account for changes in sociodemographic characteristics, including country of birth. These limitations suggest that future studies should explore trends in tooth loss by considering key population characteristics to identify factors that could be targeted through health initiatives.

5. Conclusion

This study demonstrated cohort effects in temporal changes of edentulism and underscores the critical need for sustained public health efforts aimed at preventing tooth loss, particularly as Australia's population continues to age. While progress has been made, the possible and unexpected rise in edentulism among younger generations raises serious concerns for the future of oral health. If not addressed through targeted prevention, early intervention and accessible dental care services, Australia risks reversing decades of improvements in oral health outcomes. Investing in comprehensive oral health programs, maintaining surveillance of population‐level trends and ensuring equitable access to care are essential to preserving the health and wellbeing of all Australians, now and in the future.

Author Contributions

Gloria C. Mejia: contributed to conception, design, data analysis and interpretation, drafted and critically revised the manuscript. Kym M. McCormick: contributed to the formal APC‐I analysis and interpretation, visualisation, critical review and editing of the manuscript. Sergio Chrisopoulos: contributed to the conception, data acquisition, collection and management, descriptive analysis, interpretation of findings and critically revised the manuscript. Najith Amarasena: contributed to the conception, interpretation of results, critical review and editing of the manuscript. Liana Luzzi: contributed to the data acquisition and collection, contextualised and interpreted results, and critically revised the manuscript. All authors gave their final approval and agreed to be accountable for all aspects of the work.

Funding

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.