• Open Access

Disparities in edentulism and tooth loss between Māori and non-Māori New Zealand women


Correspondence to:
Beverley Lawton, Director, Women's Health Research Centre, Department of Primary Health Care & General Practice, Wellington School of Medicine & Health Sciences, PO Box 7343, Wellington South, New Zealand. Fax: +64 4 385 5473; e-mail: bev.lawton@otago.ac.nz


Objectives: Tooth loss and edentulism are important negative health outcomes; however, there is little current information about the prevalence of these conditions among adults in New Zealand (NZ). This study describes the dentate status of Maori and non-Maori NZ women with regard to tooth loss, edentulism and denture-wearing, and investigates ethnic and sociodemographic disparities within the sample. Associations between dentate status, socio-demographic and health-related factors are described.

Methods: Participants were 1,817 women who were screened for participation in a randomised controlled trial.

Results: 9.0% of women were edentulous and 30.3% wore a denture (partial or complete). The mean number of teeth present was 24.2, and older women had fewer teeth on average. Socio-demographic and ethnic disparities in tooth loss and edentulism were observed. Maori ethnicity was strongly associated with edentulism and tooth loss, with Maori women five times more likely than NZ European women to be edentulous. These associations held after controlling for age, education, smoking, diabetes, cardiovascular disease history, and BMI.

Conclusions: Marked ethnic disparities in edentulism and tooth loss exist in New Zealand. Effective targeted programmes are needed to reduce the public health impact of poor oral health among Maori.

Edentulism is the state of having had all of the natural teeth removed, while the term ‘tooth loss’ refers to the loss (usually incremental and unplanned) of one or more natural teeth but which falls short of the entire complement. Tooth loss and edentulism are oral health states of some importance. The World Organization (WHO) recognises edentulism as a poor health outcome that not only affects oral health, but also general health status and quality of life.1 Improving oral health is a goal of Healthy People 2010 for the US Department of Health and Human Services to achieve over the first decade of the new century,2 and the reduction of disparities in oral health are a key point of public health policy in many countries around the world.

Edentulism can affect day-to-day functioning, facial appearance and the ability to speak and eat.3,4 It is a risk factor for weight loss among elderly people.5 Partial or complete loss of teeth can impair chewing ability and thus limit the types of foods that an individual chooses to eat. Edentulous individuals are likely to favour softer processed foods that are typically higher in fat and cholesterol, and lower in vitamin and mineral content over foods with higher nutritional value that may be more difficult to eat.6–8 Furthermore, dietary changes associated with the transition to edentulism may lead to increased risk of selected systemic diseases such as cancer and cardiovascular disease.6,9 Tooth loss and edentulism have been associated with coronary heart disease, hypertension and stroke.10,11 Additionally, one study found edentulism to be associated with existing and future progression of carotid stenosis,12 and another showed a significant association between tooth loss and the presence of carotid artery plaque when controlling for risk factors.13

Edentulism is more prevalent among indigenous people, rural and remote dwellers and people of lower socioeconomic status.9,14,15 Māori are the indigenous people of New Zealand (NZ) and currently make up 15% of the total population, and 9.5% of the female population aged 40–64 years in New Zealand.16 Major health disparities exist between Māori and non-Māori (principally European descent) in New Zealand. Māori have an eight to nine year lower life expectancy than non-Māori in New Zealand, and are more likely to have major chronic diseases such as cardiovascular disease and diabetes.17 Similar health disparities have occurred in other countries with a comparable history of colonisation.18,19

Historically, New Zealand has had high rates of tooth loss and edentulism by international standards.20 Two national dental surveys have been conducted in New Zealand: the Survey of Adult Oral Health (SAOH) in 197621 and the Study of Oral Health Outcomes (SOHO) in 1988.22 The earlier survey showed that 28% of 35–44 year-old adults were edentulous, and this had fallen to 12% by 1988.22 Neither of the national dental surveys identified differences in edentulism prevalence between Māori and non-Māori, but significant differences in tooth loss were observed. Among adults aged 35–44 years in the 1976 survey, Māori had a mean of 11.9 missing teeth, while NZ Europeans had a mean of 7.1 missing.21 A similar difference was obtained in 1988: Māori aged 35–44 years were reported to have a mean of 14.6 missing teeth, while that for Europeans was 12.5. However, Māori were under-represented in those surveys. It has been estimated that 9.3% of the population were Māori in both 1976 and 1988 (a possible under-estimate); however, only 116 (6.5%) and 188 (6.0%) participants in the 1976 and 1988 national surveys (respectively) were Māori. Thus, there were insufficient numbers of Māori in either survey to enable generalisation of comparisons between Māori and other ethnic groups in New Zealand.

Improving oral health and reducing disparities are key objectives of the New Zealand Health Strategy.23 Despite this official recognition of the importance of oral health, there is a paucity of recent data to enable examination of ethnic disparities in oral health among New Zealand adults. No national monitoring studies of New Zealanders’ oral health have been undertaken since 1988. The current study aims to provide much-needed information on oral health by describing the dentate status (and its associations) of a group of Māori and non-Māori Zealand women aged between 40 and 74 years, and to investigate and quantify ethnic and sociodemographic disparities within the sample.



Participants were drawn from a convenience sample of women interviewed to determine eligibility for participation in a randomised controlled trial (RCT) called the ‘Women's Lifestyle Study’. Recruitment to the RCT was carried out between November 2004 and November 2005 by postal invitation. Women were recruited from two sources. First, letters of invitation were sent to women aged 51–74 years in an existing cohort that was recruited between 1999 and 2002 from 10 general practices in the greater Wellington area of New Zealand.24 Exclusion criteria for the original cohort included some cancers, severe, debilitating or terminal clinical conditions such as breast cancer, schizophrenia or impaired mobility. The second group of women were recruited from 11 general practices and two marae health clinics in the Wellington region.25 General practitioners at participating practices were asked to identify women aged 50–69 years (40–60 years for Māori and Pacific women) from their practice register, and signed study invitation letters for their own patients. Doctors were asked to remove any patients deemed inappropriate for participation in a lifestyle study. The letter (which also contained a reply slip and pre-paid envelope) invited women to contact the researchers if interested in learning more about the study. Replies were followed by a phone call from a research nurse to determine eligibility and then to invite eligible women to an interview.

Data collection

Information was collected during a face-to-face interview conducted by primary care research nurses in hired rooms in community health care settings. Interviews lasted between 1.0 and 1.5 hours, and participants’ responses were recorded directly onto a computer via a customised Microsoft Access form. Research nurses took part in a comprehensive training session on the administration of all aspects of the data collection process prior to interviewing participants.

The educational history of participants was categorised as: no secondary school education; at least NZ school certificate in one or more subjects; or a qualification beyond secondary school. Participants were classified as current smokers if they answered “yes” to the question “do you currently smoke one or more cigarettes per day?” Participants were categorised as diabetic if they answered “yes” to the question “Have you ever been told by a doctor that you have diabetes (other than in pregnancy)?” Participants were assigned a socio-economic status (SES) score using the New Zealand Deprivation Index (NZDep2001) – a validated, census-derived,16 area-based index of deprivation.26 NZDep2001 scores range from 1 (least deprived) through to 10 (most deprived). For the purposes of this research, scores were grouped into three categories: 1–3 (Least deprived), 4–7 and 8–10 (Most deprived) for analysis. Ethnicity was self-defined using the New Zealand 2001 census question.

Dental health data

Each participant was asked whether she had any of her own natural teeth; if she did not, she was asked to estimate the age at which she had lost her last tooth. A third question related to whether she had any “dentures, plates or false teeth that you can take out” in the upper or lower jaw. Another question sought information on when she had last visited a dentist (whether in the last 2 years, more than 2 years ago, or more than 5 years ago).

The interviewer next prompted the participant to count her teeth by using the statement “I would like to ask you to count the number of teeth you have in your upper, and in your lower jaw.” Each was offered a disposable glove and the use of a mirror, and was asked to count from the back of the mouth, working her way around by moving a finger from the edge of one tooth to the next. Participants were asked not to count dentures or bridges, but to count any teeth that had ‘roots in the gum’ (in order to include crowned teeth). Separate counts of the number of teeth in the upper and lower jaws were obtained.

Data analytic approach

Data were analysed using the Statistical Package for the Social Sciences (SPSS) and Intercooled Stata version 8. Bivariate analyses were used in SPSS to explore the data. For multivariate analysis, the data were transferred to Stata and regression analyses were used to test the study hypotheses and derive adjusted estimates for the dependent variables. Poisson regression modelling was used for the number of teeth present and logistic regression was used for dichotomous oral health outcomes.


Participation rates

Letters of invitation were sent to 5913 women in total (2021 to the existing cohort and 3,892 to the new sample). Of these 317 women were no longer resident at the recorded address and could not be contacted. Replies were received from 3,036 of those contacted (54.3%), with a higher response rate (83.9%) from those in the existing cohort (1,695 of the 2021). Of those who replied, 2,464 (81.2%) were willing to be assessed for eligibility for the Lifestyle Study, and only those women who attended an interview with the research nurse contributed data to the present analysis. Some 1,036 (42.0%) of those women who replied were pre-screened for eligibility by telephone; of those, 502 were eligible and attended an interview (48.5%). An additional 1,321 women were assessed for eligibility at an interview setting; the remaining 107 women were not interviewed because recruitment into the trial closed before appointments were made. The final number included was 1,823 women; each was interviewed by a nurse, with 1,817 (99.7%) giving information on their teeth. Subsequent analyses are limited to these 1,817 (representing an overall participation rate of 32.5%).

Description of sample

The characteristics of the sample are presented in Table 1. The mean age of participants was 59.4 years (SD 6.7; range 40 to 74), and most were of NZ European ethnicity (80.2%). Māori made up 8.3% of the sample. Due to the recruitment strategy used for the Women's Lifestyle Study, proportionally more NZ Europeans fell into the older age groups, while proportionally more Māori were in the younger age groups.

Table 1.  Characteristics of participants by ethnicity.a
 NZ European N (%)Māori N (%)Pacifc N (%)Other/unknown N (%)Total N (%)
  1. Notes:

  2. (a) All differences statistically signifcant.

  3. (b) Deprivation data unavailable for 267 participants

Age group (years)     
 40-4913 (0.9)61 (40.1)11 (28.9)085 (4.7)
 50-59694 (47.5)66 (43.4)21 (55.3)82 (48.0)863 (47.3)
 60-69631 (43.2)21 13.8)6 (15.8)76 (44.4)734 (40.3)
 70-74124 (8.5)4 (2.6)013 (7.6)141 (7.7)
 Mean age (SD)60.3 (6.1)51.9 (7.7)52.6 (6.4)60.1 (6.1)59.4 (6.7)
Socio-economic statusb     
 Dep scores 1-3 (Least deprived)724 (58.3)23 (17.6)6 (18.8)61 (40.1)814 (52.3)
 Dep scores 4-7382 (30.8)54 (41.2)8 (25.0)77 (50.7)521 (33.5)
 Dep scores 8-10 (Most deprived)135 (10.9)54 (41.2)18 (56.3)14 (9.2)221 (14.2)
 No secondary school354 (24.2)63 (41.4)15 (39.5)32 (18.8)464 (25.5)
 At least school cert in 1+ subjects380 (26.0)49 (32.2)16 (42.1)60 (35.3)505 (27.7)
 Qualifcation beyond secondary school728 (49.8)40 (26.3)7 (18.4)78 (45.9)853 (46.8)
 Current smoker121 (8.3)46 (30.3)11 (28.9)14 (8.2)192 (10.5)
 Diabetic69 (4.7)11 (7.2)4 (10.5)17 (9.9)101 (5.5)
Total (percent by row)1462 (80.2)152 (8.3)38 (2.1)171 (9.4)1823

A greater proportion of NZ Europeans and participants of ‘other/unknown’ ethnicity had gained an educational qualification beyond the secondary school level (or had achieved a secondary school qualification) than Māori or Pacific participants. Over three times more Māori and Pacific women were current smokers than NZ Europeans and ‘others’. Diabetes prevalence was lowest among NZ Europeans (Table 1).

Edentulism and tooth-loss

A total of 164 (9.0%) participants were edentulous, and 159 reported the age at which they reached that state. The mean age for the transition to edentulism was 31.0 years (SD 12.7, range 16 to 68). Of the 1653 dentate participants, 11 (0.7%) had five or fewer teeth remaining, while 93 (5.6%) had ten or fewer teeth remaining, and 526 (28.9%) had at least 28 teeth present. Similar numbers of teeth were present in participants’ lower and upper jaws. Some 272 participants (15.0%) had an edentulous upper jaw, while 171 (9.4%) had an edentulous lower jaw. The edentulous maxilla/dentate mandible combination was reported by 108 (5.9%), while seven (0.4%) had an edentulous mandible opposing a dentate (or partially dentate) maxilla.

Data on dental status by the socio-demographic characteristics of participants is presented in Table 2. Women with fewer teeth (on average) were more likely to be older, of Māori or ‘other/unknown’ ethnicity, have lower levels of education, lower SES, have a BMI of <20.0 kg/m2 or >25.1 kg/m2, have diabetes, be current smokers, or have had previous cardiovascular disease (includes diagnosis of myocardial infarction, angina, coronary artery bypass graft, angioplasty or stroke). A greater prevalence of (complete) edentulism was also observed among these groups.

Table 2.  Dental status of participants by socio-demographic characteristics.a,b
 Number of remaining teeth (dentate or partially dentate arches only)cDentures (includes partial dentures)Totald
 None N (%)Upper Mean (SD)Lower Mean (SD)Overall Mean (SD)None N (%)Upper N (%)Lower N (%)Both N (%)N (col %)
  1. Notes:

  2. (a) Education unknown for one participant

  3. (b) All differences by ethnicity, education, deprivation, and age group were statistically signifcant.

  4. (c) 222 Participants excluded due to edentulism or missed the question

  5. (d) Differences that did not reach statistical signifcance are fagged with “ns”

  6. Missing data: BMI not known for 12 participants, Diabetes status for 3 participants, and current smoking status for 1 participant.

Age group (years)         
 40-495 (5.9)12.4 (2.8)12.6 (2.6)23.9 (6.2)59 (69.4)23 (27.1)11 (12.9)8 (9.4)85 (4.7)
 50-5957 (6.6)12.8 (2.3)12.9 (2.1)25.0 (5.3)663 (77.5)181 (21.1)90 (10.5)77 (9.0)861 (47.4)
 60-6980 (10.9)12.3 (2.5)12.4 (2.4)23.8 (6.0)473 (65.1)236 (32.4)130 (17.9)112 (15.4)733 (40.3)
 70-7422 (15.9)11.6 (2.8)11.5 (2.7)21.5 (6.7)70 (51.1)63 (45.7)38 (27.7)33 (24.1)138 (7.6)
 NZ European122 (8.4)12.6 (2.2)12.7 (2.2)24.6 (5.4)1039 (71.9)375 (25.8)203 (14.0)239 (16.5)1462 (80.2)
 Māori34 (22.5)11.6 (3.2)11.7 (3.0)21.8 (7.4)78 (51.3)71 (46.7)43 (28.3)40 (26.3)152 (8.3)
 Pacifc3 (7.9)13.2 (2.0)12.9 (2.5)25.7 (5.0)31 (81.6)7 (18.4)4 (10.5)4 (10.5)38 (2.1)
 Other/unknown5 (2.9)11.9 (3.2)12.2 (2.9)23.1 (7.0)117 (68.4)53 (31.0)20 (11.7)19 (11.1)171 (9.4)
 No secondary school education80 (17.3)12.1 (2.7)12.0 (2.6)22.9 (6.4)280 (60.7)175 (38.0)100 (21.7)94 (20.4)464 (25.5)
 School cert in 1+ subjects52 (10.3)12.4 (2.5)12.4 (2.5)23.8 (6.1)346 (68.9)147 (29.2)85 (16.9)75 (14.9)505 (27.7)
Qualifcation beyond secondary school32 (3.8)12.8 (2.2)13.0 (2.1)25.1 (5.1)640 (45.7)184 (21.7)85 (10.1)64 (7.3)853 (46.8)a
BMI (kg/m2)         
 <20.05 (11.4)12.9 (2.3)12.4 (2.6)24.0 (6.8)32 (72.7)12 (27.3)8 (18.2)8 (18.2)44 (2.4)
 20.1-25.032 (6.4)12.7 (2.3)13.0 (2.1)25.1 (5.1)370 (74.1)111 (22.2)65 (13.0)47 (9.4)501 (27.8)
 25.1-30.061 (8.7)12.6 (2.2)12.6 (2.3)24.4 (5.5)497 (71.1)190 (27.1)96 (13.7)83 (11.9)704 (39.0)
 >30.165 (11.7)12.2 (2.7)12.3 (2.6)23.3 (6.4)358 (65.0)186 (33.6)98 (17.8)ns90 (16.3)556 (30.8)
 Yes11 (10.9)11.6 (2.9)11.2 (3.0)20.9 (7.5)56 (56.6)41 (40.6)22 (22.2)19 (19.2)101 (5.5)
 No153 (8.9)12.5 (2.4)12.7 (2.3)24.4 (5.6)1207 (70.9)461 (27.0)247 (14.5)ns211 (12.4)1713 (94.3)
 1+ daily31 (16.2)11.6 (2.7)11.7 (2.8)21.8 (6.8)103 (54.5)81 (42.9)44 (23.3)39 (20.6)191 (10.5)
 Non-smoker133 (8.2)12.6 (2.4)12.7 (2.3)24.5 (5.6)1161 (71.9)422 (26.1)225 (13.9)191 (11.8)1625 (89.5)
Previous cardiovascular disease         
 Yes23 (17.4)11.9 (2.8)11.6 (3.1)21.9 (7.0)70 (53.8)55 (42.0)38 (29.2)33 (25.4)132 (7.3)
 No141 (8.4)12.5 (2.4)12.7 (2.3)24.4 (5.6)1195 (71.3)448 (26.7)231 (13.8)197 (11.8)1685 (92.7)
Overall164 (9.0)12.5 (2.4)12.6 (2.4)24.2 (5.8)1265 (69.6)503 (27.7)269 (14.8)230 (12.7)1817


Dentures (including partial dentures) were worn by 540 participants (30.3%), of whom 503 wore an upper denture, while 269 wore a lower denture (27.7% and 14.9%, respectively). Dentures were worn on both upper and lower arches by one in eight women. Data on denture-wearing was incomplete for 12 participants.

Dental visiting

Dental visiting data were missing for 132 participants who were edentulous or did not know when they had last visited the dentist. Among the 1595 dentate women who could recall, 1414 (77.8%) reported that they had attended a dentist for a check-up and/or treatment within the previous two years. The characteristics of those who had not been to the dentist in the past two years are presented in Table 3. Proportionally more Māori had not visited the dentist within the past two years (26.4%) than NZ Europeans (9.0%). Women who had less education, higher BMI scores, diabetes, were current smokers, or had previous cardiovascular disease (includes diagnosis of myocardial infarction, angina, coronary artery bypass graft, angioplasty or stroke) were also less likely to have attended the dentist recently.

Table 3.  Dental visiting among dentate participants.
 Not in last 2 years N (%)Not in last 5 years N (%)
  1. Notes:

  2. Statistically signifcant differences are fagged with *

  3. Missing data: BMI not known for 12 participants, Diabetes status for three participants, and current smoking status for one participant.

Age group (years)  
 40-4917 (23.3)8 (11.0)
 50-59102 (13.3)33 (4.3)
 60-6950 (7.8)17 (2.7)
 70-7412 (10.5)*4 (3.9)*
 NZ European117 (9.0)34 (2.6)
 Māori29 (26.4)12 (10.9)
 Pacifc7 (22.6)6 (19.4)
 Other/unknown28 (17.5)*10 (6.3)*
 No secondary school education71 (19.2)23 (6.2)
 School cert in 1+ subjects49 (11.3)19 (4.4)
 Qualifcation beyond secondary school61 (7.7)*20 (2.5)*
BMI (kg/m2)  
 <20.03 (8.6)0
 20.1-25.039 (8.5)16 (3.5)
 25.1-30.061 (9.9)19 (3.1)
 >30.174 (15.6)*26 (5.5)*
 Yes15 (17.2)7 (8.0)
 1+ daily31 (20.5)15 (9.9)
Had previous cardiovascular disease  
 Yes17 (16.3)3 (2.9)
 No164 (11.0)*59 (4.0)
Overall181 (11.3)62 (3.9)

Multivariate analysis

Māori ethnicity, education, smoking and age were significantly associated with negative dental health status (Table 4). Māori were five times more likely to be edentulous than NZ Europeans, while women of ‘other/unknown’ ethnicity were 0.3 times as likely to be edentulous than NZ Europeans. Māori were four times more likely than NZ Europeans to wear a denture (of any type). Lack of education was also associated with infrequent dental visiting, and high BMI was associated with edentulism and denture-wearing. After adjusting for age and other putative confounders, the association of Māori ethnicity with infrequent dental visiting did not persist (Table 4). A second set of multivariate models was constructed, restricted to individuals in the 50–59 year age range to account for the different distribution of age groups in the sample with Māori being over-represented in the younger age groups. We chose the 50–59 year age group as it is well represented in our sample, and because it is an important age group as it represents pre-retirement people with many years of retired life ahead of them. Within the 50–59 year age range, similar associations by Māori ethnicity for the tooth-count, edentulism, and denture-wearing were found to those reported for the analysis of the whole sample (Table 5).

Table 4.  Multivariate models for tooth presence, edentulism, denture wearing, and dental visiting.
 Number of teeth present IRR per tooth (95% CI)Edentulism OR (95% CI)Denture wearing OR (95% CI)Last dental visit >2 years ago OR (95% CI)
Sociodemographic characteristics    
Age (per year)0.98 (0.98-0.98)1.11 (1.07-1.13)1.09 (1.07-1.12)0.96 (0.93-0.98)
 NZ European (reference)1111
 Māori0.73 (0.70-0.77)5.0 (2.9-8.5)4.0 (2.5-6.5)1.4 (0.8-2.5)
 Pacifc1.05 (0.97-1.13)1.1 (0.3-4.0)1.1 (0.4-3.3)1.5 (0.6-3.7)
 Other/unknown1.00 (0.97-1.04)0.3 (0.1-0.8)0.9 (0.5-1.5)2.1 (1.3-3.3)
 No secondary school qualifcation0.93 (0.90-0.96)1.6 (1.0-2.3)1.3 (0.9-1.8)1.9 (1.3-2.9)
 School cert in 1+ subjects (reference)1111
 Qualifcation beyond secondary school1.09 (1.06-1.12)0.3 (0.2-0.5)0.5 (0.3-0.7)0.7 (0.5-1.1)
Health-related factors    
BMI (kg/m2)    
 <20.00.93 (0.87-0.99)2.0 (0.7-5.9)2.1 (0.9-5.2)1.2 (0.3-4.2)
 20.1-25.0 (ref)1111
 25.1-30.00.99 (0.96-1.01)1.2 (0.7-1.8)1.1 (0.7-1.7)1.1 (0.7-1.7)
 >30.10.93 (0.90-0.95)1.7 (1.1-2.8)1.7 (1.1-2.5)1.5 (1.0-2.3)
Diabetic0.91 (0.87-0.96)0.7 (0.4-1.4)1.0 (0.6-1.8)1.3 (0.7-2.4)
Smoking 1+ daily0.83 (0.80-0.86)2.1 (1.3-3.4)1.9 (1.3-2.9)1.5 (0.9-2.4)
Previous cardiovascular disease0.92 (0.88-0.97)1.5 (0.9-2.5)1.7 (1.1-2.7)1.4 (0.8-2.5)
Table 5.  Multivariate models for tooth presence, edentulism, denture wearing, and dental visiting for study participants aged between 50 and 59.
 Number of teeth present IRR per tooth (95% CI)Edentulism OR (95% CI)Denture wearing OR (95% CI)Last dental visit >2 years ago OR (95% CI)
Sociodemographic characteristics    
Age (per year)0.99 (0.98-0.99)1.08 (0.96-1.21)1.08 (0.98-1.19)0.84 (0.77-0.91)
 NZ European (reference)1111
 Māori0.64 (0.60-0.69)5.8 (2.7-12.5)5.2 (2.6-10.3)1.5 (0.7-3.4)
 Pacifc1.00 (0.92-1.10)1.2 (0.3-5.8)1.8 (0.5-6.6)1.5 (0.4-5.7)
 Other/unknown1.01 (0.97-1.06)-1.4 (0.6-3.3)1.9 (1.0-3.6)
 No secondary school qualifcation0.89 (0.86-0.93)2.3 (1.2-4.5)1.7 (1.0-3.1)2.1 (1.2-3.7)
 School cert in 1+ subjects (reference)1111
 Qualifcation beyond secondary school1.08 (1.05-1.12)0.3 (0.1-0.7)0.4 (0.2-0.8)0.8 (0.5-1.4)
Health-related factors    
 BMI (kg/m2)    
 <20.00.91 (0.82-1.01)1.5 (0.2-15.7)2.0 (0.4-10.7)1.1 (0.1-9.1)
 20.1-25.0 (ref)1111
 25.1-30.01.00 (0.97-1.04)0.7 (0.3-1.6)0.8 (0.4-1.6)1.3 (0.7-2.4)
 >30.10.94 (0.91-0.97)1.1 (0.5-2.4)1.2 (1.6-2.2)1.7 (0.9-3.0)
Diabetic0.87 (0.82-0.96)1.0 (0.3-3.6)1.3 (0.5-3.7)1.5 (0.6-4.0)
Smoking 1+ daily0.84 (0.80-0.89)1.6 (0.8-3.5)1.9 (1.0-3.6)1.1 (0.6-2.3)
Previous cardiovascular disease1.06 (1.00-1.14)0.3 (0.1-1.2)0.8 (0.3-2.0)1.6 (0.7-3.6)


This study found the prevalence of edentulism to be 9.0% among a convenience sample of New Zealand women aged between 40 and 75 years. Māori were five times more likely to be edentulous than NZ European women. Edentulism was strongly associated with older age, less education, and Māori ethnicity. This ethnic disparity is consistent with other health outcomes in New Zealand, and most likely contributes to poor health status and poorer quality of life for Māori. The data suggest that, in the 40–74 year age groups, Māori women had experienced markedly greater tooth loss than other women. This finding held after controlling for multiple putative confounding variables including age, socioeconomic status, education, smoking, diabetes, cardiovascular disease history, and BMI.

These study findings may not be generalisable to all New Zealand women aged 40–74 years, as participants were drawn from a convenience sample of women from one city (Wellington), and two different sampling strategies were used resulting in different age profiles of participants of Māori and Pacific ethnicity compared to NZ European and other ethnicities. A cross-sectional study using random selection would be more appropriate to adequately assess population prevalence; however, data of that type are not available for New Zealand at present. The analytical approach used in this study (with adjustment for age and other factors in the multivariate analyses) should have helped to address this problem. In our dataset, 7.6% (95% confidence interval 5.8, 9.4) of 863 women aged 50–59 were Māori. New Zealand census data indicate that 9.4% of women aged 50–59 in June 2005 were Māori, and it is therefore likely that Māori were adequately represented in this sample. Furthermore, Māori in this sample are over-represented in smoking rates and in the most deprived SES status which is consistent with census data.16

Our data may also underestimate the rate of tooth loss and edentulism for New Zealand women, as those who are more severely ill or immobilised were not invited to participate. The study relied upon self-reported data on the number of remaining teeth, and the interviewers were not trained dentists; however, a number of studies have found that self-reported dentate status compares well with ‘gold standard’ assessment by clinicians, particularly for measures such as the number of remaining teeth.27,28

A strength of the study is the large sample size, so Type 2 error was unlikely to be a problem. Furthermore, the wide range of health (and other) information collected enabled the examination of a broader set of determinants for tooth loss and edentulism than is usual for studies of this type. To our knowledge this is the largest published sample of Māori women showing dentate status.

The finding that Māori women in this sample were five times more likely to be edentulous than NZ European women (even after controlling for putative confounders) raises an important public health matter. Māori should enjoy at least the same level of oral health as other New Zealanders.29 While we have not specifically identified the underlying reason for the ethnic disparities observed in this study, evidence from other studies indicates that many factors may contribute to dental status, including dental attendance patterns, attitudes to dental care, differing dental care and lack of knowledge of dental procedures.15 Debate has occurred in New Zealand about the considerable negative impact of racism (both personal and institutional) on health outcomes for Māori, and it is reasonable to assume that this may also contribute to the poor dental outcomes found for Māori in this study.19,30

In New Zealand, dentistry for adults is predominantly an out-of-pocket, fee-for-service system. Dental care for children and adolescents (until age 18) is free. A study of Canadian children showed that having universal access to a publicly-financed dental program was not effective in reducing socio-economic disparities in oral health.31 Differing levels of care may also play a part, one study showed that African Americans and those with lower socio-economic status were more likely to receive a dental extraction (given similar disease extent and severity) once they entered the dental care system.15

It is encouraging that disparities may be considerably reduced when a dental system has universal care, standard treatment and required attendance. This is clearly shown by the success of the US military oral health care service. The US military has a universal access-to-care dental system, with compulsory attendance. It has been demonstrated that having standardised, regular care with high rates of utilisation and accessibility can almost eliminate disparities that traditionally exist between African American and white American adults.32


Significant ethnic disparities in edentulism appear to exist in New Zealand. The findings of this study suggest that Māori women in the 40–74 age group experience markedly greater tooth loss and edentulism than other New Zealand women. This difference by ethnicity persisted after controlling for a number of putative confounding variables (including age, education, smoking, diabetes, cardiovascular disease history, and BMI). These findings should be further investigated and programs developed to reduce the public health impact of poor oral health among Māori.


The researchers wish to thank the women who participated in this study and Selina Brown and the team of research nurses for their role in recruitment and data collection. The Women's Lifestyle Study is funded by the National Heart Foundation, a Lottery Health Research grant and the Hutt Valley District Health Board. Jonathan Broadbent is supported by Grant R01 DE-015260-01A1 from the National Institute of Dental and Craniofacial Research, National Institutes of Health, Bethesda, Maryland, USA.