The number and proportion of older Mori aged 65 years or more is growing.1 In 2006, Mori made up 6.8 percent of the older population in New Zealand and by 2026 it is predicted Mori will comprise 9.5% of older people.1Of Mori who reach age 75, many have multiple health problems, but may not have readily available whnau (extended family) to care for and support them due to the migration of whnau members from rural to urban areas often for employment. Few Mori reach 85 years of age (<0.2% of the Mori population)2 and 80 to 90 years represents advanced age for Mori.National surveys indicate Mori over the age of 50 have significant inequalities in health outcomes and a higher burden of chronic illness compared with non-Mori of the same age.1 In 2006, a quarter of Mori aged over 50 years lived in the most deprived areas3 and therefore more likely to live lives challenged with less economic wealth and resources. Among Mori aged 65 to 69 years, only a third have average material living standards and 32% of single people and 22% of couples live in hardship.4 Mori over the age of 65 years have a high level of mobility-related disability and are more likely than non-Mori to have high support needs.4 In 2006, life expectancy at birth lagged by 7 years for Mori,1 however the Mori population is growing faster than the non-Mori population, thereby potentially expanding the population of older Mori.5 The nutritional status of older Mori is unknown. Older people in general tend to be at higher risk of malnutrition, but there are limited data on older Mori because they are underrepresented in surveys. Poor nutritional status is related to an increased risk of developing health problems.6 Higher patterns of morbidity occur in malnourished older people7 and increased functional difficulties, cognitive decline and comorbidities may all lead to malnutrition in advanced age.8 Screening for nutrition risk provides a simple and rapid method to identify those at risk of becoming malnourished.9 The validated screening tool Seniors in the Community: Risk Evaluation for Eating and Nutrition (SCREENII) determines nutrition risk using four key factors: food intake; physiological; adaptive; and functional.10 The feasibility study leading to Life and Living in Advanced Age: a Cohort Study in New Zealand (LiLACS NZ)11 found of non-Mori and Mori combined, half (52%) were at high nutrition risk.12 The same sample analysis showed that nutrition risk was mildly related to cardiovascular disease.13Kai (food) has cultural importance for Mori. The traditional food systems used by Mori apply a whnau,or group, approach to the growing, procuring, cooking and eating of food. The principle of manaakitanga (sharing) ensures that food is available for all. There is little known about the significance and access to kai for older Mori.Mori aged over 65 years are significantly more likely than non-Mori of the same age to be overweight or obese.1 Mori have a different body composition to non-Mori. Younger Mori have a higher proportion of lean body mass14 compared to non-Mori and this may persist into old age. The universal Body Mass Index (BMI) cut-off points which define overweight and obesity may not be appropriate for Mori.14 Increased weight has been associated with better health outcomes for older people,15 and a higher BMI may be protective.Risk factors related to risk of malnutrition have not been examined in Mori. Depression in older people may lead to impaired appetite and food intake, leading to weight loss and decreased physical function.16 Functional status may also be related to nutrition risk.17 Poor mobility may adversely affect food procurement, preparation and cooking, and lead to dependence on support networks.This study aims to describe the prevalence of nutrition risk in older Mori and to identify cultural, social and physical factors associated with high nutrition risk. Identification of these factors may provide useful insights to improve food and nutrition intake and enhance the lives of older Mori.MethodThis cross-sectional study examines nutrition risk in older Mori from two regions of New Zealand: the Bay of Plenty (as part of the feasibility study for LiLACS NZ18,19), and in Northland, where a second group of older Mori participants were recruited to extend the study sample and increase the generalisability of results to Mori.Bay of Plenty participantsThere were a total of 186 older people invited to participate in the feasibility study for LiLACS NZ and 112 participants were recruited (response rate 60%). Of these, 79 were non-Mori and 33 were Mori. For Mori, the inclusion criteria was aged 75 to 79 years old (birth date between 1 January, 1929 and 31 December, 1933). For all other ethnicities, birth date was between 1 January and 31 December 1922, 85 years. Younger Mori participants were recruited, as the gap in life expectancy between Mori and non-Mori was 8.2 years for men and 8.8 years for women.20Whnau and local networks were used to invite Mori who fitted the age criteria in the Rotorua, Whakatne and \u014cp\u014dtiki areas. All people within the age range were eligible. Support of local general practitioners was sought, especially to identify any participants who might be too unwell to be invited to participate in the study. Overall, 45 Mori were invited to participate, 12 declined and 33 agreed (an overall response rate of 73%). There were 20 participants living in Rotorua, 8 in Whakatne and 5 in \u014cp\u014dtiki.Northland participantsIn Northland, potential participants were sought from the Northland District Health Board patient management system, the Te Tai Tokerau Mori Electoral Roll, the Mori community, iwi/hap\u016b groups, church groups, Mori non-government organisations, sporting and other social groups. To ensure the participation of healthy Mori the inclusion criterion at enrolment for the Northland Mori was no hospital admission in the previous six months and enrolled on the Te Tai Tokerau Mori Electoral Roll. For the Northland Mori participants, the age criteria were the same as the Bay of Plenty participants: birth date between 1 January 1929 and 31 December 1933, aged 75 to 79 years.A total of 44 community-living Mori, aged 75 to 79 years, living in Te Tai Tokerau/Northland, who met the eligibility criteria were approached and 34 agreed to participate (response rate 75%). Data from 33 Mori participants from the Bay of Plenty region engaged in the feasibility study were combined with the Northland sample.Ethics approval was granted by the Northern X Regional Ethics Committee in May 2011. All participants provided written informed consent.QuestionnaireDemographic data and responses to validated questionnaires were ascertained during face-to-face interviews by trained Mori interviewers in the Bay of Plenty and a Mori dietitian interviewer in Northland. The questionnaire was translated into te reo Mori (Mori language) and Mori interviewers were fluent in te reo. Interviews were conducted in the participants home or a local clinic, depending on participants preference.MeasuresDemographics: gender; age; living arrangement; and marital status were asked using standard questions from the New Zealand Census.21 Education was ascertained by any primary, secondary and/or tertiary education attendance. Self-rated standard of living was assessed by using part of the New Zealand Health, Work and Retirement study questionnaire.22Questions to reflect the specific cultural issues for older Mori were included. Items were identified from focus groups led by a Kaitiaki (guardian) group of Mori elders. The methods have been described elsewhere and link with previous research with Mori of advanced age.19 Cultural items included the importance of kai Mori (traditional food), the importance of taha wairua (spirituality) and whether te reo Mori was the first language. Whether the participant lived in their own hap\u016b (large extended family) area was recorded. Using a 5-step Likert scale (of not at all, to extremely important), the importance of: whnau; hap\u016b; language and tikanga (cultural practices); use of Rongo Mori (Mori medicine and healing) was determined. Access to kai Mori and gardening habits were recorded.19Nutrition risk assessmentThe 14-item validated questionnaire Seniors in the Community: Risk Evaluation for Eating and Nutrition, Version II (SCREENII) was used to determine nutrition risk.23 From an assessment of 21 tools which aim to assess the nutrition risk status of older adults, SCREENII was the only tool specifically designed for those living in the community.24 Based on comprehensive nutrition assessments, SCREENII has been validated among older people in Canada against the criterion of a dietitians clinical judgement of risk and has high inter-rater and test-retest reliability, as well as excellent sensitivity (94%) and specificity (78%) in detecting risk of malnutrition.23 SCREENII items are scored from 0 to 4, with high scores indicating low-risk, and scores less than or equal to two, out of a maximum of four, potentially leading to nutrition risk. The total scores range from 0 to 64. A cut-off of less than 50 is considered to be high nutrition risk.HealthDepression was assessed by the 15-item Geriatric Depression Scale (GDS-15). The GDS-15 is a reliable and valid self-rating depression screening scale developed specifically for older people.25 Scores range from 1\u201315 and correlate with depressive symptoms. A higher score indicates more depressive symptoms.Functional status was assessed with the Nottingham Extended Activities of Daily Living (NEADL),26 which is a measure of physical disability and independence. The NEADL asks whether the older person \u201cdoes\u201d a range of activities \u201con their own, on their own with difficulty, with help, or not at all\u201d. There are 22 items of activities within, covering four domains: mobility; in the kitchen; domestic tasks; and leisure activities. A higher score is indicative of a higher level of function.Physical assessments were conducted using portable equipment and included measures of height, weight, waist and hip circumference. Anthropometric measures followed the protocol advised by the National Nutrition Survey of New Zealand.27Statistical analysisDescriptive analyses were completed for socio-demographic, cultural and physical data and nutrition risk. The mean, standard deviation was calculated for the nutrition risk score.Univariate analyses were completed to examine the relationship between the SCREEN II score, socio-demographic and cultural variables using ANOVA, t-tests and Chi Square, dependent on the form of the data. For the variable \u201cspecial foods available when wanted\u201d, the missing values (those who did not report having special foods) were coded to 0. A sensitivity analysis was completed restricting the analysis to only those who answered this question. Significant variables from the univariate analyses at the level of p<0.2 and relevant variables related to the literature were entered into a generalised linear regression model with SCREEN II score as the dependent variable. IBM\u00ae SPSS\u00ae Statistics 20 (SPSS) was used for all analyses. Statistical significance was set for p-value less than 0.05 in the regression model. The regression was completed with the whole sample (see Table 3), and a sensitivity analysis also completed restricted to those who answered the question about access to important foods.ResultsThe participants comprised a total of 67 Mori, mean (SD) age 77 (1.5) years and there were 30 (44%) men. Forty-nine percent (n=33) of the participants lived in the Bay of Plenty region (14 men, 19 women) and 51% (n=34) lived in Northland (16 men, 18 women). All of the participants identified themselves as Mori, with 10 (14%) identifying themselves as Mori and other. Out of 65 participants who responded, 28 (43%) were married and two (3%) had never married. Thirty (46%) of the participants had been widowed and five (7%) had been separated or divorced. Thirty-two (49%) of the participants had lost a spouse during their lifetime and 42 (64%) lived alone. Overall, eight (12%) of the participants had received only a primary education, 35 (53%) had received a secondary education and 20 (30%) had been tertiary educated. A total of 51 (91%) of the participants were able to speak te reo Mori and it was spoken by 35 (56%) as a first language (Table 2). Table 1 shows the SCREENII subscores.Table 1: SCREENII item scores that indicate nutrition risk SCREENII Itema Participants with Scores \u22642b % (n) Milk Product Intake Drinks milk or eats milk products <1\u20132/day 75 (47) Meat and Alternatives Intake Eats meat or alternatives <1/day 60 (38) Fruit and Vegetable Intake Eats <3 serves/day 59 (37) Unintentional weight change Gain (28%) or loss (26%) of \u00b12kg in past 6 months 54 (35) Skips meals Skips meals sometimes/often/almost every day 53 (33) aSCREEN II items are the questions from SCREENII.bSCREEN II items with scores less than or equal to two, out of a maximum score of four, potentially lead to nutrition risk.34The mean weight (SD) of the men was 90 (17) kg and for women was 77 (20) kg. The mean (SD) BMI was 31 (5) for men, 31 (9) for women. The mean (SD) body fat percentage was 20 (7%) for men, 38 (9%) for women. The mean waist-to-hip ratio (WHR) was 0.98 (0.1) for men, 0.89 (0.1) for women.The overall mean SCREEN II score was 47.8 (5.3) (range 35\u201356, out of maximum 64). Nearly two-thirds (63%) of the participants were assessed as being at high nutrition risk (SCREEN II score <50). High nutrition risk was more common for women (60%) compared to men (40%).Table 2 shows the univariate analyses. Those at high nutrition risk (lower SCREENII score): were more likely to have lost a spouse; rated the importance of hap\u016b and the importance of language and culture to wellbeing more highly; and were less able to access kai Mori. Those at high nutrition risk also had a lower WHR and had a higher GDS-15 score. These variables were entered into a generalised linear regression model. Importance of hap\u016b and importance of language and culture were highly correlated (Spearman correlation coefficient 0.7). A decision was made to group these cultural variables together, and as they are positively reflective of Maori identity, were termed cultural identity in the generalised linear regression model.Table 2: Mean SCREENII score in relation to frequency counts (n) and percentage (%) for participant socio-demographic, cultural and physical characteristics. SCREEN II score Mean (SD) Participants n (%) p-value Gender 63 (94.0) 0.69 Men 48.1 (5.23) 28 (44.4) Women 47.6 (5.42) 35 (55.6) Ever lost a spouse 0.001 No 50.7 (4.57) 23 (43.4) Yes 46.2 (4.81) 30 (56.6) Marital status 0.34 - Married / Partnered 49.1 (5.37) 28 (44.4) - Widowed 47.0 (4.83) 28 (44.4) - Divorced / Separated 45.4 (6.43) 5 (7.9) - Never married 47.5 (9.19) 2 (3.2) Living arrangement 0.12 Alone 46.3 (5.53) 42 (67.7) With others 48.5 (5.15) 20 (32.3) Education 0.15 - Primary 44.9 (3.72) 8 (13.1) - Secondary 47.5 (5.04) 35 (57.4) - Tertiary 49.2 (5.90) 18 (29.5) Self-rated standard of living 0.44 High / fairly high 48.0 (4.79) 23 (36.5) Medium 47.9 (5.62) 39 (61.9) Fairly low / low 41.0 (0.00) 1 (1.6) Spirituality 0.85 Not important 48.5 (6.7) 8 (12.7) Somewhat / Moderately 48.4 (3.9) 9 (14.3) Very / Extremely 47.6 (5.3) 46 (73.0) Spirituality role in life 0.40 No part 55.0 (0.0) 1 (1.6) Somewhat / Moderate 47.6 (6.2) 10 (15.9) Very / Extreme 47.8 (5.1) 52 (82.5) Conversational language/s Te Reo Speaks English Yes 47.8 (5.4) 57 (96.6) 0.33 No 44.0 (0.0) 2 (3.4) Speaks Mori Yes 47.4 (5.5) 51 (91.1) 0.18 No 50.8 (3.6) 5 (8.9) Speaks Mori as the first language 0.11 Yes 46.9 (5.5) 35 (55.6) No 49.0 (5.0) 28 (44.4) Importance of whnau to wellbeing 0.29 Not at all 46.3 (5.3) 12 (19.4) A little / moderate 0.0 (0.0) 0 (0.0) Very / Extremely 48.1 (5.3) 50 (80.6) Importance of hap\u016b to wellbeing 0.03 Not at all 46.9 (5.6) 13 (21.0) A little / moderate 53.0 (3.2) 6 (9.7) Very / Extremely 47.2 (5.1) 43 (69.4) Importance of language and culture to wellbeing (cultural identity) 0.02 Not at all 45.5 (4.9) 11 (17.7) A little / moderate 53.2 (3.1) 5 (8.1) Very / Extremely 47.7 (5.2) 46 (74.2) Lives in own hap\u016b area 0.57 Yes 47.6 (4.9) 14 (22.2) No 48.6 (6.6) 49 (77.8) Importance of using Rongo Mori medicine 0.84 Not at all 49.0 (4.4) 5 (18.5) A little / Moderately 47.3 (4.3) 10 (37.0) Very / Extremely 47.8 (6.1) 12 (44.4) Kai Mori important 0.96 Yes 47.9 (5.4) 55 (87.3) No 47.8 (5.4) 8 (12.7) Kai Mori available when wanted 0.02 Yes 48.3 (5.1) 47 (90.4) No 42.6 (3.2) 5 (9.6) Manage your own garden 0.63 Yes 47.0 (7.6) 54 (87.1) No 48.0 (5.0) 8 (12.9) BMI kg/m2, mean (SD) 48.2 (5.21) 0.94 < 18.5 44 (0.00) 1 (1.8) 18.5 to 24.9 49.7 (5.54) 6 10.9) 25.0 to 29.9 47.2 (5.22) 21 38.2) 30.0 to 34.9 49.3 (4.60) 16 (29.1) 35.0 to 39.9 52.3 (4.32) 6 (10.9) > 40.0 43.4 (4.62) 5 (9.1) Waist to Hip ratio 48.1 (5.12) 0.03 Low <0.90 men, <0.85 women 46.8 (4.71) 12 (20.7) High \u22650.90 men, \u22650.85 women 48.4 (5.26) 46 (79.3) Depressive symptoms, GDS-15 47.8 (5.32) 0.67

The number and proportion of older Mori aged 65 years or more is growing.1 In 2006, Mori made up 6.8 percent of the older population in New Zealand and by 2026 it is predicted Mori will comprise 9.5% of older people.1Of Mori who reach age 75, many have multiple health problems, but may not have readily available whnau (extended family) to care for and support them due to the migration of whnau members from rural to urban areas often for employment. Few Mori reach 85 years of age (<0.2% of the Mori population)2 and 80 to 90 years represents advanced age for Mori.National surveys indicate Mori over the age of 50 have significant inequalities in health outcomes and a higher burden of chronic illness compared with non-Mori of the same age.1 In 2006, a quarter of Mori aged over 50 years lived in the most deprived areas3 and therefore more likely to live lives challenged with less economic wealth and resources. Among Mori aged 65 to 69 years, only a third have average material living standards and 32% of single people and 22% of couples live in hardship.4 Mori over the age of 65 years have a high level of mobility-related disability and are more likely than non-Mori to have high support needs.4 In 2006, life expectancy at birth lagged by 7 years for Mori,1 however the Mori population is growing faster than the non-Mori population, thereby potentially expanding the population of older Mori.5 The nutritional status of older Mori is unknown. Older people in general tend to be at higher risk of malnutrition, but there are limited data on older Mori because they are underrepresented in surveys. Poor nutritional status is related to an increased risk of developing health problems.6 Higher patterns of morbidity occur in malnourished older people7 and increased functional difficulties, cognitive decline and comorbidities may all lead to malnutrition in advanced age.8 Screening for nutrition risk provides a simple and rapid method to identify those at risk of becoming malnourished.9 The validated screening tool Seniors in the Community: Risk Evaluation for Eating and Nutrition (SCREENII) determines nutrition risk using four key factors: food intake; physiological; adaptive; and functional.10 The feasibility study leading to Life and Living in Advanced Age: a Cohort Study in New Zealand (LiLACS NZ)11 found of non-Mori and Mori combined, half (52%) were at high nutrition risk.12 The same sample analysis showed that nutrition risk was mildly related to cardiovascular disease.13Kai (food) has cultural importance for Mori. The traditional food systems used by Mori apply a whnau,or group, approach to the growing, procuring, cooking and eating of food. The principle of manaakitanga (sharing) ensures that food is available for all. There is little known about the significance and access to kai for older Mori.Mori aged over 65 years are significantly more likely than non-Mori of the same age to be overweight or obese.1 Mori have a different body composition to non-Mori. Younger Mori have a higher proportion of lean body mass14 compared to non-Mori and this may persist into old age. The universal Body Mass Index (BMI) cut-off points which define overweight and obesity may not be appropriate for Mori.14 Increased weight has been associated with better health outcomes for older people,15 and a higher BMI may be protective.Risk factors related to risk of malnutrition have not been examined in Mori. Depression in older people may lead to impaired appetite and food intake, leading to weight loss and decreased physical function.16 Functional status may also be related to nutrition risk.17 Poor mobility may adversely affect food procurement, preparation and cooking, and lead to dependence on support networks.This study aims to describe the prevalence of nutrition risk in older Mori and to identify cultural, social and physical factors associated with high nutrition risk. Identification of these factors may provide useful insights to improve food and nutrition intake and enhance the lives of older Mori.MethodThis cross-sectional study examines nutrition risk in older Mori from two regions of New Zealand: the Bay of Plenty (as part of the feasibility study for LiLACS NZ18,19), and in Northland, where a second group of older Mori participants were recruited to extend the study sample and increase the generalisability of results to Mori.Bay of Plenty participantsThere were a total of 186 older people invited to participate in the feasibility study for LiLACS NZ and 112 participants were recruited (response rate 60%). Of these, 79 were non-Mori and 33 were Mori. For Mori, the inclusion criteria was aged 75 to 79 years old (birth date between 1 January, 1929 and 31 December, 1933). For all other ethnicities, birth date was between 1 January and 31 December 1922, 85 years. Younger Mori participants were recruited, as the gap in life expectancy between Mori and non-Mori was 8.2 years for men and 8.8 years for women.20Whnau and local networks were used to invite Mori who fitted the age criteria in the Rotorua, Whakatne and \u014cp\u014dtiki areas. All people within the age range were eligible. Support of local general practitioners was sought, especially to identify any participants who might be too unwell to be invited to participate in the study. Overall, 45 Mori were invited to participate, 12 declined and 33 agreed (an overall response rate of 73%). There were 20 participants living in Rotorua, 8 in Whakatne and 5 in \u014cp\u014dtiki.Northland participantsIn Northland, potential participants were sought from the Northland District Health Board patient management system, the Te Tai Tokerau Mori Electoral Roll, the Mori community, iwi/hap\u016b groups, church groups, Mori non-government organisations, sporting and other social groups. To ensure the participation of healthy Mori the inclusion criterion at enrolment for the Northland Mori was no hospital admission in the previous six months and enrolled on the Te Tai Tokerau Mori Electoral Roll. For the Northland Mori participants, the age criteria were the same as the Bay of Plenty participants: birth date between 1 January 1929 and 31 December 1933, aged 75 to 79 years.A total of 44 community-living Mori, aged 75 to 79 years, living in Te Tai Tokerau/Northland, who met the eligibility criteria were approached and 34 agreed to participate (response rate 75%). Data from 33 Mori participants from the Bay of Plenty region engaged in the feasibility study were combined with the Northland sample.Ethics approval was granted by the Northern X Regional Ethics Committee in May 2011. All participants provided written informed consent.QuestionnaireDemographic data and responses to validated questionnaires were ascertained during face-to-face interviews by trained Mori interviewers in the Bay of Plenty and a Mori dietitian interviewer in Northland. The questionnaire was translated into te reo Mori (Mori language) and Mori interviewers were fluent in te reo. Interviews were conducted in the participants home or a local clinic, depending on participants preference.MeasuresDemographics: gender; age; living arrangement; and marital status were asked using standard questions from the New Zealand Census.21 Education was ascertained by any primary, secondary and/or tertiary education attendance. Self-rated standard of living was assessed by using part of the New Zealand Health, Work and Retirement study questionnaire.22Questions to reflect the specific cultural issues for older Mori were included. Items were identified from focus groups led by a Kaitiaki (guardian) group of Mori elders. The methods have been described elsewhere and link with previous research with Mori of advanced age.19 Cultural items included the importance of kai Mori (traditional food), the importance of taha wairua (spirituality) and whether te reo Mori was the first language. Whether the participant lived in their own hap\u016b (large extended family) area was recorded. Using a 5-step Likert scale (of not at all, to extremely important), the importance of: whnau; hap\u016b; language and tikanga (cultural practices); use of Rongo Mori (Mori medicine and healing) was determined. Access to kai Mori and gardening habits were recorded.19Nutrition risk assessmentThe 14-item validated questionnaire Seniors in the Community: Risk Evaluation for Eating and Nutrition, Version II (SCREENII) was used to determine nutrition risk.23 From an assessment of 21 tools which aim to assess the nutrition risk status of older adults, SCREENII was the only tool specifically designed for those living in the community.24 Based on comprehensive nutrition assessments, SCREENII has been validated among older people in Canada against the criterion of a dietitians clinical judgement of risk and has high inter-rater and test-retest reliability, as well as excellent sensitivity (94%) and specificity (78%) in detecting risk of malnutrition.23 SCREENII items are scored from 0 to 4, with high scores indicating low-risk, and scores less than or equal to two, out of a maximum of four, potentially leading to nutrition risk. The total scores range from 0 to 64. A cut-off of less than 50 is considered to be high nutrition risk.HealthDepression was assessed by the 15-item Geriatric Depression Scale (GDS-15). The GDS-15 is a reliable and valid self-rating depression screening scale developed specifically for older people.25 Scores range from 1\u201315 and correlate with depressive symptoms. A higher score indicates more depressive symptoms.Functional status was assessed with the Nottingham Extended Activities of Daily Living (NEADL),26 which is a measure of physical disability and independence. The NEADL asks whether the older person \u201cdoes\u201d a range of activities \u201con their own, on their own with difficulty, with help, or not at all\u201d. There are 22 items of activities within, covering four domains: mobility; in the kitchen; domestic tasks; and leisure activities. A higher score is indicative of a higher level of function.Physical assessments were conducted using portable equipment and included measures of height, weight, waist and hip circumference. Anthropometric measures followed the protocol advised by the National Nutrition Survey of New Zealand.27Statistical analysisDescriptive analyses were completed for socio-demographic, cultural and physical data and nutrition risk. The mean, standard deviation was calculated for the nutrition risk score.Univariate analyses were completed to examine the relationship between the SCREEN II score, socio-demographic and cultural variables using ANOVA, t-tests and Chi Square, dependent on the form of the data. For the variable \u201cspecial foods available when wanted\u201d, the missing values (those who did not report having special foods) were coded to 0. A sensitivity analysis was completed restricting the analysis to only those who answered this question. Significant variables from the univariate analyses at the level of p<0.2 and relevant variables related to the literature were entered into a generalised linear regression model with SCREEN II score as the dependent variable. IBM\u00ae SPSS\u00ae Statistics 20 (SPSS) was used for all analyses. Statistical significance was set for p-value less than 0.05 in the regression model. The regression was completed with the whole sample (see Table 3), and a sensitivity analysis also completed restricted to those who answered the question about access to important foods.ResultsThe participants comprised a total of 67 Mori, mean (SD) age 77 (1.5) years and there were 30 (44%) men. Forty-nine percent (n=33) of the participants lived in the Bay of Plenty region (14 men, 19 women) and 51% (n=34) lived in Northland (16 men, 18 women). All of the participants identified themselves as Mori, with 10 (14%) identifying themselves as Mori and other. Out of 65 participants who responded, 28 (43%) were married and two (3%) had never married. Thirty (46%) of the participants had been widowed and five (7%) had been separated or divorced. Thirty-two (49%) of the participants had lost a spouse during their lifetime and 42 (64%) lived alone. Overall, eight (12%) of the participants had received only a primary education, 35 (53%) had received a secondary education and 20 (30%) had been tertiary educated. A total of 51 (91%) of the participants were able to speak te reo Mori and it was spoken by 35 (56%) as a first language (Table 2). Table 1 shows the SCREENII subscores.Table 1: SCREENII item scores that indicate nutrition risk SCREENII Itema Participants with Scores \u22642b % (n) Milk Product Intake Drinks milk or eats milk products <1\u20132/day 75 (47) Meat and Alternatives Intake Eats meat or alternatives <1/day 60 (38) Fruit and Vegetable Intake Eats <3 serves/day 59 (37) Unintentional weight change Gain (28%) or loss (26%) of \u00b12kg in past 6 months 54 (35) Skips meals Skips meals sometimes/often/almost every day 53 (33) aSCREEN II items are the questions from SCREENII.bSCREEN II items with scores less than or equal to two, out of a maximum score of four, potentially lead to nutrition risk.34The mean weight (SD) of the men was 90 (17) kg and for women was 77 (20) kg. The mean (SD) BMI was 31 (5) for men, 31 (9) for women. The mean (SD) body fat percentage was 20 (7%) for men, 38 (9%) for women. The mean waist-to-hip ratio (WHR) was 0.98 (0.1) for men, 0.89 (0.1) for women.The overall mean SCREEN II score was 47.8 (5.3) (range 35\u201356, out of maximum 64). Nearly two-thirds (63%) of the participants were assessed as being at high nutrition risk (SCREEN II score <50). High nutrition risk was more common for women (60%) compared to men (40%).Table 2 shows the univariate analyses. Those at high nutrition risk (lower SCREENII score): were more likely to have lost a spouse; rated the importance of hap\u016b and the importance of language and culture to wellbeing more highly; and were less able to access kai Mori. Those at high nutrition risk also had a lower WHR and had a higher GDS-15 score. These variables were entered into a generalised linear regression model. Importance of hap\u016b and importance of language and culture were highly correlated (Spearman correlation coefficient 0.7). A decision was made to group these cultural variables together, and as they are positively reflective of Maori identity, were termed cultural identity in the generalised linear regression model.Table 2: Mean SCREENII score in relation to frequency counts (n) and percentage (%) for participant socio-demographic, cultural and physical characteristics. SCREEN II score Mean (SD) Participants n (%) p-value Gender 63 (94.0) 0.69 Men 48.1 (5.23) 28 (44.4) Women 47.6 (5.42) 35 (55.6) Ever lost a spouse 0.001 No 50.7 (4.57) 23 (43.4) Yes 46.2 (4.81) 30 (56.6) Marital status 0.34 - Married / Partnered 49.1 (5.37) 28 (44.4) - Widowed 47.0 (4.83) 28 (44.4) - Divorced / Separated 45.4 (6.43) 5 (7.9) - Never married 47.5 (9.19) 2 (3.2) Living arrangement 0.12 Alone 46.3 (5.53) 42 (67.7) With others 48.5 (5.15) 20 (32.3) Education 0.15 - Primary 44.9 (3.72) 8 (13.1) - Secondary 47.5 (5.04) 35 (57.4) - Tertiary 49.2 (5.90) 18 (29.5) Self-rated standard of living 0.44 High / fairly high 48.0 (4.79) 23 (36.5) Medium 47.9 (5.62) 39 (61.9) Fairly low / low 41.0 (0.00) 1 (1.6) Spirituality 0.85 Not important 48.5 (6.7) 8 (12.7) Somewhat / Moderately 48.4 (3.9) 9 (14.3) Very / Extremely 47.6 (5.3) 46 (73.0) Spirituality role in life 0.40 No part 55.0 (0.0) 1 (1.6) Somewhat / Moderate 47.6 (6.2) 10 (15.9) Very / Extreme 47.8 (5.1) 52 (82.5) Conversational language/s Te Reo Speaks English Yes 47.8 (5.4) 57 (96.6) 0.33 No 44.0 (0.0) 2 (3.4) Speaks Mori Yes 47.4 (5.5) 51 (91.1) 0.18 No 50.8 (3.6) 5 (8.9) Speaks Mori as the first language 0.11 Yes 46.9 (5.5) 35 (55.6) No 49.0 (5.0) 28 (44.4) Importance of whnau to wellbeing 0.29 Not at all 46.3 (5.3) 12 (19.4) A little / moderate 0.0 (0.0) 0 (0.0) Very / Extremely 48.1 (5.3) 50 (80.6) Importance of hap\u016b to wellbeing 0.03 Not at all 46.9 (5.6) 13 (21.0) A little / moderate 53.0 (3.2) 6 (9.7) Very / Extremely 47.2 (5.1) 43 (69.4) Importance of language and culture to wellbeing (cultural identity) 0.02 Not at all 45.5 (4.9) 11 (17.7) A little / moderate 53.2 (3.1) 5 (8.1) Very / Extremely 47.7 (5.2) 46 (74.2) Lives in own hap\u016b area 0.57 Yes 47.6 (4.9) 14 (22.2) No 48.6 (6.6) 49 (77.8) Importance of using Rongo Mori medicine 0.84 Not at all 49.0 (4.4) 5 (18.5) A little / Moderately 47.3 (4.3) 10 (37.0) Very / Extremely 47.8 (6.1) 12 (44.4) Kai Mori important 0.96 Yes 47.9 (5.4) 55 (87.3) No 47.8 (5.4) 8 (12.7) Kai Mori available when wanted 0.02 Yes 48.3 (5.1) 47 (90.4) No 42.6 (3.2) 5 (9.6) Manage your own garden 0.63 Yes 47.0 (7.6) 54 (87.1) No 48.0 (5.0) 8 (12.9) BMI kg/m2, mean (SD) 48.2 (5.21) 0.94 < 18.5 44 (0.00) 1 (1.8) 18.5 to 24.9 49.7 (5.54) 6 10.9) 25.0 to 29.9 47.2 (5.22) 21 38.2) 30.0 to 34.9 49.3 (4.60) 16 (29.1) 35.0 to 39.9 52.3 (4.32) 6 (10.9) > 40.0 43.4 (4.62) 5 (9.1) Waist to Hip ratio 48.1 (5.12) 0.03 Low <0.90 men, <0.85 women 46.8 (4.71) 12 (20.7) High \u22650.90 men, \u22650.85 women 48.4 (5.26) 46 (79.3) Depressive symptoms, GDS-15 47.8 (5.32) 0.67