Improvements in indicators of diabetes-related health status among first nations individuals enrolled in a community-driven diabetes complications mobile screening program in Alberta, Canada.

Oster, Richard T. ; Shade, Sandra ; Strong, David 等

As with many Aboriginal populations worldwide, there has been a rapid emergence of type 2 diabetes among First Nations populations. (1,2) Aboriginal prevalence rates of diabetes and its complications are reportedly 2-5 times higher than those of the general Canadian population. (2-6) The problem is expected to worsen dramatically with the increasing prevalence of obesity, metabolic syndrome and type 2 diabetes in First Nations youth. (3) In response, several community-based, primary and secondary diabetes prevention and/or screening projects have been successfully implemented in many Canadian First Nations communities. (7-10) Reported longitudinal results of such projects are scarce and have shown only limited effectiveness with respect to health outcomes. (8,10) However, modest health improvements among Aboriginals have been shown in longitudinal diabetes projects in Australia and the US. (11-13)

The Screening for Limb, I-Eye, Cardiovascular, and Kidney complications of diabetes (SLICK) project is an ongoing community-driven diabetes complications mobile screening program. This descriptive paper sought to report indicators of diabetes health status at baseline and over time among SLICK participants over a six-year time span from December 2001 to June 2007.

METHODS

SLICK travelled to 43 Alberta First Nations communities twice a year to provide diabetes complications screening, diabetes education, and community-based care starting in 2001. Rationale and methods have been reported elsewhere. (9,14) Briefly, two mobile clinics were equipped with portable lab technology and staffed by qualified health care personnel who were either First Nations (cree-speaking where possible) or had Aboriginal-specific training. Local community health representatives and home-care nurses, and/or personal care attendants collaborated within each First Nation community by assisting with each visit. Significant initial funding to SLIcK was provided by the canadian Health Infostructure partnership program (cHIpp) to University of Alberta researchers, who ran the program in collaboration with the First Nations and Health canada. SLIcK was approved by the University of Alberta Health Research Ethics Board. The project direction transitioned from the University of Alberta to Health canada in 2005.

Indicators of diabetes-related health status that were measured included height and weight (for the calculation of body mass index; BMI), waist circumference, hemoglobin A1c (HbA1c), total and fractionated cholesterol, triglycerides, blood pressure, foot abnormalities (peripheral neuropathy) and kidney damage (proteinuria or microalbuminuria). Each diabetes indicator analyzed was drawn from the recent canadian consensus for the standardized evaluation of quality improvement interventions in type 2 diabetes. (15)

Subjects were volunteers with known diabetes and enrolled through either referral by community health care workers or self-referral in response to advertising. Diabetes was confirmed by medications, chart review, or infrequently, nurse history. The vast majority of subjects had type 2 diabetes (approximately 99%). Adult subjects (18 years or older) were considered overweight if BMI was 25-29.9 and obese if BMI was [greater than or equal to] 30. (16) Children were classified as overweight or obese if BMI was between the 85th-95th percentile or [greater than or equal to] the 95th percentile, respectively. (17) Given that there are no established definitions for abnormal waist circumference in Aboriginals, the National Cholesterol Education Program (NCEP) cutoffs (males >102 cm, females >88 cm (16)) were utilized for adults. Children were considered to have abnormal waist circumferences if [greater than or equal to] the 90th percentile. (18) HbA1c was used to assess glucose control according to both the 1998 and 2003 Canadian Clinical Practice Guidelines (CPGs) cutoffs: "inadequate glucose control" (>8.4%) and "poor glucose control" (>7%), respectively. (19,20) We used cutoffs for hypercholesterolemia and hypertension of >5.2 mmol/L (total cholesterol) and >130/80, respectively. (20) Children were considered hypertensive if [greater than or equal to] the 95th percentile. (21) HbA1c and lipid concentrations were determined using the Bayer DCA2000[R]+ and the Cholestech L.D.X[TM] portable analyzers, respectively.

The presence of foot abnormalities (peripheral neuropathy) were determined by use of a microfilament wire, and World Health Organization foot grades (0 = no abnormality; 1 = sensory loss, low risk; 2 = sensory loss and presence/history of deformity, moderate risk; 3 = presence/history of plantar ulcer, high risk) were applied. (22) Kidney damage was defined as either overt proteinuria (urine dipstick >100 mg/dl) or microalbuminuria (microalbumin/creatinine ratio (ACR) >2.0 for males, >2.8 for females). Dipstick was used to screen all participants and ACR was measured (via Bayer DCA2000[R]+) only on those who were dipstick negative.

[FIGURE 1 OMITTED]

Mean baseline indicators of diabetes-related health were calculated and compared between genders. Also, mean baseline and subsequent indicators were compared for individuals over time, and clinical parameters at baseline visits, per year, were also examined, looking for a secular trend. Compared to baseline subjects, the longitudinal group (those with more than one visit) differed only in age (53.6 and 56.3 years, respectively) and duration of diabetes (8.8 and 10.5 years, respectively). Statistical analyses were done with SAS 9.1 (SAS Institute Inc, Cary, NC) and SPSS 17.0 (SPSS Inc, Chicago, IL). Results were considered significant if p<0.05. Standard deviations and 95% confidence intervals were determined for mean and prevalence values accordingly. Gender comparisons were conducted using chi-square tests for categorical variables. For continuous variables, we used a logistic regression model, where the variables included covariates. For the longitudinal analysis, univariate general linear mixed effect models with random client effect and fixed time (year) effect were used for continuous outcome variables and gender comparisons. Longitudinal analyses for baseline clinical parameters per year were adjusted for duration of diabetes. Categorical variables were not analyzed longitudinally due to the unbalanced nature of the data.

RESULTS

From 2001 to 2007, 2,102 unique subjects with known diabetes were screened and a total of 4,978 visits were completed, corresponding to approximately 62% of cases in Alberta's First Nations population on-reserve. (2,23) The mean and median number of visits per subject were 2.2 and 2, respectively. For returning subjects, the mean and median time span between visits were 441 and 351 days, respectively. Subjects ranged from 9 to 94 years of age, the mean being 54 years. The majority of subjects (62.5%) were female.

The mean duration of diabetes was 9 years, with females having a significantly longer mean duration than males (p<0.05; Table 1). We identified high baseline rates of overweight and obesity (91.8%), abnormal waist circumference (85%), poor and inadequate glucose control (60% and 35%), hypercholesterolemia (44.1%), hypertension (62.3%), foot abnormalities (32%) and kidney damage (37% with elevated ACR, 11.8% with proteinuria). The proportion of overweight subjects was higher among males (p<0.05); however obesity (p<0.05) and abnormal waist circumference (p<0.05) were more prevalent among females. The proportion with inadequate HbA1c concentrations was higher for males (p<0.05). Hypercholesterolemia, hypertension, foot abnormalities, abnormally high ACR, and proteinuria were more prevalent among males than females (p<0.05).

Figure 1 displays the trends over time for selected diabetes indicators in returning subjects enrolled in the SLICK program. Significant improvements in BMI, blood pressure, mean arterial pressure, total cholesterol and HbA1c concentrations were detected in returning subjects (p<0.01). No gender differences were apparent for returning subjects, with the exception of total cholesterol, which improved to a greater extent among returning females compared to males (p<0.05).

When baseline data for new unique subjects coming to SLICK were compared over time (adjusted for duration of diabetes), significant improvements were observed for average total cholesterol and HbA1c concentrations as yearly trends for these parameters declined over the six-year time span (p<0.01; Table 2). In other words, new subjects coming to SLICK had more favourable total cholesterol and HbA1c concentrations compared to prior unique subjects, suggesting secular improvements. For new unique subjects, no gender differences were observed in the change over time in diabetes indicators.

DISCUSSION

We observed improvements in several indicators of diabetes-related health status among First Nations individuals with diabetes living on-reserve in Alberta between 2001 and 2007. Significant improvements in BMI, blood pressure, total cholesterol and HbA1c concentrations were shown in those followed over time. Additionally, decreasing secular trends in total cholesterol and HbA1c concentrations were identified. These results are encouraging given the poor baseline results.

Our reported poor baseline diabetes health status of First Nations is consistent with recent studies of Canadian Aboriginals. Roughly 40% of First Nations individuals in Alberta with diabetes had HbA1c concentrations below the CPGs target (<7%), compared to 45% and 37% in First Nations populations of Eeyou Istchee (Quebec (4)) and Sandy Lake (Ontario (7)), respectively. Additionally, the detected high prevalence rates of overweight/obesity, abnormal waist circumference, hypercholesterolemia, hypertension and kidney damage are similar in other Aboriginal populations in North America and Australia, and are considerably higher than in non-Aboriginal populations. (4-7,11,12,24)

Providing modest improvements in BMI, HbA1c, blood pressure and cholesterol concentrations leads to considerable reductions in diabetes complications risk, (25,26) thus our reported decreases in diabetes-related measures seen over time can be viewed as progress. It is also favourable that additional outcomes that did not improve over time (waist circumference, risk of foot abnormalities, kidney damage/proteinuria) at least did not worsen. Various longitudinal diabetes projects in Australia and the US (11-13) have shown similar modest improvements in HbA1c concentrations, cholesterol and blood pressure in Aboriginals with and without diabetes. However in British Columbia, a two-year community-based diabetes prevention and control project utilizing behavioural and environmental change was found to be ineffective in improving diabetes indicators among First Nations individuals with or at risk for diabetes. (8) Also, a community-based diabetes prevention project in Quebec did not lead to sustained improvements in risk factors over 8 years among First Nations children. (10)

The secular data support results from cross-sectional studies of administrative data in both Canada and the US. (4,13,24) Dannenbaum et al. (4) showed that the proportion of Eeyou Istchee First Nations individuals with diabetes achieving HbA1c concentrations <7% significantly improved from 2002 to 2005. Significant improvements in HbA1c and cholesterol concentrations, as well as in blood pressure, have also been shown in American Indians and Alaska Natives with diabetes from 1995 to 2001. (13,24)

In the present study, subjects were educated about appropriate screening recommendations and the SLICK program itself. Subjects received a copy of their screening test results, were individually informed of the implications of the results, and were given recommendations regarding the need for follow-up with primary care providers. Part of SLICK's community care provision is to increase accessibility to and quality and quantity of diabetes screening services for rural First Nations people who may not have access to care. However, whether or not the improvements observed for returning subjects over time in the current study are the result of the SLICK project cannot be determined, as no control group was included. It is also probable that various other factors had an impact, particularly the updated Canadian CPGs, (20) improved diabetes treatments and care from other health providers, and non-SLICK activities of the federally funded Aboriginal Diabetes Initiative. These factors likely had an even more pronounced effect on secular trends observed for new subjects per year.

Gender differences in diabetes complications rates described in other populations have been contradictory. (27-29) Our results suggest that First Nations males may be at a higher risk for diabetes-related complications than females, despite having endured a shorter duration of the disease on average. The prevalence of inadequate HbA1c concentrations, hypercholesterolemia, hypertension, foot abnormalities and kidney damage were significantly higher among male subjects. It could be speculated that these are the results of males seeking care less often. Interestingly, however, although greater reductions in total cholesterol were observed for returning females compared to males, we did not observe any other differences between males and females regarding changes in diabetes indicators over time in both new subjects per year and returning subjects.

In addition to the lack of a control group, the findings are also limited as data regarding anti-hypertensive and lipid-lowering medications were not always reliably collected, thus the prevalences of hypertension and lipid abnormalities are likely underestimated. Also, subjects were not randomly selected and therefore were not necessarily a representative sample; the study probably (and anecdotally) included both those likely to be more health-conscious and some who had very poor care. Population-based studies examined prospectively are needed. Nevertheless, SLICK improves access to diabetes care and provides important community-based information. It should also be noted that SLICK only captures on-reserve individuals; caution must be used when extrapolating results to the diabetes health of off-reserve individuals.

Despite high baseline rates of obesity, poor HbA1c concentrations, hypercholesterolemia, hypertension, foot abnormalities and kidney damage, on-reserve Alberta First Nations subjects with diabetes followed over time showed improvements in BMI, blood pressure, total cholesterol and HbA1c concentrations. Moreover, decreasing secular trends in total cholesterol and HbA1c concentrations were observed. While these observations are encouraging, efforts such as SLICK and other community-based diabetes initiatives will need to be sustained and possibly expanded in order for improvements to be continued.

Acknowledgements: We thank all of the First Nations subjects and leaders who welcomed SLICK into their communities. Thanks are also due to the SLICK team members and the Aboriginal Diabetes Initiative committee, and to Xiufang Ye for statistical assistance.

Project Funding: Canada Health Infostructure Partnership Program (CHIPP), Health Canada and the University of Alberta.

Conflict of Interest: None to declare.

REFERENCES

(1.) Yu CH, Zinman B. Type 2 diabetes and impaired glucose tolerance in Aboriginal populations: A global perspective. Diabetes Res Clin Pract 2007;78(2):159-70.

(2.) Oster RT, Hemmelgarn BR, Toth EL, King M, Crowshoe L, Ralph-Campbell K. Diabetes and the Status Aboriginal population in Alberta. In: Alberta Diabetes Atlas 2009. Edmonton, AB: Institute of Health Economics, 2009.

(3.) Young TK, Reading J, Elias B, O'Neil JD. Type 2 diabetes mellitus in Canada's First Nations: Status of an epidemic in progress. CMAJ 2000;163 (5):561-66.

(4.) Dannenbaum D, Kuzmina E, Lejeune P, Torrie J, Gangbe M. Prevalence of diabetes and diabetes-related complications in First Nations communities in northern Quebec (Eeyou Istchee), Canada. Can J Diabetes 2008;32(1):46-52.

(5.) Molitch ME, DeFronzo RA, Franz MJ, Keane WF, Mogensen CE, Parving HH, et al. Nephropathy in diabetes. Diabetes Care 2004;27(Suppl 1):S79-S83.

(6.) Zhang Y, Lee ET, Devereux RB, Yeh J, Best LG, Fabsitz RR, et al. Pre-hypertension, diabetes, and cardiovascular disease risk in a population-based sample: The Strong Heart study. Hypertension 2006;47(3):410-14.

(7.) Hanley AJ, Harris SB, Mamakeesick M, Goodwin K, Fiddler E, Hegele RA, et al. Complications of type 2 diabetes among Aboriginal Canadians: Prevalence and associated risk factors. Diabetes Care 2005;28(8):2054-57.

(8.) Daniel M, Green LW, Marion SA, Gamble D, Herbert CP, Hertzman C, et al. Effectiveness of community-directed diabetes prevention and control in a rural Aboriginal population in British Columbia, Canada. Soc Sci Med 1999;48(6):815-32.

(9.) Virani S, Strong D, Tennant M, Greve M, Young H, Shade S, et al. Rationale and implementation of the SLICK project: Screening for limb, I-eye, cardiovascular and kidney (SLICK) complications in individuals with type 2 diabetes in Alberta's First Nations communities. Can J Public Health 2006;97(3):241-47.

(10.) Paradis G, Levesque L, Macaulay AC, Cargo M, McComber A, Kirby R. Impact of a diabetes prevention program on body size, physical activity, and diet among Kanien'keha:ka (Mohawk) children 6 to 11 years old: 8-year results from the Kahnawake Schools Diabetes Prevention Project. Pediatrics 2005;115(2):333-39.

(11.) Bailie RS, Si D, Robinson GW, Togni SJ, D'Abbs PH. A multifaceted health service intervention in remote Aboriginal communities: 3-year follow-up of the impact on diabetes care. Med J Aust 2004;181(4):195-200.

(12.) McDermott R, Tulip F, Schmidt B, Sinha A. Sustaining better diabetes care in remote Indigenous Australian communities. BMJ 2003;327(7412):428-30.

(13.) Shields R. Unofficial IHS diabetes care: Past results. IHS audit database, 2009. Available at: http://www.dmaudit.com/ResultsMain.htm (Accessed October 5, 2009).

(14.) Oster RT, Virani S, Strong D, Shade S, Toth EL. Diabetes care and health status of First Nations individuals with type 2 diabetes in Alberta. Can Fam Phys 2009;55(4):386-93.

(15.) Majumdar SR, Johnson JA, Bowker SL, Booth GL, Dolovich L, Ghali W, et al. A Canadian consensus for the standardized evaluation of quality improvement interventions in type 2 diabetes. Can J Diabetes 2005;29(3):220-29.

(16.) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive summary of the third report of the national cholesterol education program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (adult treatment panel III). JAMA 2001;285(19):2486-97.

(17.) Ogden CL, Flegal KM, Carroll MD, Johnson CL. Prevalence and trends in overweight among US children and adolescents, 1999-2000. JAMA 2002;288(14):1728-32.

(18.) Fernandez JR, Redden DT, Pietrobelli A, Allison DB. Waist circumference percentiles in nationally representative samples of African-American, European-American, and Mexican-American children and adolescents. J Pediatr 2004;145(4):439-44.

(19.) Meltzer S, Leiter L, Daneman D, Gerstein HC, Lau D, Ludwig S, et al. 1998 clinical practice guidelines for the management of diabetes in Canada. Canadian Diabetes Association. CMAJ 1998;159(Suppl 8):S1-S29.

(20.) Canadian Diabetes Association Clinical Practice Guidelines Expert Committee. Canadian Diabetes Association 2003 clinical practice guidelines for the prevention and management of diabetes in Canada. Can J Diabetes 2003;27(Suppl 2):S1-S152.

(21.) National High Blood Pressure Education Program Working Group on High Blood Pressure in Children and Adolescents. The fourth report on the diagnosis, evaluation, and treatment of high blood pressure in children and adolescents. Pediatrics 2004;114(Suppl 2):555-76.

(22.) World Health Organization Expert Committee on Diabetes Mellitus. Second report. World Health Organ Tech Rep Ser 1980;646:1-80.

(23.) Statistics Canada. Aboriginal peoples in Canada in 2006: 2006 census: First Nations people, 2006. Available at: http://www12.statcan.ca/english/census06/analysis/aboriginal/first.cfm (Accessed October 5, 2009).

(24.) Wilson C, Gilliland S, Cullen T, Moore K, Roubideaux Y, Valdez L, et al. Diabetes outcomes in the Indian health system during the era of the Special Diabetes Program for Indians and the Government Performance and Results Act. Am J Public Health 2005;95(9):1518-22.

(25.) UK Prospective Diabetes Study Group (UKPDS). Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet 1998;352(9131):837-53.

(26.) UK Prospective Diabetes Study Group (UKPDS). Tight blood pressure control and risk of macrovascular and microvascular complications in type 2 diabetes. UKPDS 38. BMJ 1998;317(7160):703-13.

(27.) Hayward LM, Burden ML, Burden AC, Blackledge H, Raymond NT, Botha JL, et al. What is the prevalence of visual impairment in the general and diabetic populations: Are there ethnic and gender differences? Diabet Med 2002;19(1):27-34.

(28.) Wandell PE, Gafvels C. Patients with type 2 diabetes aged 35-64 years at four primary health care centres in Stockholm County, Sweden. Prevalence and complications in relation to gender and socio-economic status. Diabetes Res Clin Pract 2004;63(3):195-203.

(29.) Huxley R, Barzi F, Woodward M. Excess risk of fatal coronary heart disease associated with diabetes in men and women: Meta-analysis of 37 prospective cohort studies. BMJ 2006;332(7533):73-78.

Received: December 8, 2009

Accepted: May 9, 2010

Richard T. Oster, MSC, [1] Sandra Shade, BSCN, [2] David Strong, MD, FRCPC, [3] Ellen L. Toth, MD, FRCPC [1]

Author Affiliations

[1]. Department of Medicine, University of Alberta, Edmonton, AB

[2]. Blood Tribe, AB

[3]. Calgary Health Region, Calgary, AB (Previously First Nations and Inuit Health Branch (FNIHB), Health Canada)

Correspondence and reprint requests: Dr. Ellen L. Toth, Division of Endocrinology & Metabolism, Department of Medicine, 362 Heritage Medical Research Centre, University of Alberta, Edmonton, AB T6G 2S2, Tel: 780-407-7745, Fax: 780-407-6702, E-mail: ellen.toth@ualberta.ca

Table 1. Baseline Characteristics of SLICK Subjects (total n = 2102)

                                    Male

Age                             54.0 [+ or -] 13.9
Age at diagnosis
  (years)                       45.9 [+ or -] 13.5
Duration of diabetes
  (years)                        7.9 [+ or -] 8.1 *
BMI (kg/[m.sup.2])              32.5 [+ or -] 6.1
  % Overweight               30.5% * (27.2-33.8)
  % Obese                    62.8% * (59.4-66.2)
Waist circumference
  (cm)                         110.1 [+ or -] 14.4
  % Abnormal NCEP            70.8% * (67.6-74.0)
HbA1c (%)                        8.0 [+ or -] 2.0
  % Poor                       62.4% (59.0-65.8)
  % Inadequate               38.1% * (35.0-41.8)
Total cholesterol
  (mmol/L)                       5.1 [+ or -] 1.2
  % Hypercholesterolemia     46.3% * (43.3-49.3)
Systolic blood
  pressure                     133.2 [+ or -] 17.9 *
  (mmHg)
Diastolic blood
  pressure                      80.0 [+ or -] 11.0 *
  (mmHg)
Mean arterial pressure
  (mmHg)                        97.6 [+ or -] 11.7 *
  % Hypertensive             70.3% * (67.1-73.5)
Foot risk category
  % with no
    abnormality (0)          63.0% * (59.5-66.5)
  % at high risk (3)            7.7% (5.8-9.7)
% High ACR                   49.0% * (45.0-53.0)
% Overt proteinuria          14.1% * (11.5-16.7)
% Taking insulin               20.9% (18.0-23.8)
% Taking oral agents           65.3% (61.9-68.7)

                                  Female

Age                             53.4 [+ or -] 14.6
Age at diagnosis
  (years)                       44.3 [+ or -] 14.0
Duration of diabetes
  (years)                        9.2 [+ or -] 9.0 *
BMI (kg/[m.sup.2])              33.7 [+ or -] 7.1
  % Overweight               23.3% * (21.0-25.6)
  % Obese                    68.4% * (65.9-70.9)
Waist circumference
  (cm)                         110.7 [+ or -] 14.8
  % Abnormal NCEP            94.1% * (92.8-95.4)
HbA1c (%)                        7.8 [+ or -] 1.9
  % Poor                       59.0% (56.3-61.7)
  % Inadequate               33.2% * (30.7-35.8)
Total cholesterol
  (mmol/L)                        5.1 [+ or -] 1.2
  % Hypercholesterolemia      43.0% * (39.2-46.8)
Systolic blood
  pressure                      128.7 [+ or -] 18.0 *
  (mmHg)
Diastolic blood
  pressure                       75.7 [+ or -] 10.3 *
  (mmHg)
Mean arterial pressure
  (mmHg)                         93.4 [+ or -] 11.3 *
  % Hypertensive              57.3% * (54.6-60.0)
Foot risk category
  % with no
    abnormality (0)           71.3% * (68.7-73.9)
  % at high risk (3)             5.2% (4.0-6.5)
% High ACR                    29.2% * (26.2-32.3)
% Overt proteinuria           10.1% * (8.2-12.0)
% Taking insulin                20.1% (17.9-22.3)
% Taking oral agents            63.6% (61.0-66.2)

Values are means ([+ or -] SD) or prevalence (95% confidence
intervals).

* Significant difference between genders (p<0.05).

Table 2. Diabetes Indicators of New Unique SLICK Subjects Over Time
Adjusted for Duration of Diabetes

                            2001/02          2003            2004

BMI (kg/[m.sup.2])               33.1            32.8            33.4
(n=1879)                 [+ or -] 0.2    [+ or -] 0.3    [+ or -] 0.4
                              (n=757)         (n=376)         (n=320)

Waist circumference             111.4           108.9           110.5
(cm) (n=1860)            [+ or -] 0.5    [+ or -] 0.7    [+ or -] 0.8
                              (n=754)         (n=372)         (n=318)

HbA1c (%)                         8.0             7.5             7.3
(n=1902)                 [+ or -] 0.0    [+ or -] 0.0    [+ or -] 0.0
                              (n=768)         (n=378)         (n=320)

Total cholesterol                 5.4             5.0             4.9
(mmol/L) (n=1913)        [+ or -] 0.1    [+ or -] 0.1    [+ or -] 0.1
                              (n=771)         (n=380)         (n=322)

Systolic blood                  131.6           129.9           130.6
pressure (mmHg)          [+ or -] 0.7    [+ or -] 0.9    [+ or -] 1.0
(n=1893)                      (n=769)         (n=372)         (n=318)

Diastolic blood                  77.3            76.7            77.1
pressure (mmHg)          [+ or -] 0.4    [+ or -] 0.6    [+ or -] 0.6
(n=1893)                      (n=769)         (n=372)         (n=318)

Mean arterial pressure           95.4            94.4            94.9
(mmHg) (n=1893)          [+ or -] 0.4    [+ or -] 0.6    [+ or -] 0.7
                              (n=769)         (n=372)         (n=318)

                             2005            2006            2007

BMI (kg/[m.sup.2])               33.6            33.7            33.4
(n=1879)                 [+ or -] 0.5    [+ or -] 0.5    [+ or -] 0.7
                              (n=167)         (n=157)         (n=102)

Waist circumference             109.9           109.9           110.2
(cm) (n=1860)            [+ or -] 1.1    [+ or -] 1.2    [+ or -] 1.9
                              (n=164)         (n=155)          (n=97)

HbA1c (%)                         7.5             7.3             7.2
(n=1902)                 [+ or -] 0.0    [+ or -] 0.0    [+ or -] 0.0
                              (n=171)         (n=161)         (n=104)

Total cholesterol                 4.6             4.8             4.7
(mmol/L) (n=1913)        [+ or -] 0.1    [+ or -] 0.1    [+ or -] 0.1
                              (n=172)         (n 163)         (n=106)

Systolic blood                  129.0           129.5           127.5
pressure (mmHg)          [+ or -] 1.4    [+ or -] 1.4    [+ or -] 1.8
(n=1893)                      (n=170)         (n=161)         (n=103)

Diastolic blood                  77.5            78.0            76.8
pressure (mmHg)          [+ or -] 0.8    [+ or -] 0.9    [+ or -] 1.1
(n=1893)                      (n=170)         (n=161)         (n=103)

Mean arterial pressure           94.7            95.1            93.7
(mmHg) (n=1893)          [+ or -] 0.9    [+ or -] 0.9    [+ or -] 1.1
                              (n=170)         (n=161)         (n=103)

                          Overall F       Gender-
                                         specific F
                          (p-value)      (p-value)

BMI (kg/[m.sup.2])              0.72           0.25
(n=1879)                    (p=0.61)       (p=0.94)

Waist circumference             1.60           0.57
(cm) (n=1860)               (p=0.16)       (p=0.73)

HbA1c (%)                       8.92           1.12
(n=1902)                    (P<0.01)       (p=0.22)

Total cholesterol              18.48           1.46
(mmol/L) (n=1913)           (P<0.01)       (p=0.20)

Systolic blood                  1.52           1.75
pressure (mmHg)             (p=0.18)       (p=0.12)
(n=1893)

Diastolic blood                 0.42           0.49
pressure (mmHg)             (p=0.83)       (p=0.79)
(n=1893)

Mean arterial pressure          0.68           0.94
(mmHg) (n=1893)             (p=0.64)       (p=0.45)

Values are means [+ or -] SD.