Otago Southland Diabetes Research Trust

Comparison of cardiovascular risk factors and outcomes at 5 years following diagnosis of type 2 diabetes between different age groups, Otago, New Zealand, 1997-2002.

Janet (EunHwa) Lee
Supervisors: Dr. Kirsten Coppell, Prof. Jim Mann
Edgar National Centre for Diabetes Research, Medical and Surgical Sciences,
Dunedin School of Medicine.

Abstract

Diabetes is a common and increasingly prevalent non-communicable disease. Assessment and treatment of known modifiable cardiovascular risk factors including blood pressure and lipid levels is an important aspect of diabetes management, but does not always happen. This study compared clinical measures at diagnosis and 5-years later for new cases of type 2 diabetes enrolled on Otago Diabetes Register, 1997-2002 (n=1,108). Those aged <40 years were obese and had the worst risk factors at diagnosis. Statistically significant improvements in modifiable cardiovascular risk factors were evident at 5-years, but blood glucose control deteriorated significantly despite more intensive medication. Obesity prevention is critical.

Comparison of cardiovascular risk factors and outcomes at 5 years following diagnosis of type 2 diabetes between different age groups, Otago, New Zealand, 1997-2002.
Janet (EunHwa) Lee
Supervisors: Dr. Kirsten Coppell, Prof. Jim Mann
Edgar National Centre for Diabetes Research, Dunedin School of Medicine.

Introduction

The relatively recent worldwide increase in the prevalence and incidence of type 2 diabetes mellitus (T2DM) is an international concern due to the multi-system complications and high costs. New Zealand is not an exception. The 2006/07 National Health Survey in New Zealand reported age-standardised prevalence of self-reported diabetes for adult aged over 15 years was 4.8% for men and 3.8% for women.1 The most common complication of diabetes is cardiovascular disease, which is the cause of death for up to 50% of the diabetic population.2,3 The assessment and treatment of known modifiable cardiovascular risk factors including hypertension and dyslipidaemia is an important aspect of diabetes management. Studies have established that improved glycaemic control, and blood pressure and lipid levels in T2DM reduces the risk of cardiovascular disease.4,5,6 However, whether demonstrated improvements in clinical process measures in the ‘real world’ have translated into better health outcomes is unknown.

The aim of this project was to determine whether improved diabetes care has resulted in a better cardiovascular risk factor profile and outcomes at 5-years following the diagnosis for T2DM for cases enrolled on the Otago Diabetes Register. Of particular interest is younger patients as several recent analyses show that they are more likely to have a worse array of cardiovascular risk factors compared to older age groups.7 However, as age is a risk factor, estimated 5-year cardiovascular risk can be low amongst younger ages despite the presence of modifiable risk factors, possibly providing false reassurance, and affecting intensity of treatment.

Methods

Study population

1687 new T2DM cases diagnosed between 1997 and 2002 enrolled on the Otago Diabetes Register were identified. 303 were excluded because they were enrolled more than 12 months following their year of diagnosis. A further 276 were excluded because they did not have 5-year follow-up data. Reasons for this included lost to follow up for unknown reasons, moved to a non-participating GP, moved outside the Otago region, and death. 1108 eligible cases remained. Using date of birth and year of diagnosis information from the Otago Diabetes Register, age at diagnosis was calculated for each individual. Each individual was assigned to one of four age groups: <40 years, 40-59 years, 60-79 years and 80-100 years.

Data source

The Otago Diabetes Register (ODR) was established in 1997, as part of a diabetes quality improvement project. The set up of the register has been previously described.8 In brief, data were collected annually from general practice for each enrolled individual, demographic details, including name, NHI, self-identified ethnicity, date of birth, smoking history, type of diabetes, year of diagnosis, dates and results of retinal and feet examinations, hospitalisation events, diabetes complications, diabetes medication and diabetes related medications (medication categories), dates and results of biochemistry tests (HbA1c, total cholesterol, HDL-cholesterol, triglycerides, creatinine, urine albumin creatinine ratio). Date and cause of death information was obtained from Births, Deaths and Marriages and added to the register. As hospitalisation data were incomplete for this project I checked Dunedin Hospital medical records for each eligible patient for details of first-ever cardiovascular event, and any missing clinical or biochemistry information. For those who had no evidence of a cardiovascular event prior to diagnosis of diabetes, documentation of first incident cardiovascular event within 5 years following diagnosis was searched for and validated using iSOFT; an electronic version of Dunedin Hospital medical records. Missing information was added to the register.

Cardiovascular risk assessment

Cardiovascular risk was estimated using the National Heart Foundation 5-year Cardiovascular Risk Charts for diabetic patients,9 which are based on the Framingham Risk Equation. Each person’s sex, age, smoking status, systolic blood pressure level and total cholesterol:HDL ratio were used to estimate risk. In accordance with the New Zealand guidelines, people with a previous cardiovascular event, genetic lipid disorder or overt diabetic nephropathy were assumed to have a very high risk (>20%). Cardiovascular risk was increased by a maximum of 5% (regardless of the number of criteria present in any individual) for the following groups: Maori, Pacific or Indian subcontinent ethnic groups, people with microalbuminuria and those who fulfilled the criteria for the metabolic syndrome.

Statistical analysis

Means and standard deviations or percentages were calculated for the variables of interest for the four age groups. A mixed model with participant as a random variable was used to analyse the data. An interaction term, group x time was included in the model in the first place. As the interaction terms were not significant, they were excluded. P-values for differences over time and among the groups were calculated. A p-value of p<0.05 was taken as statistically significant.

Results

Table 1 shows the baseline and follow up demographic characteristics and treatments of the eligible 1108 new cases of T2DM. The proportion of females in each group increased from 44.1% to 68.6% with increasing age. Overall, the majority self identified as NZ European (90.3%). Other ethnic groups included Maori, Pacific and Asian (ethnic data not shown). More than one-quarter of the youngest group smoked. At diagnosis diet only treatment was the most common diabetes treatment in all groups.

At baseline the youngest group showed the least favourable clinical and biochemical characteristics compared with the other groups (Table 2). As a group, they were obese (BMI=34.5 kg/m²), compared with the other groups who were overweight (BMI ≥25kg/m²- < 30kg/m²) or normal weight

Table 1. Demographic characteristics and treatments at diagnosis of T2DM and at 5 years follow-up

 


<40 years
(n=57)

40-59 years
(n=406)

60-79 years
(n= 594)

80-100 years
(n= 51)

Female

 

45.6

44.1

51.5

68.6

Current Smoking

Baseline
Follow up

26.3
29.8

17.5
15.8

8.6
9.3

3.9
7.8

Diet only treatment

Baseline
Follow up

61.4
21.1

55.0
28.6

63.3
40.1

47.1
43.1

OHA*

Baseline
Follow up

26.3
61.4

42.1
62.1

34.3
54.5

50.1
50.1

Insulin

Baseline
Follow up

8.8
5.3

2.7
4.4

2.2
3.0

2.7
3.9

OHA* and insulin combined

Baseline
Follow up

3.5
12.3

0.3
4.9

0.2
2.4

0
2.0

ACE inhibitor medication

Baseline
Follow up

17.5
33.3

29.6
55.1

41.6
60.4

45.1
58.8

Antihypertensive medication

Baseline
Follow up

7.0
12.3

12.6
23.7

23.1
35.9

17.7
35.3

Lipid modification

Baseline
Follow up

5.3
36.8

17.3
60.3

24.9
59.9

9.8
29.4

*OHA = oral hypoglycaemic agents.

(BMI≥18kg/m²- < 25kg/m²). The <40 year age group had the highest HbA1c (7.3%) and triglycerides (2.6mmol/L). Total and HDL cholesterol levels were similar across the groups.

All clinical and biochemical measures changed significantly over the 5-year study period for all age groups (Table 2). Compared with baseline, blood pressure and lipid levels improved and prescription of antihypertensive and lipid modifying medication increased in all age groups. Improvements were greatest in the <40 year age group, particularly blood pressure with systolic BP decreasing by 12.6mmHg and diastolic BP by 8.8mmHg. Blood pressure and lipid improvements in the youngest age group were accompanied by a 6.7kg weight loss and 2.3 unit reduction in BMI. BMI increased in the other age groups.

Table 2. Demographic characteristics and treatments at diagnosis of T2DM and at 5 years follow-up.

 


<40 years
(n=57)

40-59 years
(n=406)

60-79 years
(n= 594)

80-100 years
(n= 51)

p-value Change over time*

p-value difference among groups*

Weight (kg)

Baseline
Follow up

97.7 ± 38.5
91.0 ± 43.8

81.9 ± 33.9
84.3 ± 31.1

69.7 ± 32.9
75.0 ± 26.0

51.7 ± 31.7
54.0 ± 29.6

< 0.001

0.095

BMI (kg/m²)

Baseline
Follow up

34.5 ± 11.2
32.2 ± 13.4

28.8 ± 11.7
29.6 ± 10.5

25.8 ± 12.1
27.7 ± 9.5

22.0 ± 10.9
23.0 ± 10.0

< 0.001

0.425

Systolic BP (mmHg)

Baseline
Follow up

128.4 ± 14.9
115.8 ± 39.2

136.0 ± 16.2
132.7 ± 27.5

140.1 ± 18.2
136.7 ± 20.8

143.7 ± 19.7
137.3 ± 33.6

< 0.001

0.013

Diastolic BP (mmHg)

Baseline
Follow up

80.7 ± 10.9
71.9 ± 16.1

82.2 ± 10.4
77.7 ± 16.1

79.2 ± 9.6
75.8 ± 11.4

77.8 ± 12.7
72.4 ± 18.1

< 0.001

0.001

HbA1c (%)

Baseline
Follow up

7.3 ± 1.9
7.9 ± 1.7

7.2 ± 1.8
7.6 ± 1.3

6.8 ± 1.4
7.1 ± 1.2

7.0 ± 1.3
7.0 ± 1.5

< 0.001

0.213

Total Cholesterol (mmol/l)

Baseline
Follow up

5.4 ± 1.1
4.9 ± 0.8

5.7 ± 1.1
4.9 ± 1.0

5.7 ± 1.2
4.8 ± 1.0

5.7 ± 1.2
5.4 ± 1.5

< 0.001

0.100

HDL (mmol/l)

Baseline
Follow up

1.0 ± 0.2
1.2 ± 0.2

1.1 ± 0.3
1.3 ± 0.4

1.2 ± 0.3
1.4 ± 0.4

1.2 ± 0.3
1.4 ± 0.5

< 0.001

0.994

Triglycerides (mmol/l)

Baseline
Follow up

2.6 ± 3.6
1.9 ± 1.5

2.3 ± 1.4
2.1 ±1.2

2.0 ± 1.2
1.9 ± 1.2

1.9 ± 0.7
1.9 ± 1.1

0.001

0.667

Cardio vascular risk (%)

Baseline
Follow up

8.6 ± 4.6
7.4 ± 4.3

15.9 ± 6.2
17 ± 5.7

23.8 ± 6.2
23.9 ± 5.2

24.4 ± 4.9
23.7 ± 4.6

NS

NS

* adjusted for sex and age. NS = not significant

In contrast, glycaemic control worsened, despite more intense treatment, with the youngest age group showing the greatest deterioration. The difference in HbAlc between baseline and follow up decreased with increasing age, from 0.6%, in the youngest group to no difference in oldest group. Changes in prescription of diabetes treatment was most evident in the <80 years groups. The proportion prescribed diet only treatment declined, particularly in the youngest group (40.3% decrease), while higher proportions were prescribed OHA after 5 years. Prescription of OHA and insulin combined increased in all four groups, more so in the youngest group.

Estimated cardiovascular risk at baseline increased with increasing age, from 8.6% to 24.4%. After 5 years, estimated cardiovascular risk was almost unchanged in all age groups. It slightly decreased in the youngest and eldest groups by 1.2% and 0.7%, respectively, and slightly increased in the other groups.

Discussion

Blood pressure and lipid levels improved significantly with a concomitant increase in the prescribing of antihypertensive and lipid modification treatment in all age groups over the 5 years following diagnosis of T2DM. This finding, consistent with international cross-sectional studies,10,11 could be the consequence of several factors: the initiation of the Otago regional diabetes quality improvement project in 1997, which included guideline development and implementation, GP and patient education and establishment of a register to monitor quality of care. Increasing access to statins by Pharmac may also have contributed. However, HbA1c levels worsened over the same period, despite more aggressive prescription of hypoglycaemic treatment, and in the <40 year age group, significant weight loss. The most likely explanation for this observation is that diabetes is a progressive disease and that in general it worsens over time no matter how intense the treatment is. The UKPDS showed that although the intensive treatment regime maintained HbA1c at a lower level than conventional treatment, the recommended HbA1c<7% was only maintained for 4 years only.12

The level of adiposity was highest at diagnosis in the youngest group at, but despite weight loss in this group and weight gain in the other groups, they remained the most overweight group. Their greater weight may have contributed to their worse glycaemic control. Poor medication adherence is another possible explanation, but it appears from improvements in blood pressure and lipid levels that cardioprotective medication was being taken, at least most of the time.

There were clear differences in cardiovascular risk between age groups with risk increasing with age. In all age groups improvements in modifiable risk factors over 5-years made little apparent difference to estimated cardiovascular risk. The most likely reason for these observations is that age is a non-modifiable determinant of cardiovascular disease, and in general even healthy people will have a higher cardiovascular risk after 5 years.

This cohort study allowed comparisons within the same population over time, which eliminated the confounding and bias that can be introduced in cross-sectional studies.10,11 Only those T2DM patients who had data within a year of diagnosis were included, differentiating this study from other studies,10,11 which usually include a series of cross-sectional observations of patients with varying diabetes duration. The data source was predominantly primary care medical records and some data were missing, usually because measures were not completed or recorded. This problem lessened over time with the introduction of the regional quality improvement project and resultant improvements in processes of care. Missing hospitalisation data was addressed in this study by checking the medical records of eligible patients. Hospital data checks also provided an opportunity for correcting any register data errors, but there were few. The exclusion of the 276 with no 5-year data reduced the study size, and may have introduced bias.

This observational cohort study demonstrated that good diabetes management, principally in primary care, can lead to improvements in modifiable cardiovascular risk factors in T2DM patients in the 5 years following diagnosis. However, glycaemic control deteriorated despite more aggressive hypoglycaemic therapy, particularly in the <40 year age group who were obese. Possible explanations for this observation are a delay in diagnosis or a more aggressive form of the disease. Prevention of obesity, especially at young ages, and screening and early detection of diabetes in high risk obese individuals provides the most promising means to avoid or slow the progression of diabetes and its serious consequences.

Acknowledgements

I thank the Otago Medical Research Foundation and Otago Diabetes Research Trust for funding such an amazing opportunity. Many thanks to my supervisors, Dr. Kirsten Coppell for guidance and support, Prof. Jim Mann for his time and help, and Assoc. Prof. Sheila Williams for help with the statistical analysis. Ethical approval was obtained from the Lower South Regional Ethics Committee.

References

  1. Ministry of Health. A Portrait of Health: Key Results of the 2006/07 New Zealand Health Survey. Wellington: Ministry of Health. 2008.
  2. Coppell K, McBride K, Williams S. Under-reporting of diabetes on death certificates among a population with diabetes in Otago Province, New Zealand. N Z Med J. 2004;117(1207).
  3. Roper NA, Bilous RW, Kelly WF, et al. Cause-specific mortality in a population with diabetes: South Tees diabetes mortality study. Diabetes Care. 2002;25:43-8.
  4. UK Prospective Diabetes Study Group. Tight blood pressure control and risk of macrovascular and microvascular complications in type2 diabetes: UKPDS 38. BMJ. 1998;317:713-20.
  5. Pyorala K, Pedersen TR, Kjekshus J et al for Scandinavian Simvastatin Survival Study (4S) Group. Cholesterol Lowering With Simbstatin Improves Prognosis of Diabetic Patients With Coronary Heart Disease. A subgroup analysis of the Scandinavian Simvastatin Survival Study (4S). Diabetes Care. 1997;20:614-20.
  6. Heart Protection Study Collaborative Group. MRC/BHF Heart Protection Sudy of cholesterol lowering with simvastatin in 20536 high-risk individuals: a randomised placebo-controlled trial. Lancet 2002;360:7-22.
  7. Coppell K, Williams S, Anderson K, Lamb C, Mann J. Characteristics and cardiovascular risk of new cases of type 2 diabetes in Otago, New Zealand, 1998-2004. Diab Res Clin Pract 2008;82:396-401.
  8. Coppell K, Manning P. Establishing a regional diabetes register and a description of registered population after one year. N Z Med J. 2002;115(1160).
  9. New Zealand Guidelines Group. New Zealand Cardiovascular Guidelines Handbook; a summary resource for primary care practitioners. 2nd ed. Wellington: New Zealand Guidelines Group. 2009.
  10. Wan Q, Taggart J, Harris MF, Jayasinghe UW, et al. Investigation of cardiovascular risk factors in type 2 diabetes in a rural Australian Division of General Practice. MJA. 2008;182 (2):86-89.
  11. Gudbjörnsdottir S, Eeg-Olofsson K, Cederholm J, et al. Risk factor control in patients with type 2 diabetes and coronary heart disease: findings from the Swedish National Diabetes Register (NDR). Diabetic Med. 2009;26:53-60
  12. UK Prospective Diabetes Study Group. 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;352:837-53.