Chapter 1 Introduction
1.2 Study purposes
Using a random sample of nationwide dataset, the major goal of the present study is to describe the annual probability that general population have type 2 diabetes, annual probability that individuals without type 2 diabetes will develop type 2 diabetes, and annual prevalence of process and outcome measures for diabetes care in the Taiwan during 2000 to 2007. The specific objectives of this study are as follows:
1. To estimate annual prevalence and incidence rates of type 2 diabetes, and to describe their secular trends in different gender, age, insurance and urbanization degree during 2000-2007.
2. To estimate annual prevalence and incidence rates of complications in type 2 diabetes during 2000-2007.
3. To measure annual prevalence rates in laboratory tests, to describe their secular trends according to groups of different gender, age, insurance and urbanization degree during 1997-2007, and to
examine the effect of Pay-for-performance (P4P) program on these secular trends.
6
Chapter 2 Literature Review
1.1 Definition of Diabetes
WHO describes diabetes as “a chronic disease that occurs when the body cannot effectively use the insulin it produces, or when the pancreas does not produce enough insulin.” Hyperglycaemia or raised blood sugar is a common effect of uncontrolled diabetes. Over time it leads to serious damage to many of the body's systems, especially the nerves and blood vessels (WHO, 2010).
There are three types of diabetes, and they are as follows:
1. Type 1 diabetes (insulin-dependent, IDDM; or juvenile diabetes):
Type 1 diabetes is a polygenic disease, and it is usually diagnosed in children and young adults. In type 1 diabetes, the body does not produce insulin. The subsequent lack of insulin leads to increased blood and urine glucose. The classical symptoms consist of
polyuria (frequent urination), polydipsia (increased thirst),
polyphagia (increased hunger), and weight loss result. Only 5-10%
of people with diabetes have this form of the disease.
2. Type 2 diabetes (non-insulin-dependent diabetes mellitus, NIDDM;
or adult-onset diabetes):
In type 2 diabetes, either the body does not produce enough insulin or the cells ignore the insulin. It comprises 90% of people with diabetes around the world, and it is more common in the world, as well as the aged population. Symptoms may be similar to those of type 1 diabetes, but are often less marked.
7
3. Gestational diabetes:
Gestational diabetes is a condition in which women without previously diagnosed diabetes exhibit high blood glucose levels during pregnancy. The cause is unknown, but it is thought that some hormones from the placenta increase insulin resistance in the mother, causing elevated blood glucose levels. Gestational diabetes is most often diagnosed through prenatal screening, rather than symptoms.
This study focuses on the type 2 diabetes. Thus, 4 criteria of diagnosing type 2 diabetes proposed by four associations were introduced (Table 1). First, American Diabetes Association (ADA) recommended 4 tests for diagnosing diabetes (ADA, 2010): (1) diabetes symptoms: polyuria, polydipsia, polyphagia, increased fatigue, weight loss, blurred vision and casual plasma glucose 200 mg/dl (or 11.1 mmol/l), casual is defined as any time of day without regard to time since last meal or (2) fasting plasma glucose (FPG)
>126 mg/dl (or 7.0 mmol/l) or (3) plasma glucose 200 mg/dl (or 11.1 mmol/l) during an oral glucose tolerance test (OGTT). If any of these test results occur, testing should be repeated on a different day to confirm the diagnosis. Second, WHO recommended for 2 tests
diagnosing diabetes (WHO, 2006): (1) FPG ≥7.0mmol/l (or 126mg/dl) or (2) 2-h plasma glucose ≥ 11.1mmol/l (or 200 mg/dl), one of the above should exit. Third, National diabetes education program (NDEP) recommended 2 tests for diagnosing diabetes (NDEP, 2010): (1) A1C
≥ 6.5% or (2) FPG > 126 mg/dl or (3) 2-hr plasma glucose > 200 mg/dl post 75g oral glucose challenge or (4) random plasma glucose >
200 mg/dl and the symptoms of type 2 diabetes include polyuria, polydypsia and unexplained weight loss. For criteria of (1) to (3),
8
repeat test to confirm unless symptoms are present. It is preferable that the same test be repeated for confirmation. If two different tests are used and both indicate diabetes, consider the diagnosis confirmed.
If the two different tests are discordant, repeat the test above the diagnostic cut-point. Last, International Diabetes Federation (IDF) recommended two tests for diagnosing diabetes: (1) FPG>7.0 mmol/l (or >126 mg/dl) or (2) 2–h plasma glucose >11.1 mmol/l (or >200 mg/dl), one of the above should exist.
9
Table 1 Criteria for Diagnosis of Type 2 Diabetes by American Diabetes Association, World Health Organization, National Diabetes Education Program and International Diabetes Federation
American Diabetes Association
World Health Organization
National Diabetes Education Program
International Diabetes Federation
Glycated hemoglobin (HbA1C)
6.5%
Fasting plasma glucose (FPG)
≧126mg/dl (7.0 mmol/l)
≧126mg/dl (7.0 mmol/l)
> 126mg/dl (7.0 mmol/l)
> 126mg/dl (7.0 mmol/l) Oral glucose tolerance
test (OGTT)
200 mg/dl
(11.1 mmol/l)> 200 mg/dl (11.1mmol/l) Casual plasma glucose
200 mg/dl
(11.1 mmol/l)
> 200 mg/dl (11.1 mmol/l)
2-h plasma glucose
200 mg/dl
(11.1 mmol/l)
> 200 mg/dl (11.1 mmol/l)
10
2.2 Prevalence and incidence of diabetes in Taiwan
After thorough reviewing literature, a total of 14 articles estimating prevalence or incidence of type 2 diabetes in Taiwan had been reported (Table 2). Among them, a total of 10 studies, including a meta-analysis, reported prevalence rates whereas a total of 4 studies estimated incidence rates.
These prevalence estimates had been conducted in populations of Pu-Li during 1987-1988 (Chou, et al., 1992; Chou, et al., 1997), Hualien County during 1994-1995 ( Chen, et al., 1997), Ann-Lo during 1988-1990 ( Lin, et al., 1993), Kinmen during 1991-1994 (Chou, et al., 1994), Tainan in 1996 (Lu, et al., 1998) and Shonsun during 1996-1997 (Chen, et al., 2001). These studies had been conducted in various age groups, such as participants aged 20 years and over (Lu, et al., 1998), 30 years and over (Chou, et al., 1992; Chou, et al., 1997; Chou, et al., 1994), 40 old and over (Chen, et al., 1997; Lin, et al., 1993) and 50-79 years (K. T. Chen, et al., 2001). All of them were community-based studies with either random sampling or recruiting all eligible residents. The diabetes measurement used fasting glucose tolerance test ( Chen, et al., 2001; Wang, et al., 1997), 2-hr postprandial blood sugar ( Lin, et al., 1993), 75-g oral glucose
tolerance test (Chou, et al., 1994; Lu, et al., 1998; Tseng, et al., 2000), WHO criteria (Chen, et al., 1997; Chou, et al., 1992; Chou, et al., 1997), ADA criteria (Chen, et al., 1999) and ICD-9-CM in outpatient and
inpatient claim data (Chang, et al., 2010).
The prevalence of diabetes was different according to age, gender and races. These studies conducted before 1990 reported crude
prevalence rates ranged from 2.6% to 6.9% (Chou, et al., 1992; Lin, et al., 1993) and those after 1990 ranging from 5.6% to 11.0% (Chang, et al.,
11
2010; Chou, et al., 1997; Chou, et al., 1994; Lu, et al., 1998; Tseng, et al., 2006; Wang, et al., 1997). In studies comparing prevalence rates of
various race groups, one reported age-adjusted prevalence during
1994-1995 for Han Chinese was 9.8% in men and 12.3% in women; for aboriginal groups was 11.5% in men and 8.5% in women (Chen, et al., 1997); the other reported the age-standardized prevalence during 1996-1997 for Hakaas was 17.9% in men and 15.5% in women; for Fukienese was 14.5% in men and 12.8% in women; for aborigines was 10.0% in men and 13.3% in women (Chen, et al., 2001).
For incidence estimates, they had been determined in Chu- Dung and Pu-Tzu townships during 1993-1996 (Wang, et al., 1997), in Pu-Tai
(Tseng, et al., 2000), in the entire Taiwan during 1992-1996 (Tseng, et al., 2006) and during 1999-2004 (Chang, 2010). All of these studies were community-based (Chang, et al.,2010; Tseng, et al., 2000; Tseng, et al., 2006; Wang, et al., 1997). The source of participants were 35-74 years (Wang, et al., 1997), ≧35 years (Tseng, et al., 2006). The diabetes status had been determined by FPG and self-report diabetic medication (Wang, et al., 1997), OGTT (Tseng, et al., 2000), self-report or ICD-9-CM in outpatient and inpatient claim data (Chang, et al., 2010).
Two of these studies reported cumulative incidence and two reported incidence density. Two studies reporting cumulative incidence had been conducted in the same time period but they had different estimates. One reported a cumulative incidence of 9.8 per 1000 persons per year in men and 9.0 per 1000 persons per year in women during 1993-1996 (Wang, et al., 1997). The other reported a cumulative incidence of 1.871 per 1000 persons per year in men and 2.184 per 1000 persons per year in women (Tseng, et al., 2006). The other two studies reporting incidence density were different in their sample size. The one with small sample size
12
reported an overall incidence density rate of 27.4 per 1000 person-years (Tseng, et al., 2000), whereas the one with large sample size reported the age-standardized incidence density rate of approximately 7.6 per 1000 person-years in men and 7.7 to 6.9 per 1000 person-years in women during 1999-2004 (Chang, et al., 2010).
13
Table 2 Previous studies estimated prevalence and/or incidence of type 2 diabetes in Taiwan, 1993-2010
©
Authors Objectives Study
design
DM measurement
Source of participants
Sample size Findings
(Chou, et al., 1992)
Prevalence of type 2 diabetes
(Chou, et al., 1992)
Chou, P.et al., 1992
To determine the prevalence and possible risk factors associated with diabetes in Pu-Li from 1987-1988
Cross-sectional study
WHO criteria Residents aged 30 yrs and over in Pu-Li selected by stratified cluster sampling: 1152 registered residents
The age-adjusted prevalence:
6.9% for previous diabetes 4.4% for new diabetes
(
To determine the prevalence and possible risk factors associated with diabetes in Pu-Li from 1987-1988 and 1991-1992
Cross-sectional study
WHO criteria Residents greater than or equal to 30 yr of age in Pu-Li selected by stratified cluster sampling
1987-1988: 1152 registered residents 1991-1992:2719 registered residents
To determine the prevalence of diabetes in three different ethnic groups in Hualien County in the eastern Taiwan during 1994-1995
Cross-sectional study
WHO criteria Six rural village inhabitants aged 40 years and over of Han Chinese, aboriginal Ami and aboriginal Atayal selected by random sampling: 1013 adults (460 men and 553 women)
To estimate the prevalence of type 2 diabetes and IFG in Penghu, Taiwan
Cross-sectional study
ADA criteria Residents 40-70 years of age in Penghu Islands, Taiwan- 2500 residents
Age-adjusted prevalence by world adult population : 16.8% (95% CI 15.0–18.6)
(C. Chang, et al., 2000)
Chang, et al., 2000
To investigate the prevalence of diabetes and complications in Taiwan from1985 to 1996
Meta-analysis Paper review
(
To investigate the prevalence of diabetes in the Ann-Lo district (northern Taiwan) from July 1988 to June 1990
Cross-sectional study
2-hr postprandial blood sugar
Residents ≧40 years of agein Ann-Lo district, a suburban area of Northern Taiwan: 9087 subjects
Overall prevalence: 2.6%;
Residents aged 40 years and older:
8.0%
To investigate the prevalence of type 2 diabetes in Kin-Hu, Kinmen in 1991-1994
Cross-sectional study
75-g oral glucose tolerance test
Residents ≧30 years of age in Kin-Hu, Kinmen: 7415 eligible residents
Crude prevalence rate : 6.5%
Age-adjusted prevalence: 4.9%
IFG: impaired fasting glucose;
14
Table 2 Previous studies estimated prevalence and/or incidence of type 2 diabetes in Taiwan, 1993-2010 (Continued)
©
Authors Objectives Study
design
DM measurement
Source of participants
Sample size Findings
(L
To investigate the prevalence of diabetes in southern Taiwan in 1996
Cross-sectional study
75-g oral glucose tolerance test
Residents aged ≧20 years in Tainan city selected by stratified systematic cluster sampling: 1638 subjects (780 men and 858 women) in Taiwan from1996 to 1997
Cross-sectional study
Fasting plasma glucose
Residents aged 50-79 years in three townships of Shonsun, Kuanhsi and Fushin selected by random sampling: 1293 persons (468 Hakaas, 440 Fukienese, and 385 Aborigines)
Age-adjusted prevalence
Hakaas: 17.9% in men and 15.5%
in women
Fukienese: 14.5% in men and 12.8% in women
Aborigines: 10.0% in men and 13.3% in women
To evaluate annual prevalence and incidence of type 2 diabetes and to examine possible trends among adults in Taiwan from1999 to 2004
Retrospective
Insurers aged ≧20 years from Taiwan NHIRD: 15,270,726-16,709,375 insurers from 1999 to 2004
The age-standardized prevalence Men: 4.7%-6.5%
Women: 5.3%-6.6%
Incidence of type 2 diabetes
(
To determine type 2 diabetes incidence in Taiwan;
The first survey: 1990-1993, The second survey: 1993-1996
Cohort Study First survey : FPG;
Second survey:
diabetic medication
Residents aged 35-74 years free from diabetes of two townships selected by random sampling (Chu-Dung and Pu-Tzu):
2190 subjects (995 men and 1195 women)
Crude incidence rates:
To investigate the incidence of type 2 DM in Taiwan
Cohort Study Oral glucose tolerance test
Non-diabetic residents in Pu-Tai for a period of up to four years: 446 residents
Incidence density rate:
27.4 per 1000 person years
(Tseng , et al., 2006)
Tseng, et al., 2006
To assess the yearly incidence of type 2 diabetes in Taiwan from1992 to 1996
Cohort study Telephone interviews of 93,484 diagnosed diabetic patients
Patients aged ≧35 years selected from Taiwan NHIRD,
Population from household registration system
Incidences
Men: 87.1 per 100,000 persons Women: 218.4 per 100,000 persons
(C. H.
Chang, 2010)
Chang, et al., 2010
To evaluate annual prevalence and incidence of type 2 diabetes and to examine possible trends among adults in Taiwan from1999 to 2004
Retrospective
Insurers aged ≧20 years from Taiwan NHIRD: 15,270,726-16,709,375 insurers
Age-standardized incidence rates Men: 7.6 per 1000 person-years and
then remain stable Women: 7.7 to 6.9 per 1000
person-years
NHIRD: National Health Insurance Research Database; FPG: fasting plasma glucose;
15
2.3 Prevalence and incidence of complications in patients with type 2 diabetes
Type 2 diabetes is an important cause of complications (Lusignan, et al., 2005), the consequences of which include blindness, kidney damage, and foot ulcers that can result in amputation (IDF, 2010). Diabetic
retinopathy is an important cause of blindness (Haik, et al., 1989), and it occurs as a result of long-term accumulated damage to the small blood vessels in the retina. After 15 years of diabetes, approximately 2% of people become blind, and about 10% develop severe visual impairment.
Diabetes is among the leading causes of kidney failure and 10-20% of people with diabetes die of kidney failure. Diabetic neuropathy is damage to the nerves as a result of diabetes, and affects up to 50% of people with diabetes. Many different problems can occur as a result of diabetic
neuropathy. Its common symptoms are tingling, pain, numbness, or weakness in the feet and hands. The overall risk of dying among people with diabetes is at least double the risk of their peers without diabetes (WHO, 2010). Diabetes increases the risk of heart disease and stroke, and 50% of people with diabetes die of cardiovascular disease, primarily heart disease and stroke.
The studies investigating the prevalence and incidence of
complications in type 2 diabetes can be divided into acute complications, microvascular diseases and macrovascular diseases (Fasanmade, et al., 2008; Kar, et al., 2008; Rosolova, et al., 2008). In Taiwan, the studies invested the prevalence and incidence of complications in type 2 diabetes (Table3). There were five and five articles, reporting prevalence and/or incidence of amicrovascular diseases, and macrovascular diseases, respectively.
16
Ketoacidosis is an acute complication of diabetes. A retrospective cohort study analyzed the occurrence of diabetic ketoacidosis in Chinese adults from 1992 to 1997 and it showed 54.6% of type 2 diabetes had events of ketoacidosis (Yan, et al., 2000).
Previous studies reporting the prevalence and incidence of the
microvascular disease included diabetic retinopathy (Chang, et al., 2000;
Tung, et al., 2007), renal disease (Chang, et al., 2000; Lin, et al., 2007;
Shen, et al., 2009), neuropathy (Chang, et al., 2000; Hsu, et al., 2009).
Among these studies, there were 5 studies estimating prevalence.
However, these prevalence studies were based upon small size of samples and most of them were conducted in outlying islander. In Lin’s study, they estimated that the prevalence of renal impairment in patients with diabetes in Kinmen County in 1999-2001 was 15.1% ( Lin, et al., 2007);
the prevalence of diabetic retinopathy in Kinmen County was 18.5% in 1999-2002 (Tung, et al., 2007); the prevalence of neuropathy in Mastu islanders with type 2 diabetes was 9.0% (Hsu, et al., 2009). The
nephropathy prevalence was 21.8 % in severe albuminuria, 9.8% in insulin use and 35.3% in use of albumin excretion rate blockades in 2004 (Shen, et al., 2009).
Previous studies reported prevalence or incidence of the macrovascular diseases included amputation (Chen, et al., 2006), myocardial infarction (Hsiao, et al., 2009), stroke (Hsiao, et al., 2009;
Tseng, et al., 2000; Tseng, et al., 2005), coronary heart disease (Chang, et al., 2000) and peripheral artery disease (Chang, et al., 2000). Two of them reported prevalence of ischemic heart disease, stroke, leg vessel disease and large vessel disease and their prevalence rates were 15.8%, 0.4-11.8%, 1.7% and 20%, respectively (Chang, et al., 2000; Tseng, et al., 2005). In Chen’s study, they found theincidence density of non-traumatic
17
lower-extremity amputation for diabetic men and women were 4.103 and 3.170 per 1,000 patient-years from 1997 to 2002, respectively (Chen, et al., 2006). The estimated incidence that had been reported were 6-year cumulative incidence of 2.10% and 1.68% in hemorrhagic stroke for diabetic men and women (Chen, et al., 2009), of 12.71% and 8.89% in MI and 0.80 and 0.41 in stroke for patients with rosiglitazone
monotherapy and pioglitazone monotherapy, respectively (Hsiao, et al., 2009).
18
Table 3 Previous studies estimated prevalence and/or incidence of complications in patients with type 2 diabetes in Taiwan, 1992-2005
Authors
Authors Objectives Study design Complications
measurement Source of participants Findings Microvascular diseases
(C. Chang, et al., 2000)
Chang et al., 2000
To investigate the prevalence of DM and its complications in Taiwan from 1985 to 1996
Meta-analysis Paper review Diabetic retinopathy: 35%
Nephropathy: 12.9%
To estimate the prevalence of renal impairment in type 2 diabetes patients in 1999–2001
Cross-sectional study
GFR less than 60 ml/min per 1.73 m2
Residents with diabetes in Kinmen County: 763 diabetics
The prevalence of renal impairment Overall: 15.1%
To explore whether insulin resistance and beta-cell dysfunction are both related to diabetic retinopathy in type 2 diabetics between 1999 and 2002
Cross-sectional
Residents with diabetes in Kinmen County: 715 diabetics
Diabetic retinopathy at first eye screening: 18.5%.
(Hsu, et al.,
2009)
Wei-Chih Hsu et al., 2009
To investigate the prevalence of autonomic neuropathies concurrently in pre-diabetic and diabetic subjects from October 2002 to December 2003
Cross-sectional study
Both with an abnormal sural NCS and an abnormal peroneal NCS
All adult residents aged older than 30 years in Matsu islands with type 2 diabetes: 133 type 2 diabetics
The prevalence rates Definite neuropathy: 9.0%
To determine the prevalence of diabetic nephropathy in type 2 diabetes in 2004
Cross-sectional study
Urinary albumin excretion rate,
Patients with type 2 diabetes from Chang Gong Memorial
Hospital-Kaohsiung Medical Center:1069 patients
The nephropathy prevalence In severe albuminuria: 21.8 % In insulin use: 9.8%
In use of albumin excretion rate blockades: 35.3%
NHIRD: National Health Insurance Research Database;
19
Table 3 Previous studies estimated prevalence and/or incidence of complications in patients with type 2 diabetes in Taiwan, 1992-2005 (Continued)
Macrovascular diseases
Autho rs
Authors Objectives Study design Complications measurement
Source of participants Findings
(C.
To investigate the prevalence of DM and it complications in Taiwan from 1985 to 1996
Meta-analysis Paper review Leg vessel disease: 1.7%
Large vessel disease: 20%
Ischemic heart disease: 15.8%
Stoke: 2.5%
v C-H
Tseng et al., 2005
To determine the prevalence for stroke in patients with Type 2 diabetes from 1995-1998
Cross-sectional study
Diagnosis by a physician or conform to the definition made by the WHO
Insurers with type 2 diabetes aged 45 years and over from NHIRD: 12,531 type 2 diabetics
Prevalence of stroke in type 2 diabetic Women-< 45 years old: 0.4%
To investigated the age- and sex-specific incidence density of lower-extremity amputation (LEA) of the diabetic population in Taiwan from 1997-2002
Prospective study LEA (ICD-9:
84.1 ,84.10–84.18)
Insurers aged 30 years and over with diabetes from Taiwan NHIRD: 500,868 diabetic patients
Estimated incidence density of non-traumatic LEA Men: 4.103 per 1,000 patient-years
Women: 3.170 per 1,000 patient-years (H.
To explore the impact of gender on incidence of hemorrhagic and ischemic stroke among the diabetic population in Taiwan from 1997 to 2002.
Prospective study Nontraumatic
hemorrhagic stroke:
ICD-9: 430–432) ischemic stroke: ICD-9:
433–438
Insurers with type 2 diabetes aged 35 years and over from NHIRD: 500,868 diabetic patients
The 6-year cumulative incidence
Hemorrhagic stroke-Men: 3.55%; Women: 2.83%
Ischemic stroke-Men: 30.46%; Women: 30.06%
(Hsiao, et
al., 2009) F-Y Hsiao et al., 2009
To investigate the association between oral
antihyperglycaemics with MI and stroke from 2000 to 2005
Retrospective
Insurers with type 2 diabetes aged 35 years and over from NHIRD: 473 483 with type 2 diabetes
MI incidence rates
In rosiglitazone monotherapy: 12.71%
In pioglitazone monotherapy: 8.89%
Stroke incidence rates
In rosiglitazone monotherapy: 0.80%;
In pioglitazone monotherapy: 0.41%;
NHIRD: National Health Insurance Research Database;
20
2.4 The importance of time trend studies
Time trend is one of seven uses of epidemiology (Thomas, 2008). It is an important method in epidemiology, which studies past and future trend in human and illness from the rise and fall of disease and changes in their characters. From time trend study we can make useful projections into the future.
Several previous studies have examined time trend of diabetes worldwide, and aggregating evidence across studies had shown that incidence (Burke, et al., 1999; Chang, 2010; Dahlquist, et al., 2000;
Harjutsalo, et al., 2008; Ishak, et al., 2003; Jansson, et al., 2007;
Lipscombe, et al., 2007; Tseng, et al., 2006)and prevalence (Chang, 2010;
Cooper, et al., 2000; Jansson, et al., 2007; Lipscombe, et al., 2007;
Lusignan, et al., 2005; Mokdad, et al., 2000) of diabetes were increasing annually.
We had summarized some scholars who used time trends to describe the trends in prevalence and incidence of either type 1, type 2 or gestation diabetes after 1999 (Table 4). Among them, there were a total of 4 studies reporting type 1 diabetes, including those conducted in Oxford from 1985 to 1996 (Gardner, et al., 1997), in Sweden from 1978 to 1997 (Dahlquist, et al., 2000), in Yorkshire from 1978 to 2000 (Cohen, et al., 2003) and in Finnish from 1980 to 2005 (Harjutsalo, et al., 2008). Using nationwide registration data, an annual increase of 4% was reported from 1985 to 1996 in Oxford (Gardner, et al., 1997), a mean annual increase 1.7% from 1985 to 1996 was observed in Sweden (Dahlquist, et al., 2000), and an average annual increase of 2.9% and 5.9% was reported from 1978 to 2000 in kids aged 0-14 and 15-29 year olds in West Yorkshire (Cohen, et al., 2003). A cohort study in Finnish, including children who were
younger than 15 years had an average age-standardized incidence from 31.4 to 64.2 per 100 000 persons per year from 1980 to 2005 (Harjutsalo,
21
et al., 2008).
A total of 7 studies took advantage of time trends to describe trends of type 2 diabetes in San Antonio Heart Study between 1979 and 1996 (Burke, et al., 1999), in USA between 1990 and 1998 (Mokdad, et al.,
A total of 7 studies took advantage of time trends to describe trends of type 2 diabetes in San Antonio Heart Study between 1979 and 1996 (Burke, et al., 1999), in USA between 1990 and 1998 (Mokdad, et al.,