The British Diabetic Association Cohort Study, I: all-cause mortality in patients with insulin-treated diabetes mellitus.

AIMS: To assess mortality in patients with diabetes incident under the age of 30 years. METHODS: A cohort of 23 752 diabetic patients diagnosed under the age of 30 years from throughout the United Kingdom was identified during 1972-93 and followed up to February 1997. Following notification of deaths during this period, age- and sex-specific mortality rates, attributable risks and standardized mortality rates were calculated. RESULTS: The 23 752 patients contributed a total of 317 522 person-years of follow-up, an average of 13.4 years per subject. During follow-up 949 deaths occurred in patients between the ages of 1 and 84 years, 566 in males and 383 in females. All-cause mortality rates in the patients with diabetes exceeded those in the general population at all ages and within the cohort were higher for males than females at all ages except between 5 and 15 years. The relative risk of death (standardized mortality ratio, SMR), was higher for females than males at all ages, being 4.0 (95% CI 3.6-4.4) for females and 2.7 (2.5-2.9) for males overall, but reaching a peak of 5.7 (4.7-7.0) in females aged 20-29, and of 4.0 (3.1-5.0) in males aged 40-49. Attributable risks, or the excess deaths in persons with diabetes compared with the general population, increased with age in both sexes. CONCLUSIONS: This is the first study from the UK of young patients diagnosed with diabetes that is large enough to calculate detailed age-specific mortality rates. This study provides a baseline for further studies of mortality and change in mortality within the United Kingdom.     

Mortality of South Asian patients with insulin-treated diabetes mellitus in the United Kingdom: a cohort study.

AIMS: To investigate mortality in South Asian patients with insulin-treated diabetes and compare it with mortality in non South Asian patients and in the general population. METHODS: A prospective cohort study was conducted of 828 South Asian and 27 962 non South Asian patients in the UK with insulin-treated diabetes diagnosed at ages under 50 years. The patients were followed for up to 28 years. Ethnicity was determined by analysis of names. Standardized mortality ratios (SMRs) were calculated, comparing mortality in the cohort with expectations from the mortality experience of the general population. RESULTS: SMRs were significantly raised in both groups of patients, particularly the South Asians, and especially in women and subjects with diabetes onset at a young age. The SMRs for South Asian patients diagnosed under age 30 years were 3.9 (95% CI 2.0-6.9) in men and 10.1 (5.6-16.6) in women, and in the corresponding non South Asians were 2.7 (2.6-2.9) and 4.0 (3.6-4.3), respectively. The SMR in women was highly significantly greater in South Asians than non South Asians. The mortality in the young-onset patients was due to several causes, while that in the patients diagnosed at ages 30-49 was largely due to cardiovascular disease, which accounted for 70% of deaths in South Asian males and 73% in females. CONCLUSIONS: South Asian patients with insulin-treated diabetes suffer an exceptionally high mortality. Clarification of the full reasons for this mortality are needed, as are measures to reduce levels of known cardiovascular disease risk factors in these patients.     

The British Diabetic Association Cohort Study, II: cause-specific mortality in patients with insulin-treated diabetes mellitus.

AIMS: To measure cause-specific mortality, by age, in patients with insulin-treated diabetes incident at a young age. METHODS: A cohort of 23 752 patients with insulin-treated diabetes diagnosed under the age of 30 years, from throughout the United Kingdom, was identified during 1972-93 and followed to February 1997. Death certificates have been obtained for deaths during the follow-up period and cause-specific mortality rates and standardized mortality ratios by age and sex are reported. RESULTS: During the follow-up period 949 deaths occurred and at all ages mortality rates were considerably higher than in the general population. Acute metabolic complications of diabetes were the greatest single cause of excess death under the age of 30 years. Cardiovascular disease was responsible for the greatest proportion of the deaths from the age of 30 years onwards. CONCLUSIONS: Deaths in patients with diabetes diagnosed under the age of 30 have been reported and comparisons drawn with mortality in the general population. To reduce these deaths attention must be paid both to the prevention of acute metabolic deaths and the early detection and treatment of cardiovascular disease and associated risk factors.     

Lipid but not glycaemic parameters predict total mortality from Type 2 diabetes mellitus in Canterbury,New Zealand

A cohort of 447 subjects with Type 2 diabetes mellitus (208 male, 239 female; age range 30–82, median 62 years; and of predominantly European origin) was characterized in a clinic survey in 1989. Individual status (dead or alive) at 1 June 1995 was ascertained. Mortality rates were compared with the general New Zealand population by calculating standardized mortality ratios (SMR) and the hazard ratio (HR) of prognostic factors evaluated with Cox’s proportional hazards model. At 6 years, 289 subjects were confirmed as alive and 133 as dead; only 25 were untraceable. Six-year survival for the cohort was 70 % (95 % CI 66–74). SMR was 2.53 (95 % CI 1.99–2.68) for the female cohort and 2.03 (95 % CI 1.60–2.59) for the male cohort. Factors assessed at baseline (1989) that were independently prognostic of total mortality included age, male sex, pre-existing coronary artery disease (CAD) (HR 2.2, 95 % CI 1.5–3.3) and plasma cholesterol (HR for 1.4 mmol l⁻¹ change: 1.49, 95 % CI 1.2–1.9). HDL-cholesterol was protective in women (HR for 0.4 mmol l⁻¹ change: 0.72, 95 % CI 0.51–1.00) but not men. Glycated haemoglobin was not a significant predictor of total mortality. Predictors of CAD mortality (in those subjects free of CAD in 1989) included plasma cholesterol (HR for 1.4 mmol l⁻¹ change: 1.86 95 % CI 1.20–2.89), glycated haemoglobin (HR for 1.8 % change: 1.9 95 % CI 1.04–3.47), male sex, peripheral vascular disease, and smoking. There is therefore increased mortality in Type 2 diabetic subjects in Canterbury, New Zealand. HDL-cholesterol is protective against total mortality in females. © 1998 John Wiley & Sons, Ltd.     

High impact of nephropathy on five-year mortality rates among patients with Type 2 diabetes mellitus from a multi-ethnic population in New Zealand.

AIMS: Type 2 diabetes mellitus and its complications are common among Polynesians in New Zealand. This study investigated the mortality from diabetes among indigenous Maori and recent migrants from the South Pacific. METHODS: Death certificates and other reports were collected to enumerate those who had died in an across-community cohort study of 765 diabetic patients aged 40-79 years in 1991. Five year mortality status was ascertained in 99.7% and death certificates were obtained from 129 (88%) of the 146 who had died. Diabetes was missed from 36% of death certificates. RESULTS: Compared to Europeans with Type 2 diabetes, Maori with Type 2 diabetes were 2.66 (1.63-4.35) fold as likely to die from diabetes-related conditions, including a 13.1 (3.7-46.4) fold greater risk of death from nephropathy. Pacific Islands Polynesians with Type 2 diabetes had a similar mortality to Europeans with Type 2 diabetes (hazards ratio 1.06 (0.68-1.65)). After 6 years, 10.7 (2.2-19.3)% more Maori had died than Pacific Islands Polynesians. CONCLUSIONS: Maori with Type 2 diabetes are dying from diabetic complications, particularly nephropathy, at an alarming rate. The magnitude of the difference between Maori and Pacific Islands Polynesians suggests environmental rather than inherited factors are involved and these need further investigation.     

Mortality in Israeli Jewish patients with Type 1 (insulin-dependent) diabetes mellitus diagnosed prior to 18 years of age: a population based study

Summary A total of 614 Jewish patients under the age of 18 with Type 1 (insulin-dependent) diabetes mellitus, diagnosed in Israel during the period 1 January 1965 to 31 December 1979, were identified by exhaustive screening of all possible sources. Mortality experience of this cohort was updated to 31 March 1988 through the Central Population Registry and 14 deaths were identified. The ascertainment rate for diagnosed cases as well as for deaths is estimated at about 95%. There was a significantly higher (p<0.001) by 3.2-fold excess mortality relative to the age and sex-adjusted mortality as expected the general Jewish population in Israel. This excess was due to three cause-of-death categories: diabetic ketoacidosis (n = 3; p<0.001), cardiovascular diseases (n – 3; p<0.001) and infections (n = 2; p = 0.03). The rate of malignancies (n = 2), external causes (n = 3) and other general causes (n = 1) did not differ significantly from that expected. During the first 15 years of the disease cumulative mortality resembled that of the general population, with a subsequent steep increase so that by 20 years disease duration, the rate was four-fold higher than expected. This mortality pattern was similar irrespective of age at onset, sex and ethnic group (Ashkenazi vs non-Ashkenazi Jews). A factor contributing to the lack of increase in mortality rate in the first 15 years of Type 1 diabetes may be the comprehensive multidisciplinary treatment approach employed for most juvenile diabetic patients in Israel leading to early referral and an overall better metabolic control.The study was supported by a grant from the NIH — National Institute of Diabetes and Digestive and Kidney Diseases, No. 5 RO1 DK3905-04.     

Insulin use and increased risk of mortality in type 2 diabetes: a cohort study

Aim: To compare population‐based rates of all‐cause and cardiovascular (CV) mortality in newly treated patients with type 2 diabetes according to levels of insulin exposure. Methods: Using the administrative databases of Saskatchewan Health, 12272 new users of oral antidiabetic therapy were identified between 1991 and 1996 and grouped according to cumulative insulin exposure based on total insulin dispensations per year: no exposure (reference group); low exposure (0 to <3); moderate exposure (3 to <12) and high exposure (≥12). Time‐varying multivariable Cox proportional hazards models were used to examine the relationship between insulin exposure and all‐cause, CV‐related and non‐vascular mortality after adjustment for demographics, medications and comorbidities. Results: Average age was 65 (s.d. 13.9) years, 45% were female, and mean follow‐up was 5.1 (s.d. 2.2) years. In total, 1443 (12%) subjects started insulin, and 2681 (22%) deaths occurred. The highest mortality rates were in the high exposure group; 95 deaths/1000 person‐years compared with 40 deaths/1000 person‐years in the no exposure group [unadjusted hazard ratio (HR): 2.32; 95% confidence interval (CI): 1.96–2.73]. After adjustment, we observed a graded risk of mortality associated with increasing exposure to insulin: low exposure [adjusted HR (aHR): 1.75; 95% CI: 1.24–2.47], moderate exposure (aHR: 2.18; 1.82–2.60) and high exposure (aHR: 2.79; 2.36–3.30); p = 0.005 for trend. Analyses restricted to CV‐related (p = 0.042 for trend) and non‐vascular (p = 0.004 for trend) mortality showed virtually identical results. Conclusions: We observed a significant and graded association between mortality risk and insulin exposure level in an inception cohort of patients with type 2 diabetes that persisted despite multivariable adjustment.     

Criteria for previously undiagnosed diabetes and risk of mortality: 15-year follow-up of the Edinburgh Artery Study cohort.

AIMS: To compare risk of all-cause and cardiovascular mortality associated with different criteria for undiagnosed diabetes and glucose tolerance. METHODS: A population-based cohort of 758 men and 738 women of 55-74 years of age who had an oral glucose tolerance test or known diabetes at baseline were followed up until death or for 15 years. Mortality outcomes were compared by baseline diabetes status using people with normal glucose tolerance (i.e. those without diabetes, impaired fasting glucose or impaired glucose tolerance) as the reference group. RESULTS: Prevalence of undiagnosed diabetes using World Health Organization (WHO) criteria (fasting glucose of > or = 7.0 mmol/l and/or a 2-h post-challenge glucose of > or = 11.1 mmol/l) was 6.6%, of which 81% was associated with fasting glucose > or = 7.0 mmol/l and 19% was associated with isolated post-challenge hyperglycaemia. Hazard ratios (95% CI) for all-cause mortality adjusted for age and sex were 1.51 (1.09-2.08) for new diabetes by the American Diabetes Association (ADA) criterion (fasting glucose of > or = 7.0 mmol/l regardless of post-challenge glucose), 1.60 (1.20-2.13) for new diabetes by WHO criteria and 1.98 (1.14-3.44) for isolated post-challenge hyperglycaemia. Hazard ratios (95% CI) for cardiovascular mortality adjusted for age and sex were 1.89 (1.17-3.00), 1.73 (1.12-2.66) and 1.08 (0.34-3.40) for new diabetes by ADA and WHO criteria and for isolated post-challenge hyperglycaemia, respectively. CONCLUSIONS: Undiagnosed diabetes was associated with increased risk of all-cause mortality by any criteria but significantly increased cardiovascular disease mortality was only associated with diabetes diagnosed using the fasting glucose criterion. Mortality risks were similar in this population using either ADA or WHO criteria for diagnosis of diabetes.     

Clinical and biochemical outcomes of Type 2 diabetes mellitus in Canterbury, New Zealand: a 6-year cohort study

There is a paucity of data regarding outcomes of Type 2 diabetes mellitus. A cohort of 447 Type 2 diabetic subjects (208 male, 239 female; age range 30–82 years, median 62 years; and of predominantly European origin) was characterised in a clinic survey in 1989. Individual status (dead or alive) at 1 June 1995 was ascertained. At 6 years, 289 subjects were confirmed as alive and 133 as dead—only 25 were untraceable. Of those subjects identified as alive, follow-up clinical and biochemical data were obtained for 253 (87.5%) individuals. In those subjects, glycated haemoglobin deteriorated from 63.1±18.7 mmol/mol haem in 1989 to 71.7±24.4 in 1995, P<0.0001. An increased prevalence of retinopathy was evident at 6-year follow-up, 59.7% cases in 1995 compared with 39.5% in 1989, P<0.001. Similarly there was an increased prevalence of coronary artery disease (CAD) (33.6 vs 18.2% of cases), albuminuria (26.5 vs 19% of cases; P<0.001), and hypertension (71.5 vs 54.9% of cases; P<0.001) in 1995 vs 1989, respectively. Multiple logistic regression analysis showed that glycated haemoglobin (odds ratio (OR) for 18 mmol/mol haem change, 1.78; 95% CI, 1.15–2.85), hypertension (OR, 3.33; 95% CI, 1.40–8.41) and known duration of diabetes (OR for 7 year change, 2.12; 95% CI, 1.24–3.80) were predictors for development of retinopathy. There is therefore a deterioration in glycaemic control in Type 2 diabetes over 6 years and an increased prevalence of complications that present strategies in a multidisciplinary specialist diabetes clinic are unable to prevent on a sustainable basis.     

Apolipoprotein B as a long-term predictor of mortality in type 1 diabetes mellitus: a 15-year follow up

OBJECTIVES: To evaluate the association of apolipoprotein B (apo B) with mortality due to all causes, to cardiac disease and to ischaemic heart disease (IHD) in subjects with type 1 diabetes mellitus. SUBJECTS: 165 subjects with type 1 diabetes included in the Swiss Cohort of the WHO Multinational Study of Vascular Disease in Diabetes were followed for 14.7+/-0.45 years. METHODS: Causes of death were obtained from death certificates, hospital records and postmortem reports. Using a parametric proportional hazards model the association of apo B with mortality rates was assessed by time-to-event analysis, including the absolute cumulative mortality risk over time for various apo B levels at baseline. RESULTS: Apo B was positively associated with all-cause mortality [hazard ratio (HR) 2.65 per g L-1 increase of apo B, 95% CI: 1.11-6.36, P=0.029], cardiac mortality (HR 11.64, 1.03-131.11, P=0.047) and IHD mortality (HR 9.36, 1.26-69.66, P=0.029). An apo B>or=0.96 g L-1 translated into a duplication of overall mortality hazard (HR 1.93, 1.00-3.72, P=0.050), and a sevenfold increase of mortality because of cardiac disease or IHD (HR 7.44, 1.44-38.42, P=0.017 and HR 7.38, 0.78-69.82, P=0.081). A baseline apo B of 1.5 g L-1 predicted an absolute cumulative risk to die over the next 10 years of 12.1% (5.2-31.7) for male and of 10.4% (4.7-26.1) for female subjects whereas risks were 6.3% (1.8-21.4) and 5.4% (0.8-15.8) for an apo B of 0.8 g L-1. CONCLUSION: Apo B is consistently associated with an increased mortality in type 1 diabetes.     

Respiratory symptoms as predictors of all-cause mortality in an urban community: a 30-year follow-up

OBJECTIVE: We investigated the relationship between respiratory symptoms and mortality from all causes in a large Norwegian population. We also examined mortality during separate periods of follow-up. DESIGN: Population-based, prospective cohort study. SETTING AND SUBJECTS: A total of 19,998 men and women were randomly selected from the general population of Oslo. They received a postal respiratory questionnaire. The response rate was 88%. MAIN OUTCOME MEASURES: The relationship between 11 respiratory symptoms and 30 years of total mortality was investigated separately for men and women by multivariate analyses with adjustment for age, smoking habits and occupational exposure to air pollution. RESULTS: The relative mortality risk in comparison with asymptomatic subjects varied from 1.36 (95% confidence interval 1.25-1.48) for cough symptoms to 2.46 (2.13-2.85) for severe dyspnoea amongst men; the corresponding rates amongst women were 1.28 (1.16-1.40) and 1.52 (1.31-1.75), respectively. The relative risk of mortality in individuals with 1-3, 4-6 and 7 or more symptoms was 1.20, 1.60 and 2.53 (P for trend 0.000) in men and 1.14, 1.47 and 1.84 (P for trend 0.000) in women. Except for cough, the mortality rates associated with respiratory symptoms decreased significantly during follow-up. The positive association between respiratory symptoms and mortality was observed in people with and without cardiopulmonary diseases. CONCLUSIONS: Respiratory symptoms were significant predictors of mortality from all causes over 30 years, decreased during follow-up and were still increased after 30 years.     

Islet cell and insulin autoantibodies in organ-specific autoimmune patients. Their behaviour and predictive value for the development of Type 1 (insulin-dependent) diabetes mellitus. A 10-year follow-up study

Summary To evaluate the behaviour and predictive value of islet cell and insulin autoantibodies in patients with organspecific autoimmune diseases, we followed 21 non-diabetic subjects for a mean period of 84±27 months. Ten patients were persistently seropositive for complement-fixing islet cell antibodies and high titres of immunoglobulin G islet cell antibodies ( 18). The prevalence of persistent insulin autoantibodies in this group was 67%. Seven patients (70%) developed Type 1 (insulin-dependent) diabetes mellitus after a latency period of 2–60 months. The predictive value of complement-fixing islet cell antibodies was 65%, and in the presence of both complement-fixing islet cell and insulin autoantibodies the predictive value rose to 76%. Eleven patients were seronegative for complement-fixing islet cell antibodies and had low immunoglobulin G islet cell antibodies titres (< 18) that were either persistent or transient, or that fluctuated during follow-up. The prevalence of persistent insulin autoantibodies in this group was 45%; only one subject developed Type 1 diabetes. The predictive value of persistent islet cell antibodies (complement-fixing positive/negative) was 54%, and it rose to 70% when both islet cell and insulin autoantibodies were present. Individuals with only insulin autoantibodies or immunoglobulin G islet cell antibodies did not develop diabetes mellitus. A high frequency of HLA-DR3 and/or DR4 was found in patients who developed diabetes mellitus. Thus, the presence of both islet cell and insulin autoantibodies in patients with organ-specific autoimmune disease appears to confer the highest risk of progression toward Type 1 diabetes.Deceased 29 June 1986     

Functional tooth number and 15-year mortality in a cohort of community-residing older people

Background:   To study how dental status can become a predictor of overall mortality risk.
Methods:   Community residents ( n  = 5730) over 40 years old in the Miyako Islands, Okinawa Prefecture, Japan were followed up for 15 years, 1987–2002. Functional tooth numbers were examined by dentists and overall mortalities of subjects with functional tooth numbers of <10 and ≥10 were compared in the age groups 40–49, 50–59, 60–69, 70–79 and 80 years or more in both males and females.
Results:   Groups of 80 years or more showed a significantly higher rate of overall mortality in subjects with functional tooth numbers of less than 10 than 10 or more, and there was no significant difference in the other age groups.
Conclusion:   The present study suggests that systemic attention to dental status should be recommended in older males.     

Mortality in African-Americans with Type 1 diabetes: The New Jersey 725.

AIM: To determine rates and risk factors for all-cause mortality in African-Americans with Type 1 diabetes from a 3-year observational follow-up study of 725 African-Americans with Type 1 diabetes conducted between 1 January 1999 and 31 December 2001. METHODS: Date of death was ascertained either from telephone contact with the patient's family or from relatives or on line review of the US Social Security death index. RESULTS: Since the initial examination, 131 (18.1%) patients, 60 (20%) men and 71 (17%) women, have died. At the time of death, the mean age of the men was 40.7 +/- 10.6 years and that of the women 39.4 +/- 10.5 years. The median duration of diabetes at the baseline examination was 8.04 years, interquartile range (IQR) 3.76-15.22 years for men and median 10.54, IQR 4.49-18.36 years for women. Three-year mortality rates were 7.1% for women and 10.6% for men. Age-adjusted mortality rates were not significantly different between men and women. Relative to the general US and the New Jersey African-American population, standardized mortality ratios of African-Americans with Type 1 diabetes were 12 and six times greater for women and men, respectively. Older age, low socio-economic status, low body mass index, high diastolic blood pressure, macroangiopathy, proteinuria, severe diabetic retinopathy and heavy alcohol consumption were independent risk factors for all-cause mortality. In patients with microproteinuria at initial examination, the mortality rate for men was twice that of women. CONCLUSION: Microproteinuria and other potentially modifiable factors, including hypertension, macroangiopathy and heavy alcohol consumption, are independent risk factors for mortality in this ethnic group.     

Diabetic retinopathy, visual acuity, and medical risk indicators. A continuous 10-year follow-up study in Type 1 diabetic patients under routine care

The objective of this study was to describe incidence and progression of diabetic retinopathy in relation to medical risk indicators as well as visual acuity outcome after a continuous follow-up period of 10 years in a Type 1 diabetic population treated under routine care. The incidence and progression of retinopathy and their association to HbA_1c, blood pressure, urinary albumin, serum creatinine levels, and insulin dosage were studied prospectively in 452 Type 1 diabetic patients. The degree of retinopathy was classified as no retinopathy, background, or sight-threatening retinopathy, i.e. clinically significant macular edema, severe nonproliferative, or proliferative retinopathy. Impaired visual acuity was defined as a visual acuity <0.5 and blindness as a visual acuity ≤0.1 in the best eye. In patients still alive at follow-up (n=344), 61% (69/114) developed any retinopathy, 45% (51/114) background retinopathy, and 16% (18/114) sight-threatening retinopathy. Progression from background to sight-threatening retinopathy occurred in 56% (73/131). In 2% (6/335), visual acuity dropped to <0.5 and in less than 1% (3/340) to ≤0.1. Patients who developed any retinopathy and patients who progressed to sight-threatening retinopathy had higher mean HbA_1c levels over time compared to those who remained stable (P<.001 in both cases). Patients who developed any retinopathy had higher levels of mean diastolic blood pressure (P=.036), whereas no differences were seen in systolic blood pressure levels between the groups. Cox regression analysis, including all patients, showed mean HbA_1c to be an independent risk indicator for both development and progression of retinopathy, whereas mean diastolic blood pressure was only a risk indicator for the incidence of retinopathy. Metabolic control is an important risk indicator for both development and progression of retinopathy, whereas diastolic blood pressure is important for the development of retinopathy in Type 1 diabetes. The number of patients who became blind during 10 years of follow-up was low.     

Relative mortality of Type 1 (insulin-dependent) diabetes in Denmark: 1933–1981

Summary The relative mortality of Type 1 (insulin-dependent) diabetes in Denmark during the period 1933–1981 was studied using a modification of Cox's regression model on the basis of two patient populations, ascertained in different ways and independently of each other. Initial analysis showed that the two groups could be combined completely into one common analysis. Relative mortality was the same for both sexes. The additional variables studied were age at diagnosis, current age, calendar year at diagnosis and calendar time during follow-up. All these interrelated variables were accounted for in the analysis. The analysis showed that relative mortality (a) decreased with increasing age at diagnosis; (b) increased from 1933 to a maximum in about 1965, after which it decreased; (c) increased with increased duration of diabetes to a maximum at 15–25 years, after which it declined.     

Sleep complaints predict coronary artery disease mortality in males: a 12-year follow-up study of a middle-aged Swedish population

OBJECTIVES: Only a few prospective surveys have been performed to investigate the relationship between sleep complaints and coronary artery disease (CAD) mortality. This study was conducted to determine whether sleep complaints in a middle-aged population predicted total mortality and CAD mortality. DESIGN: A population-based prospective study. Setting. The County of Dalarna, Sweden. SUBJECTS AND METHODS: In 1983, a random sample of 1870 subjects aged 45-65 years responded to a postal questionnaire (response rate 70.2%) including questions about sleep complaints and various diseases. Mortality data for the period 1983-95 were collected, and Cox proportional hazard analyses were used to examine the mortality risks. RESULTS: At 12-year follow-up 165 males (18.2%) and 101 females (10.5%) had died. After adjustment for a wide range of important putative risk factors, difficulties initiating sleep (DIS) were related to CAD death in males [relative risk (RR), 3.1; 95% confidence interval (CI), 1.5-6.3; P < 0.01], but not in females. Short or long sleep duration did not influence risk of CAD mortality or total mortality for either gender. Depression in males increased the risk of death attributed to CAD (RR, 3.0; 95% CI, 1.1-8.4; P < 0.05) and total mortality (RR, 2.2; 95% CI, 1.1-4.5; P < 0.05). CONCLUSION: These results provide evidence that there is an association between difficulties falling asleep and CAD mortality in males.     

Cancer mortality reduction and metformin: a retrospective cohort study in type 2 diabetic patients

Aims: Few studies suggest that metformin decreases cancer mortality in type‐2 diabetic patients (T2DP). We explored the association between the type and duration of antidiabetic therapies and cancer and other‐than‐cancer mortality in a T2DP cohort, taking into account the competing risks between different causes of death and multiple potential confounding effects. The mortality rates were compared with the general population from the same area. Methods: In 1995, all T2DP (n = 3685) at our diabetes clinic in Turin (～12% of all T2DP in the city), without cancer at baseline, were identified. Vital status was assessed after a mean 4.5‐year follow‐up. Results: Metformin users had greater adiposity, while insulin users had more co‐morbidities. All‐cause‐ and cancer‐related deaths occurred in: 9.2 and 1.6% of metformin users, 13.1 and 3.0% of sulfonylureas users and 26.8 and 4.8% of insulin users, respectively. In a Cox regression model for competing risks, adjusted for propensity score, metformin users showed a lower cancer mortality risk [hazard ratio (HR) = 0.56; 95% confidence interval (CI) 0.34–0.94], while insulin was positively associated with other‐than‐cancer mortality (HR = 1.56; 95%CI 1.22–1.99). Each 5‐year metformin exposure was associated with a reduction in cancer death by 0.73, whereas every 5‐year insulin exposure was associated with 1.25‐fold increase in other‐than‐cancer death. Standardized mortality ratios for cancer and other‐than‐cancer mortality in metformin users were 43.6 (95%CI 25.8–69.0) and 99.1 (95%CI 79.3–122.5), respectively, in comparison with the general population. Conclusions: Metformin users showed a lower risk of cancer‐related mortality than not users or patients on diet only; this may represent another reason to choose metformin as a first‐line therapy in T2DP.     

Relationship between blood glucose level and mortality in Type 2 diabetes mellitus: a systematic review

Aim To review the relationship between blood glucose level and mortality in patients with Type 2 diabetes mellitus (DM) as reported in the literature. Methods Literature search using Medline Search: January 1966 – April 1998. Keywords: Diabetes, Non Insulin Dependent, Mortality. Inclusion criteria for papers were: Type 2 DM; follow-up for at least 3 years; glucose or glycated haemoglobin (HbA_1c) was used as parameter; published in the form of an article. Additionally all references in the selected articles that dealt with the relationship between blood glucose level and mortality in Type 2 DM were included in the search. Results Twenty-seven eligible articles were found. Twenty-three of them showed a positive association: measures of elevated blood glucose concentrations were associated with higher mortality; in 15 out of 23 studies the positive association was statistically significant, in two only for postprandial blood glucose. One study found a nonsignificant negative relationship in a very old population. Conclusion In the literature there is a positive, but rather weak, association between the measures of blood glucose control and the risk of dying of patients with Type 2 DM. In the six larger studies (more than 100 deceased patients) that used a continuous categorization of glycaemia, the Risk ratio per unit varies from 1.03 to 1.12. Diabet. Med. 16, 2–13 (1999)