Table 1 Characteristics of the included studies
First author, year | Country | Study design | Control origin | Contrast agent | No.of cases | No.of control | OR (95%CI) | Baseline age (year) | Mean age (years) (case/control) | Percentage of males (case/control) | Study period | Adjusting variables | New diabetes | NOS scores |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Ruiter 2012 | Netherlands | Cohort | Population | Sulfonylureas | 52,689 | 32,591 | 0.73(0.66–0.80) |  ≥ 18 | 61.8/65.6 | 46.4/48.2 | 1998–2008 | Age, gender, hypoglycemic agent duration, other drugs use, previous hospitalization | No | 9 |
Bodmer 2011 | Britain | Case–control | Population | No metformin | 2763 | 16,578 | 0.83(0.57–1.21) |  > 0 | NA | 46.2/46.2 | 1995–2009 | BMI, smoking, drinking and the course of diabetes, congestive heart failure, ischemic heart disease, ischemic or hemorrhagic heart disease, transient ischemic attack, arterial hypertension and dyslipidemia, aspirin, other NSAIDs, statins or estrogen use | No | 9 |
Sung 2020(a)A | Hong Kong, China | Cohort | Hospital | No metformin and aspirin | 11,365 | 277,932 | 1.45(0.83–2.53) |  ≥ 18 | NA | 46.5/53.3 | 2000–2004 | Age, gender, comorbidities, and baseline medications (including histamine 2 receptor antagonists (H2 antagonists), statins, nonsteroidal anti-inflammatory drugs (NSAIDs), and anticoagulants | No | 9 |
Sung 2020(b)A |  |  |  |  | 6630 | 277,932 | 0.58(0.20–1.65) |  | NA | 52.8/53.3 |  |  | No | 9 |
Zhao 2022 | China | Cohort | Population | Sulfonylureas | 16,982 | 19,285 | 1.01(0.51–1.98) |  ≥ 18 | 58.1/61 | 53.2/51.0 | 2009–2020 | Age, gender, education level, duration of smoking, drinking, T2DM, blood glucose level, blood lipid level and blood pressure, Charlson's complication index, BMI, and utilization rate of medical care; Sulfonylurea and metformin removal(α- Glucosidase inhibitors, thiazolidinediones, dipeptidyl peptidase 4 inhibitors, Grinnerd, and insulin), commonly used drugs for cardiovascular diseases (diuretics, β- Blockers, calcium channel blockers, angiotensin-converting enzyme inhibitors (ACEI), angiotensin receptor blockers (ARB) and aspirin), commonly used antibiotics (penicillins, cephalosporins, macrolides, quinolones, and other antibiotics), statins and proton pump inhibitors (PPIs) | Yes | 9 |
Valente 2017 | Europe | Case–control | Hospital | No metformin | 164 | 529 | 1.35(0.68–2.66) | NA | 59.6/59.5 | 51.0/51.0 | 2013–2015 | Smoking, drinking, height and weight, body mass index (BMI), chronic pancreatitis, acute pancreatitis, peptic ulcer disease, biliary calculus and previous surgical history, gender, age, and inclusion in the center | No | 7 |
Oh 2020 | Korea | Cohort | Population | No metformin | 19,546 | 19,546 | 0.88 (0.70–1.11) |  ≥ 18 | 60.5/60.3 | 53.0/52.6 | 2011–2015 | Gender, socio-economic information (income level and residence in 2010), complications (hypertension, coronary artery disease, cerebrovascular disease, psychological and behavioral disorders, musculoskeletal diseases, chronic kidney disease, dyslipidemia, anemia, chronic obstructive pulmonary disease, arrhythmia, and liver cirrhosis); received surgery in 2010; and total hospital stay in 2010 | No | 9 |
Murff 2018 | America | Cohort | Population | Sulfonylureas | 42,217 | 42,217 | 0.85(0.57–1.27) |  ≥ 18 | 66.2/65.4 | 97.2/97.2 | 2001–2008 | Age, gender, race (white, black, other), cohort entry date, body mass index, blood pressure, glomerular filtration rate, hemoglobin A1c (HbA1c), low-density lipoprotein level, smoking status, drug selection (statins, aspirin, antihypertensive drugs, anticoagulants, antiarrhythmic drugs, diuretics, antipsychotics, glucocorticoids), Times of medication and outpatient visits for comorbid diseases (cardiovascular disease, serious mental disease, heart valve disease, arrhythmia, Parkinson's disease, chronic obstructive pulmonary disease, liver disease) | No | 9 |
Tsilidis 2014 | Britain | Cohort | Population | Sulfonylureas | 51,484 | 18,264 | 0.70(0.45–1.07) |  ≥ 35 | 61.1/65.3 | 56.1/57.9 | 1987–2010 | Age, gender, body mass index, smoking. Alcohol consumption, aspirin or nonsteroidal anti-inflammatory drugs (NSAIDs), statins, and exogenous hormones | No | 9 |
Wang 2013 | Taiwan, China | Case–control | Population | No metformin | 2158 | 8609 | 1.14(0.68–1.91) | N.A | NA | NA | 1998–2009 | Age, gender, and occupation | Yes | 5 |
Liao 2012 | Taiwan, China | Cohort | Population | No metformin | 42,754 | 7049 | 0.85(0.39–1.89) |  ≥ 20 | NA | NA | 1998–2007 | Age, gender, chronic pancreatitis, hepatitis C infection, gallstones | Yes | 8 |
Oliveria 2008 | America | Cohort | Population | No metformin | N.A | NA | 1.26(0.80–1.99) |  ≥ 18 | NA | NA | 2000–2004 | Age, gender, partial gastrectomy, chronic pancreatitis, deep venous thrombosis, dermatomyositis/polymyositis, alcoholism, hepatitis B/C, history of polyps | No | 9 |
Tseng 2018 | Taiwan, China | Cohort | Population | No metformin | 12,616 | 12,616 | 0.49(0.25–0.96) | NA | NA | NA | 1999–2005 | Age, gender, occupation, residential area, hypertension, dyslipidemia, obesity, kidney disease, eye disease, stroke, ischemic heart disease; peripheral artery disease; chronic obstructive pulmonary disease, tobacco abuse; history of Helicobacter pylori infection; drugs (insulin, sulfonylurea, metronidazole, acarbose, rosiglitazone, pioglitazone, and angiotensin-converting enzyme inhibitors/angiotensin receptor blockers, calcium channel blockers, statins, fibrin, and aspirin) | Yes | 6 |
Currie 2009(a)B | Britain | Cohort | Population | Sulfonylureas | 31,421 | 7439 | 0.20(0.11–0.36) |  ≥ 40 | 58.6/70.0 | 51.1/54.9 | 2000-mid2000 | Age, gender, systolic blood pressure, total cholesterol, weight, weight change, BMI, smoking status, baseline general incidence rate, previous major vascular disease (LVD), retinopathy, kidney damage, glycosylated hemoglobin, and previous solid tumor records | No | 9 |
Currie 2009(b)B |  |  |  | Insulin | 31,421 | 10,067 | 0.22(0.12–0.38) |  | 58.6/63.7 | 51.1/55.4 |  |  | No | 9 |
De 2017 | Netherlands | Cohort | Population | No metformin | 37,215 | 19,899 | 1.11(0.72–1.71) |  ≥ 30 | 63.5/67.0 | 48.8/47 | 1998–2011 | Age, duration of diabetes (time since NIAID dispensing was first recorded), other drugs (statins, aspirin, nonaspirin nonsteroidal anti-inflammatory drugs (NSAIDs), proton pump inhibitors, bisphosphonates, tamoxifen, oral contraceptives, and insulin) | No | 9 |
Farmer 2019 | Britain | Cohort | Population | No use of any medicine | 6105 | 49,524 | 3.11(1.24, 7.76) |  ≥ 30 | 57.6/62.2 | 58.9/56.1 | 1990–2014 | Age, gender, smoking status and alcohol status, year of onset of diabetes, HbA1c, BMI, previous year's use of other drugs (NSAIDs, statins, antihypertensive drugs), chronic kidney disease (CKD), and cardiovascular disease (CVD) history | No | 9 |
Lee 2018 | Korea | Cohort | Population | No metformin | 688,656 | 277,797 | 0.86(0.77–0.96) |  ≥ 30 | NA | NA | 2009–2012 | Age, gender, chronic pancreatitis, acute pancreatitis, hepatitis B, hepatitis C, biliary disease, alcoholism, NAFLD, lowest quartile income, place of residence, and number of ADMs with different exposure | Yes | 9 |
Lee 2011 | Taiwan, China | Cohort | Population | No metformin | 11,212 | 4194 | 0.15(0.03–0.79) |  ≥ 20 | NA | NA | 2000–2007 | Age, gender, another oral antidiabetic agent, CCI score, duration of metformin exposure | Yes | 9 |
Lu 2015 | Britain | Case–control | Population | No metformin | 175 | 856 | 1.50(1.07–2.09) |  ≥ 20 | NA | NA | 1996–2010 | Age, gender, BMI, smoking, drinking; Townsend deprivation index, and diabetes | Yes | 8 |
Vicentini 2018 | Italy | Cohort | Population | No use of any medicine(Dietary treatment) | 7460 | 4060 | 1.51(0.59–3.89) |  ≥ 20 | NA | NA | 2009–2012 | Gender, age, nationality, and time after diagnosis of diabetes | No | 8 |
Walker 2015 | America | Case–control | Hospital | No metformin | 81 | 89 | 1.01(0.61–1.68) |  ≥ 21 | NA | 53/48.3 | 2006–2011 | Age, gender, race, BMI, history of pancreatitis, alcohol, smoking, P.C. family history, other diabetes drugs, diabetes duration | No | 9 |
You 2020 | Korea | Cohort | Population | No metformin | 131,877 | 131,877 | 1.34(1.21–1.48) |  > 0 | 60.7/60.9 | 49.9/50.9 | 2005–2014 | Age, gender, economic status, and residential area | Yes | 9 |
Hsieh 2012(a)C | Taiwan, China | Cohort | Population | Sulfonylureas | 3963 | 6072 | 0.63(0.28–1.42) |  ≥ 20 | NA | NA | 2000–2008 | Age, gender | No | 8 |
Hsieh 2012(b)C |  |  |  | Insulin | 3963 | 751 | 1.44(0.18–11.5) |  ≥ 20 | NA | NA |  |  | No | 8 |
Van 2011(a)D | Britain | Cohort | Population | Sulfonylureas | 109,708 | 68,029 | 0.60(0.52–0.70) |  > 40 | 63.0/65.0 | 56.3/56.1 | 1997–2006 | Age, gender, past years, social and economic status of small regions, smoking status, alcohol consumption, BMI, previous medical history (history of coronary heart disease, coronary artery reconstruction, hyperlipidemia, hypertension, peripheral vascular disease, renal damage, stable angina pectoris), previous medication (angiotensin II receptor blocker, antiplatelet β Receptor blockers, calcium channel blockers, diuretics, nitrates, NSAIDs, aspirin or statins) | No | 9 |
Van 2011(b)D |  |  |  | Thiazolidinediones | 109,708 | 31,372 | 1.16(0.91–1.48) |  > 40 | 63.0/63.0 | 56.3/57.3 |  |  | No | 9 |
Van 2011(c)D |  |  |  | Insulin | 109,708 | 23,005 | 0.46(0.38–0.56) |  > 40 | 63.0/65.0 | 56.3/55.8 |  |  | No | 9 |
Li 2009 | America | Case–control | Hospital | No metformin | 255 | 106 | 0.38(0.22–0.69) | NA | NA | NA | 2004–2008 | Age, race, gender, smoking, alcohol, BMI, family history of cancer, diabetes duration, use of insulin |  No |  8 |