Table 1.

First-level Features constructed from source “demographic information”, “communication reports”, “outpatients diagnosis reports”, “inpatients diagnosis reports”, “inpatients discharge summaries”, “prescription reports” and “laboratory test reports”.

Source	category	Feature
Demographic information		De-identification ID, age and gender of a subject. (feature ranging from f1 to f3 as shown in Table A1 of Appendix A)
Communication report	Self-reporting note	Number of times a subject reporting body weight loss, persistent hunger, polyuria, polydipsia, prescribed diabetes medicine or returning visits for diabetes in communication report. (feature ranging from f4 to f9 as shown in Table A1 of Appendix A)
	Diagnosis code	Number of times codes of type 2 diabetes, diabetic retinopathy, diabetic neuropathy, diabetic eye disease, diabetic kidney disease, diabetic cerebral vascular disease or diabetic peripheral circulation appeared in communication report. (feature ranging from f10 to f17 as shown in Table A1 of Appendix A)
	Diagnosis note	Number of times communication report containing notes of type 2 diabetes, diabetic retinopathy, diabetic neuropathy, diabetic eye disease, diabetic kidney disease, diabetic cerebral vascular disease or diabetic peripheral circulation disease. (feature ranging from f18 to f25 as shown in Table A1 of Appendix A)
Outpatient diagnosis record	Diagnosis code	Number of times codes of type 2 diabetes, diabetic retinopathy, diabetic neuropathy, diabetic eye disease, diabetic kidney disease, diabetic cerebral vascular disease or diabetic peripheral circulation were appeared in outpatient diagnosis record. (feature ranging from f26 to f33 as shown in Table A1 of Appendix A)
	Diagnosis note	Number of times outpatient diagnosis record containing notes of type 2 diabetes, diabetic retinopathy, diabetic neuropathy, diabetic eye disease, diabetic kidney disease, diabetic cerebral vascular disease or diabetic peripheral circulation disease. (feature ranging from f34 to f41 as shown in Table A1 of Appendix A)
Inpatient discharge summary	Diagnosis notes	Number of times inpatient discharge summary containing notes of type 2 diabetes, diabetic retinopathy, diabetic neuropathy, diabetic eye disease, diabetic kidney disease, diabetic cerebral vascular disease or diabetic peripheral circulation disease. (feature ranging from f42 to f49 as shown in Table A1 of Appendix A)
Inpatient diagnosis record	Diagnosis codes	Number of times codes of type 2 diabetes, diabetic retinopathy, diabetic neuropathy, diabetic eye disease, diabetic kidney disease, diabetic cerebral vascular disease or diabetic peripheral circulation were appeared in inpatient diagnosis record. (feature ranging from f50 to f57 as shown in Table A1 of Appendix A)
	Diagnosis notes	Number of times inpatient diagnosis record containing notes of type 2 diabetes, diabetic retinopathy, diabetic neuropathy, diabetic eye disease, diabetic kidney disease, diabetic cerebral vascular disease or diabetic peripheral circulation disease. (feature ranging from f58 to f65 as shown in Table A1 of Appendix A)
Prescription record	Medication	Number of prescriptions appearing in prescription report for oral hypoglycemic, insulin, Chinese traditional hypoglycemic, a mixture of western and Chinese traditional oral hypoglycemic, Epalrestat, Alpha-glucosidase inhibitor, Dipeptidyl peptidase IV(DPP-IV) inhibitors, Meglitinides, Sulfonylureas, Thiazolidinedione, Biguanides, Incretin Mimetics, GLP-1 (glucagon-like peptide 1) mimetics, compounds of sulfonylurea and thiazolidinedione, compounds of Biguanides and Dipeptidyl peptidase IV(DPP-IV) inhibitors, compounds of Biguanides and Sulfonylureas, or compounds of Biguanides and Thiazolidinedione. (feature ranging from f66 to f82 as shown in Table A1 of Appendix A)
Laboratory test report	Venous plasma glucose test	Number of times for 2-h venous plasma glucose test, 2-h peripheral plasma glucose test ≥11.1 mmol/l (200 mg/dl), fasting venous plasma glucose test, fasting venous plasma glucose test ranging from 6.1 to 7.0 mmol/l (110 and 126 mg/dl), random venous plasma glucose test, or random venous plasma glucose test ≥11.1 mmol/l (200 mg/dl). (feature f83, f84, f87, f88, f91, and f92 as shown in Table A1 of Appendix A) The maximum value of 2-h venous plasma glucose test, fasting venous plasma glucose test, or random venous plasma glucose test. (feature f85, f89, and f93 as shown in Table A1 of Appendix A) The minimum value of 2-h venous plasma glucose test, fasting venous plasma glucose test, or random venous plasma glucose test. (feature f86, f90, and f94 as shown in Table A1 of Appendix A)
	Peripheral plasma glucose test	Number of times for 2-h peripheral plasma glucose test, 2-h peripheral plasma glucose test ≥11.1 mmol/l (200 mg/dl), fasting peripheral plasma glucose test, fasting peripheral plasma glucose test ranging from 6.1 to 7.0 mmol/l (110 and 126 mg/dl), random peripheral plasma glucose test, or random peripheral plasma glucose test ≥11.1 mmol/l (200 mg/dl). (feature f95, f96, f99, f100, f103, and f104 as shown in Table A1 of Appendix A) The maximum value of 2-h peripheral plasma glucose test, fasting peripheral plasma glucose test, or random peripheral plasma glucose test. (feature f97, f101, and f105 as shown in Table A1 of Appendix A) The minimum value of 2-h peripheral plasma glucose test, fasting peripheral plasma glucose test, or random peripheral plasma glucose test. (feature f98, f102, and f106 as shown in Table A1 of Appendix A)
	HbA1Ctest	Number of times for HbA1c test, HbA1C test ≥ 6.5%. (feature f107, and f108 as shown in Table A1 of Appendix A) The maximum value of HbA1C test (feature f109 as shown in Table A1 of Appendix A) The minimum value of HbA1C test (feature f110 as shown in Table A1 of Appendix A)