Objective To investigate the compliance of follow-up and the associated factors of diabetic retinopathy screening among diabetes patients in Muping District,Yantai City.Methods 438 diabetes patients who received free diabetic retinopathy (DR) screening in Traditional Chinese Medicine Hospital of Muping District of Yantai City from April 2012 to June 2013 were enrolled in the cross-sectional study.The re-examination rate at five years later,as well as the possible associated factors,including age,gender,height,weight,education level,occupation,past medical history,smoking and drinking habits,diabetes process,diabetes medication history,severity of DR,glycosylated hemoglobin (HbA1 c),renal functionand blood lipid were analyzed and evaluated.Results Among 438 patients,159 returned for check up five years later,with a follow-up rate of 36.3％.Further analysis revealed that follow-up rate was related to occupation types (P =0.003) and HbAlc level (P =0.006).Other factors did not show statistical significance (P ＞ 0.05).Conclusions The compliance of DR screening visits is related to occupational type and blood glucose control.It is suggested that in the screening of DR in China,we should pay attention to the propaganda and education of diabetic patients to control blood glucose,and focus on some occupational diabetes people,such as housework workers and farmers.

OBJECTIVE To establish the fingerprints of Tianma jiannao granules （TJG） and the method for content determination to evaluate the quality of TJG comprehensively combined with chemometric analysis. METHODS High-performance liquid chromatography （HPLC） was used to establish the fingerprints of 13 batches （S1-S3） of TJG and determine the contents of inosine， gastrodin， parishin B and parishin E. Cluster analysis， principal component analysis， and orthogonal partial least squares- discriminant analysis were performed using SPSS 20.0 and SIMCA 18 software； using variable importance projection （VIP） value greater than 1 as a criterion， marker components that affected quality were screened. RESULTS A total of 28 common peaks were identified in the 13 batches of TJG with similarities greater than 0.9， and 7 common peaks were identified， which were gastrodin， p-hydroxybenzyl alcohol， parishin B， parishin E， rhynchophylline， inosine and salidroside. The 13 batches of TJG were clustered into 3 categories， S1-S2， S8-S10 and S12 were clustered into one category； S3 and S7 were clustered into one category； S4-S6， S11 and S13 were clustered into one category. VIP of inosine was greater than 1. The contents of inosine， gastrodin， parishin B and parishin E were 62.637-176.677， 17.821-37.642， 5.748-16.077 and 5.660-13.510 μg/g. CONCLUSIONS The established HPLC fingerprints and content determination method are stable， reliable and highly reproducible， which can be used to evaluate the quality of TJG in combination with chemometric analysis. Inosine may be a marker component that affects the quality of TJG. There are differences in the quality of 13 batches of TJG.