Aim: Effect of health education and screening location on compliance with diabetic retinopathy screening in a rural population of Bihar. Methods: The present study was conducted in the Department of Ophthalmology, NMCH, Patna, Bihar, India, for 1 year. The population is predominantly rural. The blocks were grouped as facilities A and B: screening for DR in CHCs. Health education was not imparted in A but was imparted in B. Facilities C and D: screening in PHCs. Health education was not imparted in C block but was imparted in D. The health education intervention in the two settings was delivered by Village Level Health Workers (VHWs). Blindness and visual impairment were classified as per the WHO International Classification of Diseases. Results: The number of people screened in both blocks with PHCs screening was similar (530and 250) and higher than in the blocks with CHC screening (170 and 320). A total of 1270 people with diabetes out of 7310 registered (17.37%) were screened for DR in the four blocks over the 3‑month period. The highest uptake was in the block with PHC level screening with health education and provision of transport to PHCs from villages (29.03%) while the lowest was in the block with CHC level screening without health education (9.79%). The uptake was significantly higher in the facilities with health education than in those without (18.69% and 16.43%, respectively, P < 0.01), and was significantly higher in blocks with PHCs level screening with provision of transport to PHCs from villages than CHCs level screening (23.64% and 12.22%, respectively, P = <0.001). A third of those screened had some degree of visual impairment: 8.03% (102) were blind, 6.30% (80) had severe visual impairment, 20.08% (255) had moderate visual impairment, and 65.59% (833) had mild or no visual impairment. There was not much difference in visual status between the people who did or did not receive health education. Fundus images were gradable in 81.50% (1035/1270 of those screened. In the gradable images, 14.57% (185/1270) had any DR and 9.94% (50/1270) had STDR. Conclusion: Conducting DR screening closer to the place of living at PHCs with the provision of transport and health education was more effective, resulting in an increase in the uptake of DR screening by people with known diabetes in rural Bihar.