Introduction: There is no debate that people with diabetes should be screened for the development of retinopathy which can threaten their sight. However, there is no routine screening for retinopathy in Hong Kong at present. Many overseas countries find that they miss a large proportion of their target population and, with reliance on co-payments for screening, as is the case with the limited opportunistic screening at present, the cost-effectiveness of any routine service in Hong Kong could be reduced as is predicted by Hart's inverse care law.

Aim: This study will determine the potential cost-effectiveness of screening for retinopathy in Hong Kong under a free system and one in which a co-payment is charged.

Methods: Primary care patients attending General Outpatient Clinics on Hong Kong Island for their routine diabetic care will randomly be offered screening either at no charge or with the normal co-payment of $65. Those who are willing and unwilling to be screened will be compared for their clinical, lifestyle and socioeconomic characteristics and those unwilling will be asked their reasons. The uptake of screening at no fee and with a payment will be compared as will the prevalence of retinopathy in the two fee groups. Subsequent screening at one year will be offered at the same fee and uptake again compared.

The principal analyses will (a) identify the characteristics of those willing to be screened and reasons for not being screened (b) the uptake of screening when a co-payment is charged compared to when it is free (c) whether there is a difference in the prevalence of retinopathy between the group willing to pay and those who accept free screening and (d) the uptake of re-screening in year 2.

The resulting cost-effectiveness model will use these data, the cost data collected during the study and overseas data on benefits of treatment to model the cost-effectiveness of screening for retinopathy in Hong Kong if it were to be offered free or with a co-payment. This information will be important to determine the most cost-effective means of implementing this preventative strategy to preserve sight and quality of life.

Inclusion Criteria:

• All diabetic patients who attend the clinic during the study period.

Exclusion Criteria:

• Only if not competent to give informed consent