Identification of Fatty Liver With Advanced Fibrosis in Patients With Type 2 Diabetes Using Simple Fibrosis Scores and Electronic Reminder Messages: A Randomized Controlled Trial

This will be a parallel group, randomized controlled trial. Patients fulfilling the inclusion and exclusion criteria above will be randomized 1:1 to two groups. Randomization will be carried out through the use of computer-generated list of random numbers in variable blocks of 4 to 10. Concealment of group allocation will be achieved through putting the group allocation cards in consecutively-numbered and sealed envelopes. The patients and physicians will know that the patients are in the intervention group if they see the reminder messages. When they do not see a reminder message, there will not be a specific indicator of whether the patient is in the control group or in the intervention group but having low fibrosis scores. Furthermore, the outcome assessors will be blinded to the group assignment.

For patients in the intervention group, we will calculate the fibrosis scores. For patients with increased fibrosis scores, we will type the following pop-up message in our electronic clinical management system:

"This patient has high Fibrosis-4 index (and/or AST-to-platelet ratio index) of xxx suggestive of significant liver fibrosis. Please consider referring the patient to the hepatology clinic or arranging further test such as FibroScan."

The reminder message will pop up when physicians see the patient at the clinic and use the electronic clinical management system. The message will remain active for one year. Although the message is entered manually at this stage, the arrangement mimics an automated computer system. If the study results are positive, the next step is to modify the system to automate the process.

Patients in the control group will undergo the same assessments as patients in the intervention group. Although physicians will have access to the raw liver biochemistry results and platelet count, the fibrosis score results will not be specifically shown, and there will be no electronic reminder messages regardless of the fibrosis scores. This is to mimic usual care when there is no dedicated care model for case identification.