The Influence Of Artificial Intelligence Auto Verification As An Intervention On Bpjs Claim Output: A Case Study At Murni Teguh Memorial Hospital Medan

Abstract

Background: Hospitals are healthcare facilities that provide specialized medical treatment, care, and services to individuals in need of diagnosis, treatment, and recovery from illness, injury, or other health conditions. Hospitals are usually equipped with a variety of medical professionals, including doctors, nurses, and support staff, and offer a wide range of services (Peraturan Menteri Kesehatan Republik Indonesia, 2020).  Murni Teguh Memorial Hospital Medan is a leading healthcare facility located in Medan, North Sumatra, Indonesia. Established under the ownership of PT Murni Sadar Tbk, the hospital aims to provide high-quality medical services to the community. For the continuity of health services provided and legal interests, hospitals are required to conduct medical record activities to ensure patient safety. According to the Regulation of the Minister of Health Number 24 of 2022, Medical Records are files containing records and documents on patient data containing patient identity, examination, treatment, actions, and other services that have been provided to patients (Regulation of the Minister of Health of the Republic of Indonesia, 2022).   Murni Teguh Memorial Hospital Medan develops electronic medical records to perform official patient documentation. An Electronic Medical Record (EMR) is a digitized version of a patient¹s paper medical record.

Objective: This study aims to analyze the influence of several factors on the output of BPJS claims at Murni Teguh Memorial Hospital Medan, especially in overcoming the problem of pending claims that can hamper hospital cash flow.

Methods: This study used an experimental approach with pre-test and post-test methods to measure the effectiveness of implementing the AI Auto Verification model in overcoming obstacles that cause pending claims on BPJS claims at Murni Teguh Memorial Hospital Medan. This approach was conducted by comparing conditions before and after the use of AI through data collection using a questionnaire before the use of the AI Auto Verification model (control) and a questionnaire after the use of the AI Auto Verification model (experimental). Thus, this study can provide a more in-depth analysis of the extent to which the application of the Auto Verification model is able to overcome inhibiting factors, such as the completeness of administration/files, errors in coding, and factors related to machines and manpower.

Results: Based on the results of previous research conducted by Cathryn Gabriella (2020) on the factors affecting the return of BPJS claims for inpatients at Fatmawati Government General Hospital in March - May 2020 there were 218 inpatient claim files returned, consisting of three types of returns, namely due to Membership Administration Factors 1.8%, Service Administration Verification 17%, and Service Verification 81.2%.² The return factor caused by service verification is the most common return factor at Fatmawati Government General Hospital in 2020. Based on other research at Fatmawati Government General Hospital in 2016 on the return of BPJS claims for inpatients in January - April there were 1719 claims returned, consisting of four categories, namely Administrative Improvements 19%, Borrowed Status 7%, Confirmation of Coding 36%, and Completeness of Resume 38%, of the four factors, the most returns came from the Completeness of Resume category 38%.

Conclusion: This analysis will change the BPJS claim variables after the implementation of the Auto Verification model, to test whether there is an increased impact of file completeness, coding accuracy, and optimization of human and machine resources in the BPJS claim process. The results of the analysis show a significant difference between before and after the implementation of AI, so it can be concluded that the AI Auto Verification model is an effective solution to overcome the various obstacles that cause BPJS claims.