The most concrete published evidence about clinical decision support in Vietnamese hospitals begins with a written examination, not routine clinical use.
In 2013, a multicenter study analyzed results from 203 physicians at 11 hospitals. The intervention used a 20-minute software training video in Vietnamese followed by a 60-minute scripted discussion of simulated cases in both English and Vietnamese. Participants then used PEMSoft, an English-language pediatric emergency program, on the posttest and scored higher than on the pretest.
That result is useful because its boundary is clear. The study measured written-exam performance in a pretest-posttest design. It did not measure decisions in live cases, patient outcomes or sustained use. A 2022 scoping review of digital health in Vietnamese hospitals added the deployment detail: PEMSoft was used for testing and was not implemented in the hospitals studied. Better answers on an exam and a system used during routine care are different achievements. The first shows what clinicians could do with the tool under study conditions. The second would require evidence from the workflow where clinical decisions are made.
A newer hospital snapshot still does not give a national count
A 2025 assessment of five major Vietnamese public hospitals looked beyond one training intervention. Hospital, laboratory and radiology information systems were widely used across the five sites. More advanced use of clinical data through clinical data repositories and CDSS remained limited.
The authors explicitly warn that five hospitals cannot represent the country. Their findings do not tell us how many Vietnamese hospitals use clinical decision support. They show a narrower pattern within the hospitals assessed: systems for collecting and accessing digital information had progressed further than systems using that information for clinical decisions.
These papers support a narrow conclusion. Vietnam has tested clinical decision support, and specific hospitals show limited decision-support use amid uneven digital capability. They do not support a claim that CDSS is nationally absent.
Circular 13 changes the infrastructure question
Circular 13/2025/TT-BYT took effect on 21 July 2025. Licensed hospitals were required to implement electronic medical records by 30 September 2025. Other facilities providing inpatient, day and outpatient treatment must complete implementation by 31 December 2026.
The Circular also preserves a transition path for paper. For a patient whose treatment began before 21 July 2025 and continued beyond that date, a paper chart already in use may remain only until discharge or the end of that outpatient episode, unless the facility can convert it. For older paper records, the head of the facility decides whether to convert them according to local conditions and the rules for electronic transactions.
The mandate therefore changes the implementation baseline. It requires IT infrastructure, clinical applications, backup storage, security and retrievable electronic records. It does not make every historical chart complete, normalize data across systems or prove that a hospital is ready to operate a CDSS.
A deployable CDSS has a harder contract
The Office of the National Coordinator for Health IT defines clinical decision support around timely, person-specific information delivered at an appropriate point in care. The output may be a drug-interaction alert, an order set, a patient summary, diagnostic support or relevant clinical guidance. ONC also notes that CDS may sit inside an EHR, operate independently or connect as a plug-in.
An electronic record is therefore one input, not the finished product. Decision support still has to work with the data the hospital can provide, expose missing context, use knowledge appropriate to its intended scope and reach the clinician while a decision can still change. Translating the interface of a foreign reference product does not localize its clinical content. Making a clinician leave the record and re-enter the case returns the integration work to the clinician.
Meddies is being built to bring decision support into the hospital workflow. The intended system carries patient context into the answer, attaches sources to its claims and leaves the final decision with the doctor. These are design goals, not evidence that Meddies improves clinical decisions or patient outcomes.
Only deployment can test whether the design works. An evaluation would need to examine accuracy, actual use, ignored alerts and effects on clinical work. Until that evidence exists, we can say only that Circular 13 requires facilities to build an electronic-record base that integrated decision support could use. The clinical value of any particular system remains to be demonstrated.