Tagged: #CDSS
- This topic has 18 replies, 12 voices, and was last updated 1 year, 6 months ago by Tanatorn Tilkanont.
-
AuthorPosts
-
-
2022-09-23 at 1:05 pm #38256SaranathKeymaster
Can you give an example of Decision Support System used in your organization?
Is it working well?
Are there any factors that might influence the decision support system implementation in your organization? -
2022-09-26 at 6:34 pm #38364Zarni Lynn KyawParticipant
Although we don’t have a proper Decision Support System used in my organization. We are currently developing a EMR systems with CDSS built into the systems.
We plan to provide tablets to health workers and the health workers can use the tablets for data entry of patients data, diagnosis and treatments. During the entry of diagnosis data entry, we will display differential diagnosis according to the guidelines and during treatment data entry, we will display possible drug-to-drug interactions.
We are piloting the system in two clinics and our preliminary feedback indicate that we need to regularly update the program with the health workers inputs. At the same time, we need to provide more technical support to the clinics as the health workers computer literacy is limited. Clinic-in-charge from both participating clinics suggest that using a tablet instead of a computer is easier but the intuitiveness of program is still needs improvements as well as language barrier still exit, we can only provide English and Burmese program but the clinics are located in Karen area and if the interface is in a local language, it was reported that all the health workers will find it easier.
There are many factors that influence the decision support system implementation in my organization, but after reading the Factors influencing implementation success of guideline-based clinical decision support systems: A systematic review and gaps analysis, I’ve been reanalyzing our implementation using HOT-fit framework and human, technology and organizational factors play a very important role in the implementation.
My analysis is as follow,
Human – we managed the system use and user satisfaction well as we provide continuous technical support and training to the health workers
Technology – we’ve seen system, information and service quality improved (although service quality improvement needs more time to be quantified properly), at the system level, use of data is improved by health workers themselves, completeness of information is better compared to a previous system.
Organization – we have yet to properly analyze the structure and environment, but it gave us a lot to think about and in the future when we evaluate the system, we will use this dimension as well.
Net benefits – as the project is very small currently and just started, we have yet to measure the net benefits but it gave me food for thought.
In conclusion, we found many areas to improve in the future with our fairly new CDSS system and we are planning to improve it in the future.
-
2022-10-01 at 11:52 am #38482Boonyarat KanjanapongpornParticipant
Thank you for sharing, your information is always interesting and inspiring.The system might be very supportive because there aren’t negative effects or resistance from human factors to the system implementation.
-
-
2022-09-28 at 10:00 am #38384ABDILLAH FARKHANParticipant
The example of DSS in my organization (Ministry of Health) is being applied in a mobile phone app for Covid-19 contact tracing purposes. It is useful to support the epidemiological investigation and monitoring of close contact in the field. The app works with Personal Area Network (PAN) where the Bluetooth network can exchange personal information of two or more people when they were at less than 2 meters distance in a moment. If there is a person identified as a confirmed case during the next 14 days, the DSS will identify all people who have been close to him/her and will notify them through the app. DSS will provide an alert to the close contact people and recommend them to take further precautions.
My evaluation for this DSS is as follows:
Human
-System Use: the existence of this system may be surprising for those who are not knowledgeable about Covid-19 epidemiology and control so it needs massive education
– User Satisfaction: people feel well-satisfied because they can be aware more when they are present in public and populous spaceOrganization
– Structure: the app can improve the performance of MoH in actively detecting close contact people, and enhance decision-making for case management
– Environment: the app can promote the efficiency of resources for contact tracing, including human resources, logistics, financial, and timeTechnology
– System Quality: the app can work easy, fast, secure, and feasible
– Information Quality: the app provides an alert of epidemiological information that is complete, relevant, understandable, and on time
– Service Quality: the app facilitates remote consultation that ensures further precautions for those in close contactNet Benefits: Improved speed of early warning system and behavioral health of people becomes the most significant benefit of this DSS
Factors that might influence the decision support system implementation mostly come from external, such as pandemic situations and policy, such as:
1. The system presents at the right moment when we are still facing Pandemic occurrence
2. The urgency to transform health-related activities into the digital method
3. The national strategy encourages the citizens to have the digital literate skill-
2022-10-05 at 6:40 pm #38544Tanyawat SaisongcrohParticipant
Thank you for sharing. This Covid-19 tracing DSS is very useful for users and also community health in terms of disease control during the pandemic. I agree that implementation of the right particular DSS in the right time is one of the key success factors.
-
-
2022-09-29 at 9:57 am #38408PREUT ASSAWAWORRARITParticipant
In my organization, one of the complex clinical decision support system is alert system for potential drug interaction prescribed to the patient. The system needs lots of information about adverse effects of each drugs, integrates and analyses them. The output must be a meaningful alert that warns physicians about drug interaction among their prescribed medication.
According to HOT-fit model, the challenges and obstacles of implementation this clinical decision support system include:
1. Information quality. The pop-up from the system lacks of relevant information with the patient. For example, prescribing medication that has minor drug interaction which previously prescribed without any adverse event would be too much information.
2. There are too many details on pop-up. Since some patients had several medication which has drug interaction, too many pop-up will make physicians ignore and not focus on the information alert from the system.
3. Implementation of this system may not alter the clinical management because many drugs that have interaction must be prescribed to the patients.In conclusion, some drug interaction is life-threatening. We must put this interaction in the first priority to alert physician and mark them with high sensitivity of alert (the system must alert for every prescription of these drugs). However, other agents with minor drug interaction which has been previously prescribed without any adverse event should be put into further research to find other factors that make patients be vulnerable to such drug interaction.
-
2022-09-29 at 9:56 pm #38448Siriphak PongthaiParticipant
I agree with you that too many pop-up could cause over ignored by healthcare providers and sometimes we might miss an important one. In addition, it will be so wonderful if we have system that can prioritized major to minor interactions specifically for individual patient.
-
2022-10-05 at 9:31 pm #38549SIPPAPAS WANGSRIParticipant
Thanks for sharing. I think the problem of dismissing alert popups are every where! It is the most basic feature which I believe most HIS program would have but it really up to the UX/UI design whether it is good enough to catch users’ attention or not.
-
-
2022-09-29 at 9:50 pm #38447Siriphak PongthaiParticipant
Again, I would like to share one decision support system (CPOE) when I used to work at hospital.
In the hospital database system, there are patient’s information such as patient’s allergy history, current medications, past medical history, etc.
As I worked in pharmacy department, definitely medications are involved. When I check completeness and correctness for individual patient, there are popup alert for drug allergy or drug interactions. Sometimes patients allergic to shrimp and medicine composes of chitosan so, the popup will help in detecting and preventing adverse event and allergy. However, in some cases, patients allergic to one of NSAIDS but they can take another in the same drug group, the popup is still alert in all steps. It makes physicians and healthcare providers annoyed of this duplicated popup alert because they have to check and acknowledge boxes. This is also called alert fatigued.
Factors that influence decision support system implementation include those HOT-fit framework mentioned in Kilsdonk E, et al. But what I most concern is how organization provide hands-on user training prior to system implementation, how easily IT service or support when providers encounter technical problems as well as how easy of system to user, and how much time providers have to spent on the program.
-
2022-10-01 at 6:19 am #38478Hazem AbouelfetouhParticipant
I don’t have any experience with DSS. However, I think modern EMR systems provide physicians with many alerts and reminders in the workflow in prescription, laboratory results, and writing clinical observation which is considered a part of Clinical Decision Support.
-
2022-10-02 at 8:17 am #38490Boonyarat KanjanapongpornParticipant
At, my current working place, Pharmacy store, there is one main software application to support Pharmacist which isn’t a sophisticated system but quite enough to enhance quality and safety of medicine used. This application could be used to do many tasks such as printing medicine label, recording patient history and calculating medicine dosage for children which are useful and support the decision for medicine dispensing. I would like to mention the function for calculating medicine dosage for children by weight since it had assisted me with clinical decisions many times. Personally, I rarely meet young patients so I sometimes feel nervous about the precise dosage especially when I had limited of time. With the support from this calculating function, precise and fast services could be made. So far, I have seen other Pharmacists use this function to support their decision as well.
I consider the affecting factor for implementing the system, medicine dosage calculation for children, by using HOT-fit framework which is the framework for HIS evaluation. There are few factors from Human, Organization and Technology which affect (positive and negative) to the performance and effectiveness of this decision support system.
Human – Giving myself as an example, I didn’t realize the existing function until other staff introduced me. So, Application training is the factor that would increase the compliance to the system.
Organization – Period of service time for one patient would be tighter if there are limited staff. Patients generally expect quick service from pharmacy store, so sometimes the system has to be skipped to reduce the process of service.
Technology – System designed is user friendly and no ambiguous functions. This only needs a few clicks to complete all the calculation. However, system error and delay can occur. I had experience with the system error once which caused the delay of the service and therefore dissatisfied the patient.
With the evaluation above, If I could support the adjustment of the factors that create negative effects to the system, the capability and effectiveness of the system implementation would be increased. -
2022-10-03 at 2:32 am #38504Kawin WongthamarinParticipant
When I was a first-year intern I found it really difficult to choose the right diabetes medication for my patient because there are many different diabetes medications, contraindications, and the need to adjust the dosage according to GFR (kidney function value). For this reason, I consulted a nephrologist and created a mini-CDSS program for me and my friends (intern-1 doctors) to use. I coded my own program in Python. Just filling in the GFR and related medical conditions, the program will calculate the appropriate dosage and does not recommend drugs that are contraindicated ( Picture ).
My evaluation for this CDSS is as follows:
Human
-System Use: This program is used by just graduated physicians to reduce mistakes in prescribing diabetes medications. Experts or experienced doctors do not need to use this program.
-User Satisfaction: People who use this program are very satisfied because it is easy to use compared to other dosage calculator programs.Organization – This program is not run by my hospital. This makes disadvantages of the program not being updated and maintained after I had moved out from this hospital.
Technology – This program works easily and fast. It can be improved or bug fixed by myself. But there is a challenge in data validity as new research comes out every day so the accuracy of the program also changes over time.
Net Benefits – Since there is no concrete evaluation of outcomes, I think patients receive more accurate treatment and increase patient safety.
-
2022-10-08 at 10:01 pm #38592Kansiri ApinantanakulParticipant
Thank you for sharing.
Your project is interesting.
I think this system could be developed and support in real clinical practice.
-
-
2022-10-05 at 6:06 pm #38541Tanyawat SaisongcrohParticipant
There are several examples of DSS that I experience in my practice;
(1)Medication alert box embedded in EMR
-Drug allergy history pop-up alert box at initial access to EMR of particular patient. It’s a must-read to continue accessing EMR
-Duplicate therapy pop-up alert box that need override at medication order page, for example, NSAIDs used in the setting of initial intravenous stat dose then continue in oral form.
-Remainder medication (from last visit) notification when using re-medication order.
(2)Medical reconcile form for each in-patient, both physician and nurse have to review and in some cases might need additional review by clinical pharmacist. The printed form is an automated current medication in EMR.
(3)Pink sticky note from clinical pharmacist on in-patient charts’ cover. It’s about possible drug interaction, proper adjusted dosage recommendation or another particular alert and physicians need to review.
(4)Re-visit pop-up alert box embedded in EMR for those patients who re-visit within 24 hours, either the same department or different departments.
(5)In some private practice, there’s an exclamation sign button for the patient who has remarkable insurance coverage issue.In my opinion, all the alerts and reminders mentioned above work very well. It does improve the outcome especially patient safety. It takes less than a minute to read and take action to those boxes compare to the management time spent for the sequelae or adverse events that might happen to the patients.
According to HOT-fit model, it is more positive than negative in my community I believe.
Technically, those are embedded in existing EMR, user-friendly and minimised interaction. Its design is visually oriented and pop up at the corner on screen. The available system is quite fitted to our routine care. For the system use, we received initial training session (for EMR first use) which include those alert boxes. -
2022-10-05 at 9:28 pm #38548SIPPAPAS WANGSRIParticipant
At my current hospital, the most closest thing to CDSS would be an alert system for drug allergies and potential drug interactions. When doctors are prescribing medicines using HIS system, if there are known drug allergies for a specific patient, it will pop up. Sometimes, some clinicians would just ignore and to be honest, there are prone to be error such as prescribing a drug that patient is allergic to. I have experienced and I think the cause of ignoring them would be that the alert is not intuitive, too much text and irrelevant information and speaking technically, they are too easy to dismiss without user action (such as entering a PIN, or countdown at least 3 seconds, etc.) to tell whether they are really acknowledge or not.
In my opinion, the factor that influence clinical decision support system would be the knowledge and logic behind them provided that (to tell the truth) many senior clinicians are very conservative and most likely to decide which treatment to give to the patient based on their experiences. They are not likely to believe the computer anyway. But for other system such as alert messages for drug interactions and allergies, dosage calculations based on patient underlying diseases or laboratory values are still very important.
-
2022-10-08 at 9:59 pm #38591Kansiri ApinantanakulParticipant
Thank you for sharing.
I agree that the user acceptance is the important key to efficient system implementation
The people in X generation may not familiar with technology and tend to opposed something like this.
-
-
2022-10-08 at 9:57 pm #38590Kansiri ApinantanakulParticipant
I personally not quite experience in CDSS using since the worked in clinical research fields after graduation.
However, I would like to share one of my heath informatics project which I used to work with during start up process of the project. This may considered the closest CDSS for me.This system is aimed the recommended the most relevant trade drug product available in drug store to the pharmacist and customer.
The rationale of using is based on indications, dosage form, generic name, strength.
The program would calculate the similarity score and create list of recommended productsThe prototype one is quite work well. The recommendation is rationale and reasonable in both pharmacist and patient perspective.
Factor influencing implementation:
1) The user acceptance of the system: since the system would be available for both pharmacist and patient end. It’s necessary to ensure that the recommendation is rationale and the user interface is easy to use.
2) The maintenance of the system: the system must be tested for several times in various drug indication to ensure generalizability. The developer and pharmacist team need to ensure that the system is easy to maintenances. For example, adding new products or deleting products.-
2022-10-20 at 1:01 pm #38777Tanatorn TilkanontParticipant
Thank you for sharing your interesting system. I believe this system is very benefit to both the pharmacist and customer, because sometimes it takes time for pharmacist as well to think of another available trade name in the pharmacy store if the store’s system did not provide any suggestion or search engine.
Since the users are both pharmacist and patient, the user interface should as simplest as possible to ease the user to get the most benefit from the system. Moreover, the update of data (available trade name) should be continuously updated due to the on-going of new drug produced in everyday. Please feel free to correct me if I misunderstand your point. Thank you.
-
-
2022-10-20 at 12:34 pm #38776Tanatorn TilkanontParticipant
In the private hospital that I used to work at as a part-time, I saw that they have implemented the decision support system which was embedded with the EMR system, such as the following.
– Pop-up Warning for renal insufficiency patient and drug allergy history of patient
– Alert notification for high risk of drug interaction, drug allergy and duplicated prescription.
Firstly, it was implemented by showing pop-up alerts. Later, there is a color highlighted on the drug name in the system to notify that there’s something we need to be concerned about in ordering.Another decision support system are Smart ward web application – physicians can prescribe medicines via a single iPad, and the web application which provides the patient drug profile for medical reconciliation and dose calculation.
Factors influencing these implementations might be the user’s awareness of the change and ability to get the most benefit from this support. Another factor might be the difficulty of opening many applications to get everything they would like to have.
In my point of view, it would be better if everything was combined in one application. The potential healthcare users could be involved from the early stage and get all the feedback for improvement. There will need to provide training prior to implementing and acknowledging users about all the features they have and how to get most benefits from this system.
-
-
AuthorPosts
You must be logged in to reply to this topic. Login here