Introduction
Maine state attempted to develop a new system for handling patients’ and providers’ claims to meet the compliance with the Health Insurance Portability and Accountability Act of 1996 (HIPAA). The system was expected to give online access to the users and process the claims more accurately with a high processing rate. When the system was launched, it ended up with a loss and despair causing billions of payback by the government and leaving the providers frustrated. The state officials invested capital and manpower to recover the system due to which it was proven to be highly expensive than it was anticipated. The current paper discusses the points of failure of the system and what could have been done to prevent such loss.
Background
In the late 19th century, the states were very enthusiastic in upgrading their Medicaid claim processing systems. It should be noted that HIPPA played an imperative role in governing the changes that took place in the management of health records of patients. At the same time, it was observed that the records were kept private. Like other states, Maine also had to improve its basic claim handling system to meet HIPAA requirements. The federal Medicaid programs were also getting more demanding due to the additional health services. The purpose for improving the systems was to save states money that is spent on handling number of calls to the Bureau of medical Services by launching more effective systems and giving online access to the users.
On Jan 21, 2005, the state of Maine launched its brand new Web-based Maine Medicaid Claims System. The system was designed to facilitate the processing of Medicaid claims and payments. The fact remains that the previous Act that was implied for management of medical aid was Honeywell Framework. Later on, the Act was replaced with new frameworks because it was merely limited in terms of its operational pace. The system was also meant to comply with the Health Insurance Portability and Accountability Act of 1996 (HIPAA) which the state was not in compliance with. After declaring the need of a new system instead of upgrading the previous one, the state awarded the contract to CNSI for building a new high-end processing system for handling medical claims. When the system went, live, it met severe collapse causing a loss of around 30 million to the state. Right from the beginning, the system offered complicated issues and was unable to process the claims on time which resulted in thousands of complaints and system failure (Holmes, 2006).
According to Health Insurance Portability and Accountability Act of 1996 (HIPAA), all states have to upgrade their systems for data management related to patient health. In Maine the Honeywell mainframe was implemented which had certain glitches due to which it did not comply with HIPAA, therefore, the state decided to launch a completely new high tech system to improve health record management for the providers and patients. State officials realized that upgrading the existing software would be more troublesome as it was a very basic system (1970 vintage Honeywell mainframe) which would not be able to process around 12000 claims every week. So the IT experts of the state claimed that developing a new system would be cost effective and as well as easy to operate. However, states other than Maine concluded different results for improving their systems. Therefore, the state planned to give away the construction orders to the IT experts in DHS. The aim was to allow the IT experts to come up with rules based system that could be updated without a lot of hassle (Holmes, 2006).
The failure of Medicaid claim processing system occurred due to multiple reasons that made its success impossible. Right from the very beginning when the decision was made that instead of hiring professional claim processing system developers DHS will be building its system; things started to get absurd. DHS had intense lack of manpower and no experience for developing a system. Based on the fact that they have been operating the previous software for 25 years, they can build a new system. The claim proved wrong when in the first few weeks 50% of the claims were directed to suspend instead of being accepted or rejected. Although it was suggested that companies having experience in Medicaid system designing should be outsourced for such a huge project. Outsourcing is a process in which organizations hire external experts to take the service instead of using its internal resources (Laudon & Laudon, 2014). In the case of Maine’s Medicaid system designing the DHS team had no experience of developing a new program, and CNSI had no track record in developing Medicaid systems so the decision for not outsourcing the professionals proved to be very wrong.
The process of selecting the appropriate vendor against the quotation has to be conducted very carefully as it is the base of initiating a new project. For the development of a system package, the evaluation effort is a part of system analysis. The key elements which should be analyzed before commencing a project are resources, user friendliness, database requirements, documentation, vendor quality, cost effectiveness, installation efforts, maintenance requirements and flexibility in the system. The process of package evaluation is based on Request of Proposal (RFP), it a descriptive document which entails all the requirements of the software program and is submitted to the packaged software vendors (Laudon & Laudon, 2014). Therefore, preparing a practical and realistic RFP is an integral step in system development. When the state issued an RFP for a new system only two proposals were obtained. It can be notes that one is from CNSI ($ 15 million) and the other from Keane ($ 30 million).
Receiving only two proposals was a clear indication that the requirements of the RFP are not reasonable according to the relevant professionals. At this instant, the state IT experts should have reconsidered the demands of the project they have set up in order to make it more realistic and practical. As building a hi-tech system was not required, it was necessary that the system must be easy to use, more efficient and practical. The HIPAA act did not ask for any particular technology to be implemented rather the target was to improve patient health record maintenance and management for the users. If the project is failed to achieve its core target, it is considered as a failed project. For any technology to be adapted and operated successfully it is important that it should meet its objective as well as fulfill user requirements (Laudon & Laudon, 2014).
In addition, another flaw that was noted was when the orders for construction was given to the low bidder CNSI. It was noted that the bidders had little or no understanding of the medical aid. On the other hand, Keane had some experience in building Medicaid systems. Competitive bidding makes the decision making more rational and validated (Laudon & Laudon, 2014). The multiple bids offer a chance for detailed evaluation and risk analysis. This factor lacks in case of Maine’s Medicaid system as there were only two bids received against the RFP. The difference between the two bids also indicates that the low bid must not have taken all the aspects and intricacies of developing a Medicaid system into account. These factors should have been monitored in the initial phase before assigning the contract in order to ensure the practicality of the system.
For meeting the scalability scale asked by state officials, CNSI suggested using J2EE software language. It was a very risky decision to use such high technology in a complicated project. Medicaid claim systems involve various codes and decodes for tackling rules, services and rates. To translate all those codes into a system developed from scratch was again very difficult. Selection of appropriate software language is very important to ensure system efficiency and reduce errors. The Medicaid system must have been developed on programming language that runs successfully on Medicaid claim processing software. Easy software language makes the process smooth and more effective (Laudon & Laudon, 2014). The selection of high technology language again increased error probability in the system that was apparent in the system flaws.
Another big issue related to the project was narrow timeline for building such a project. In order to meet the target date given by HIPAA, the DHS’s head did not take the time for revisiting RFP although if the problems associated with the system and the loss it has caused are considered taking time for revising the RFP was worth investing. It could have prevented million of Dollars to the state if the project details had been reconsidered at that instant. Due to the understaffed DHS and CNSI representatives the team would not be able to meet the deadline because there was enormous data that had to be translated into a new language. As a result, the team started to take their decisions on satisfying Medicaid requirements and then reprogram the system after having guidance from Medicaid experts. This further delayed the development process of the system (Holmes, 2006).
Narrow timelines have been declared as one of the four main factors which result in system failure. Around 30 to 40 percent software technology projects are declared as ‘runaway’ projects as they have surpassed their target dates and anticipated budgets (Laudon & Laudon, 2014).
Another key issue related to the system design was its extreme intricacy. A lot of details had been incorporated into the system so that it could comply with HIPAA security requirements and become more accurate. The legacy system used to check the claims for three basic things: entry of the provider in the system, patient eligibility and if the service is claimable. The new Medicaid system was designed to check very minute details of the claims. It has 13 checks to verify before a claim is accepted. This made the system more sensitive and complicated which was beyond the Medicaid system requirement.
As the DHS and CNSI programmers could not meet the target date given by HIPAA, they started to look for shortcuts to get the system launched. It was another huge mistake. For instance: To save time rechecking of the system from end to end was not performed. A pilot test was conducted to check the processing with ten providers. Since most of the system was not ready by that time therefore; the claims were not checked to clear all the points in the system (Holmes, 2006). This is very important to invest time before giving a green signal to the main project (Laudon & Laudon, 2014). In this case, the state officials did not consider the importance of pilot testing and let the system go live; however, the c0onsequences they faced made them realize their big mistake.
The combined department of Health and Human Service HHS also did not provide any official training or guidance for operating the system to the providers and the staff who had to answer the calls in the Bureau of Medical Services (Holmes, 2006). For such a complex technology implementation staff training should have been given to avoid human errors (Laudon & Laudon, 2014). A number of errors would have been appeared due to mishandling and improper use of the system.
The state officials planned to pay back the providers if the system could not run successfully. As planned, a timeline was considered to make sure that all the claims are approved. In the case, there were some of the claims unapproved then the provider had to be paid by the state. After the system had gone live in the very first week, it was found that around 24000 of the claims were declared as suspended and could not be processed. The suspended claims are those which could not be regarded as approved due to lack of certain elements and could not be rejected as well. So, these files made a huge pile in the system which was a warning sign for the developers (Holmes, 2006).
Such an error could have been corrected, but the high percentage of unprocessed claims made it difficult. The reason was very complicated system design with detailed codes which kept the claims unprocessed declaring them suspended. This shows the importance of system design and language used to build the software. It should be compatible and user friendly to increase the efficiency of the system. The basic legacy system although was not very high tech but it had a percentage of 20% of the suspended claims. Again, this shows the incompetency of the developers who selected inappropriate programming language for the system that did not comply with the requirements of the Medicaid claim processing system. This decision making in the initial level is very important to ensure the implementation of technology and prevent its failure (Laudon & Laudon, 2014).
This system error resulted in thousands of unprocessed claims with each passing day, and tons of complaints about pending payment claims were received at the Bureau of Medical Service. The number of suspended claim was very high because of another system design flaw. It was programmed that if a claim is declared as suspended than each time it is entered in the system it will be directly rejected. As a matter of fact, there were only 1000 claims that were being considered. This raised a concern that the implementation of operations would require more than six months to visit all claims. On rechecking the issue, it was found that there was a severe problem in code and design of the system. An example could be taken about the claims that were 1000 lines. The system could only take the claims that were 1000 lines long. Any claim that was more than 1000 lines was programmed to get rejected automatically by the software.
Considering the problems that appeared in the system due to system design flaw it is found that developing a prototype before launching a project is extremely necessary. The preliminary model can be rechecked and transformed according to the requirements and it also saves a lot of money and time. Investing some time to build a prototype after finalizing the design would have been worthy particularly in this case where incomplete pilot testing and no prototype resulted in massive destruction and system failure. With planned iteration process, one can replace the unexpected rework which reflects providers’ requirements (Laudon & Laudon, 2014).
It is very common that systems do come up with errors, but these are fixed in order to keep it running. In the case of Medicaid system developed by DHS and CNSI the problem was that the number of errors was many, and the programmers were unable to resolve them. The programmers tried their best but could not find a solution to halt the continuously rising number of unprocessed claims that made the payback of $ 310 million to the providers by the state. Finally, the DHS department and state officials hired XWave for project consultancy and changed the project leader. The new team leader and consultancy firm found out that the reason behind so many issues related to the system and its repair are due to poor project management and lack of communication between the staff (Holmes, 2006).
Laudon declares that the reason for poor project management in most of the cases comes with restricted budget and timeline. Both of these can be observed in the case of Maine’s Medicaid system failure. Projects are launched with missing functional features due to lack of time that poses errors later in operating the system. It is reported that only 29 percent of the IT projects are delivered on time and within the budget with all the requirements met (Laudon & Laudon, 2014).
Conclusion
Maine’s Medicaid system of claim processing resulted in total failure in terms of cost effectiveness, efficiency and even in the output. A number of reasons have appeared on closely analyzing the facts that resulted in technology failure. The most important among them are the system design flaws which caused major issues in operating the system. Use of improper program language which could not meet the Medicaid requirements further enhanced the chances of errors in the system. The inappropriate RFP which caused low number of bids also show the impracticality of the project requirements.
The choice of vendor made on the basis of budget ignoring the expertise and experience was another major factor which lead to the disaster, Maine Medicaid system failure was not only a software collapse affecting an organization rather it crumpled the entire health care system of the state affecting the patients, providers as well as the state officials. The state also had to bear enormous capital loss as a result of system failure. The providers clearly declared the system flop and ineffective in accomplishing its claims that are to be easier, efficient and accurate. Although the state officials put efforts and money to recover the system and make it efficient but still it could not satisfy the user’s expectations.
References
Holmes, A. (2006). Maine’s Medicaid Mistakes. Web.
Laudon, K., & Laudon, J. (2014). Management information systems : managing the digital firm. New York: Prentice Hall.