Healthcare IT: or how to exceed the budget by 420%
For the seed, to have the desire to read further: “ So far, ... has cost ... at least $ 115 million and annual maintenance costs are approximately $ 14 million. The project is five years late, incomplete, riddled with deficiencies, and much of the technology may already be out dated. "
And so, as you might already have guessed, we’ll talk about a rather common occurrence, although speakers at numerous conferences try to humbly keep silent about this, when an IT project (not even a Russian one) inflated and inflated for a long time, and it took, but suddenly burst .
- Eeyore, I brought you a present! Balloon.
- Thank you, Piglet, you are a true friend, not like some! Ball, it's so big, beautiful!
- Interestingly, it so loudly boomed! And where is my balloon, and where did this rag come from?
By the way, in the old days, your humble servant participated in testing some modules of this project when he went in some other beta. Even then it was clear that something was wrong here, but to prove this to a very large and motley project team, hung with certificates like a Christmas tree, where MBA and PMP were simply must-have by definition, it was quite difficult, if at all possible. Since then, I briefly followed its development and have a sufficient idea of its structure. And now, when publicly available data has appeared, we can talk more about this.
Let's start in order. In 2003, the whole world shuddered at the epidemic of the next super virus SARS or SARS. In Canada, which will be discussed later, about 400 people have become infected with the virus, 44 have died, and 25,000 Toronto residents have been quarantined. In such circumstances, the Ministry of Health could not but respond, and it was decided to create a national system for early warning of epidemics, the purpose of which was very concise " seamless public health system that will allow public health professionals to coordinate activities in a carefully planned infrastructure ".
The project began in 2004, initially it was expected that it would include 6 large modules, such as: immunization management (vaccination of the population), outbreak management (tracking of epidemics), communicable disease case management (tracking of sexually transmitted diseases), etc. The developer was IBM, which signed a fixed contract for $ 27.8 million (hereinafter Canadian dollars). Another $ 9.9 million was allocated for other needs, amounting to $ 37.7 million for the project. According to this contract, IBM was, roughly speaking, to go shopping, to pick up all kinds of products on the shelves and to make a yummy out of it, from which everyone would be indescribable delight. In IT language, ""IBM committed to integrating a number of different COTS (commercial off-the-shelf) products into a single solution that would meet the pan-Canadian system needs ."
It soon became clear that the semi-finished products on the shelves were somehow not at all suitable for the expected gourmet candlelight dinner at the national level. After which, IBM, without thinking twice, suggested creating the required product from scratch. The Ministry of Health did not come up with anything better than to agree (because the system was needed). After quickly rewriting the contract, breaking it into a series of Change Orders, IBM, following all the newfangled agile and other technologies in both project management and development, was supposed to provide a new system within a year (in 2007). The contract increased to $ 47 million.
And then came the financial outbreak (that is, 2008). Funding for all projects has been revised. The new contract provided for the creation of a system for $ 16.2 million by 2009. In reality, the ministry paid IBM $ 73.5 million.
In addition, IBM signed another contract for the development of test scenarios and conducting acceptance testing. But even this did not help much, at the time the system was commissioned (in 2010), 1,200 errors were detected, although the terms of the contract assumed that " the system contains no severity 1 or 2 defects and 5 or less severity 3 defects ". You will never guess what it was decided to do with these errors:
• allowed severity level 1 and 2 defects to be downgraded to level 3 with the identification of workarounds and removed ministry approval of workarounds
• changed severity deficiency levels, resulting in defects being assigned to lower severity levels ????
• limited the number of defects that IBM was required to fix to a combined national total of 1,450 annually and allowed jurisdictions to prioritize defects
Having taped all the unsightly parts with tape, the system went into production in 2011. Immediately another 11 thousand defects were discovered, including 63 Severity 1 and almost 2 thousand Severity 2. Among these 11 thousand, 6 thousand defects made the work of medical personnel with the system impossible. By December 2014, 2.5 thousand defects remained open.
And so, thanks to skillful computerization, the time of admission of one patientdecreasedincreased from 25-35 minutes to 35-52 with the new system. According to the estimates of doctors, when using the system, they will lose annually up to $ 4.5 million in profit.
It is worth noting that the problem of the speed of the system even with the simplest user actions, such as selecting a patient from the list, as well as the very confusing logic for performing everyday actions, such as creating a patient and linking it to another patient, was obvious even at the very early stages of testing. However, the desire to create a super-flexible system, each aspect of which can be customized according to the wishes of the customer, has done its insidious business.
What about finances? Initially allocated $ 27 million for 6 + 1 modules of the system at the time of the audit turned into $ 113 million. At the same time, only part of the agreed functions of the system turned out to be implemented. The cost of annual system support is $ 14 million.
The results of the audit of the system can be found at: www.bcauditor.com/pubs/2015/audit-panorama-public-health-it-system
And so, as you might already have guessed, we’ll talk about a rather common occurrence, although speakers at numerous conferences try to humbly keep silent about this, when an IT project (not even a Russian one) inflated and inflated for a long time, and it took, but suddenly burst .
- Eeyore, I brought you a present! Balloon.
- Thank you, Piglet, you are a true friend, not like some! Ball, it's so big, beautiful!
- Interestingly, it so loudly boomed! And where is my balloon, and where did this rag come from?
By the way, in the old days, your humble servant participated in testing some modules of this project when he went in some other beta. Even then it was clear that something was wrong here, but to prove this to a very large and motley project team, hung with certificates like a Christmas tree, where MBA and PMP were simply must-have by definition, it was quite difficult, if at all possible. Since then, I briefly followed its development and have a sufficient idea of its structure. And now, when publicly available data has appeared, we can talk more about this.
Let's start in order. In 2003, the whole world shuddered at the epidemic of the next super virus SARS or SARS. In Canada, which will be discussed later, about 400 people have become infected with the virus, 44 have died, and 25,000 Toronto residents have been quarantined. In such circumstances, the Ministry of Health could not but respond, and it was decided to create a national system for early warning of epidemics, the purpose of which was very concise " seamless public health system that will allow public health professionals to coordinate activities in a carefully planned infrastructure ".
The project began in 2004, initially it was expected that it would include 6 large modules, such as: immunization management (vaccination of the population), outbreak management (tracking of epidemics), communicable disease case management (tracking of sexually transmitted diseases), etc. The developer was IBM, which signed a fixed contract for $ 27.8 million (hereinafter Canadian dollars). Another $ 9.9 million was allocated for other needs, amounting to $ 37.7 million for the project. According to this contract, IBM was, roughly speaking, to go shopping, to pick up all kinds of products on the shelves and to make a yummy out of it, from which everyone would be indescribable delight. In IT language, ""IBM committed to integrating a number of different COTS (commercial off-the-shelf) products into a single solution that would meet the pan-Canadian system needs ."
It soon became clear that the semi-finished products on the shelves were somehow not at all suitable for the expected gourmet candlelight dinner at the national level. After which, IBM, without thinking twice, suggested creating the required product from scratch. The Ministry of Health did not come up with anything better than to agree (because the system was needed). After quickly rewriting the contract, breaking it into a series of Change Orders, IBM, following all the newfangled agile and other technologies in both project management and development, was supposed to provide a new system within a year (in 2007). The contract increased to $ 47 million.
And then came the financial outbreak (that is, 2008). Funding for all projects has been revised. The new contract provided for the creation of a system for $ 16.2 million by 2009. In reality, the ministry paid IBM $ 73.5 million.
In addition, IBM signed another contract for the development of test scenarios and conducting acceptance testing. But even this did not help much, at the time the system was commissioned (in 2010), 1,200 errors were detected, although the terms of the contract assumed that " the system contains no severity 1 or 2 defects and 5 or less severity 3 defects ". You will never guess what it was decided to do with these errors:
• allowed severity level 1 and 2 defects to be downgraded to level 3 with the identification of workarounds and removed ministry approval of workarounds
• changed severity deficiency levels, resulting in defects being assigned to lower severity levels ????
• limited the number of defects that IBM was required to fix to a combined national total of 1,450 annually and allowed jurisdictions to prioritize defects
Having taped all the unsightly parts with tape, the system went into production in 2011. Immediately another 11 thousand defects were discovered, including 63 Severity 1 and almost 2 thousand Severity 2. Among these 11 thousand, 6 thousand defects made the work of medical personnel with the system impossible. By December 2014, 2.5 thousand defects remained open.
And so, thanks to skillful computerization, the time of admission of one patient
It is worth noting that the problem of the speed of the system even with the simplest user actions, such as selecting a patient from the list, as well as the very confusing logic for performing everyday actions, such as creating a patient and linking it to another patient, was obvious even at the very early stages of testing. However, the desire to create a super-flexible system, each aspect of which can be customized according to the wishes of the customer, has done its insidious business.
What about finances? Initially allocated $ 27 million for 6 + 1 modules of the system at the time of the audit turned into $ 113 million. At the same time, only part of the agreed functions of the system turned out to be implemented. The cost of annual system support is $ 14 million.
The results of the audit of the system can be found at: www.bcauditor.com/pubs/2015/audit-panorama-public-health-it-system