Open Source Could Have Saved Ontario Hundreds of Millions

Platinum Dragon writes "Ontario's auditor-general released a blistering report this week detailing how successive governments threw away a billion dollars developing an integrated electronic medical record system. This CBC article highlights an open source system developed at McMaster University that is already used by hundreds of doctors in Ontario. As one of the developers points out, 'we don't have very high-priced executives and consultants,' some of whom cost Ontario taxpayers $2,700 per day." The McMaster University researchers claim their system could be rolled out for two percent of the billion-dollars-plus already spent on the project. The report itself (PDF) also makes note of the excessive consultation spending: "By 2008, the Ministry’s eHealth Program Branch had fewer than 30 full-time employees but was engaging more than 300 consultants, a number of whom held senior management positions."

Comments

Re:Government at its finest

[Interoperable]

This one's been a classic example of government at work. From dubiously awarded contracts and an unusually early bonus, both which contributed to the resignation (firing) of the head of eHealth, to the boondoggle in the article. This thing has been mismanaged from the get-go and it's reflecting pretty poorly on the premier and government.

Re:Perfect Example

[c6gunner]

And it isn't like private healthcare is not around when there is a socialized system anyway. You get a choice.

Not in Canada you don't. The only way for me to have a choice is to go to another country.

The issues are not simple

[XB-70]

I have had meetings with senior bureaucrats and politicians in Ontario about FOSS. Moving free software into the government sphere is really difficult. Firstly, the biggest fear is change. If you implement a change which impacts 60,000 employees, if there is a problem, the person who made the decision to change software will be 'implicated' and blamed etc. etc. No one wants to stick their necks out.

The only way to counteract the problem is if you get backing at a very high level (from the Premier and his Cabinet). During the late '90s all the ministries had to convert and conform to one accounting standard. The push-back from all levels was incredible. It was only because people were threatened with being fired that the project got enough traction to be implemented.

This is what Open Source software is up against. It's truly brutal. That said, never give up fighting, but it has to be done at the highest levels.

Re:Could open source really do the job?

[h4rm0ny]

Also don't forget that somebody had to pay for the open-source system to be developed. I somewhat doubt that anybody spends their spare time hacking away on electronic health record databases.

Hey now - I did! :)

This was in the UK and I no longer work in the NHS. But in Primary Care (i.e. your local doctor's, not a hospital), there are a handful of providers of medical systems with a couple of really big ones (EMIS and Synergy). I have a lot of experience with Synergy and quite frankly, it's a mess. I think it got a little better, but in the modern age we could do so very much better. And it wouldn't actually take much developer resource to come up with it. I noodled around with writing an alternative one, but I simply couldn't dedicate the time. But it would have taken a small team of decent programmers (maybe four of us) around a year to make a functionally equivalent system that did the job better, was open source and a good platform on which to build further. And I could have written conversion tools for the back end databases myself fairly easily. The issues are twofold. Firstly, the perceived hassle of migrating to a new system and secondly getting the license from the UK's Department of Health. The latter would have been the showstopper. It's a Boy's Club. There are a lot of very hard-working people at the low levels of the NHS and - under Labour - quite a lot of over-spend and corruption at the top. Particularly in the area of IT.

I'm no longer in the NHS. I got fed up with the problems I had to deal with being caused at a level above where I could fix them. I would love to manage the development of an open source alternative to the existing systems though and I could do it for a tiny fraction of the budget allocated to other NHS IT projects - a sort of skunk works project.

Unfortunately, the New Labour government has been determined to do everything it can to privatise the NHS without committing the political suicide of admitting they're doing so. The damage behind the scenes to British healthcare is enormous.

Re:Its not just Ontario. The whole of the Australi

[FireFury03]

Developing an eHealth system costs money. End of story. At the end of the day it is better to roll out a eHealth system that is secure, reliable and well integrated than a system that is unreliable, unsecure and convoluted.

Here in the UK, the government has been putting billions into the NHS computer systems. From talking to people who work with them, the consultants responsible basically have no clue about PKI, so there goes your security. As for being reliable and well integrated, experience of past (very expensive) government IT projects makes me doubt that this is likely too.

At the end of the day, the government goes to one of the really big 2 or 3 IT companies to develop a system (I'm talking about you, EDS, Capita, etc.), get quoted a crazy amount of money, accept the quote and then watch as the whole thing becomes a disaster and goes many times over budget. Then when the next IT project comes up they go back to exactly the same company. It is true that there are a limited number of huge IT companies to choose from, but many of the IT projects could be done just fine by smaller companies, and wouldn't cost the earth, with the advantage that supporting small businesses is a Good Thing for the economy. However, the government won't use small businesses to do these jobs because doing so is seen as high risk - personally, I don't see how you can get much higher risk than using one of the big companies that seem to have a 100% record of screwing up projects. Hell, for the amount these big companies get paid, you could probably get 4 or 5 small companies doing exactly the same job as each other and then actually roll out the project that looks the most likely to succeed.

Re:Project was a flop... open source wouldn't save

[jimicus]

The same mindset that would have allowed for open source would have allowed for other "breaking the government waste" pattern activities.

Why buy and maintain MS-Office licenses when there's a better, free, alternative? Teh "Because ..." mentality.

I'm having exactly this conversation at work right now - and the economic climate means it's a much easier conversation for me than it was 2 years ago. However, it goes something like this:

Me: Yes, could roll out Open Office to everyone. No problem. And it's free.
PHB: Good, so what do we need to consider in terms of compatibility?
Me: You'll see 95-98% MS Office compatibility easily. Of course, if you want 99-100% compatibility with MS Office, it's going to have to be MS Office. This is true for all office suites - hell, it's true between different versions of Office.
PHB: Right, so anyone who deals with outside organisations on a regular basis needs MS Office.
Me: Well, you could exchange PDFs...
PHB: Anyone who deals with outside organisations on a regular basis needs MS Office. Who else?
Me: Okay.... well, you need to consider if less than 100% compatibility with existing files is a problem. Things like spreadsheets, big fancy documents...
PHB: Spreadsheets? OK, so the finance people need MS Office. Any others?
Me: Well, engineering say that having to deal with different formats internally will be a PITA...
PHB: So we get MS Office for the engineers....
Me: Right, you do realise that there's only one license being saved now?

Re:Government at its finest

[KillerBob]

Yup... it's actually illegal in Canada for corporations to make political campaign contributions to candidates or parties (either monetary or in goods/services). It's considered serious fraud to conceal the source of a donation (as in, something that could get a party de-listed as an official party*, and could get individual candidates sent to jail). And any donation over $200 is a matter of public record that anybody with an Internet connection can see and review, just by going to the Elections Canada website.

Because of that, while there are cases where a coproration tries to buy an individual candidate, it's a *lot* harder for a corporation to dictate public policy. It's just the way the party system works in this country.

*De-listing an official party has some serious negative effects on the party... First, they can't put their party name on the ballot with the candidates they endorse. That'll make it harder for voters who vote party line to know who they're going to vote for. Their party logo is no longer recognized as a trademark, meaning anybody can use it. Also, listed parties are entitled to have up to 60% of their campaign costs reimbursed on a riding-by-riding basis if they get more than 10% of the popular vote... when you consider that an election could, theoretically, cost a party upwards of $30million across all 308 ridings, that's pretty big. Listed parties are also entitled to free airtime on tv/radio in which to advertise, and they can transfer funds between ridings. Not only that, but the personal donation limit on campaign contributions is separated into several categories, each of which has the same upper limit, and is cumulative. For an independant, you can only donate to the candidate. For a registered party, you can also donate to the riding association and to the national party. (meaning you can donate 3x as much money). Nobody's ever been stupid enough to try the kind of electoral fraud that'd get a party de-listed.