It's not that nobody is building the technology because they're concentrating on other things, but rather nobody is building it because they know it won't sell, because it would have to go sixteen thousand layers of red tape and many tours through Congress before anything got done.
I think there's probably a lot of scope for solving problems with fairly simple apps, but this example seems like a terrible one.
VA's got no end of computer systems, where its staff hand-enter information in one system, print out its report, only to scan it, fax it across the country, where someone else will hand-enter it into... A different system.
Engineers, consultants, and Serious Business Contracts were involved in the construction of this abomination - and while they were all in compliance of the standards that the government expects from its vendors, none of them delivered a good product.
On the other hand, something like a basic CRUD app could really help out a charity or non-profit that serves disadvantaged populations, and that could be built in Rails or Django for just a few days of developer time.
The trouble was each flipping charity/bank wanted a bespoke version some of the changes required by the banks where absoluty nit picking.
The product had some good press for the Minister but could we get charity's/banks to hold their nose and just use one form! NO
Wonder if any VC in old street/silicon roundabout want to try again I still have few contacts form my days at Poptel
The problem with vets is not the absence of software. I am rather tired of seeing this particular horse beaten.
The screening part is tricky though, as it can quickly get political.
The main problem lies with the paper backlog of attempting to get people into the system. I'm personally not aware of how/why military personnel's medical records don't transition directly into the VA, but what happens in practice is a gap between military med. and the VA, where vets find themselves forced to file claims. I suspect that this is in large part because the set of "Vets" and the set of people qualifying for VA care, while overlapping, are not the same (that assertion comes from anecdote: I had an ex in the military, and when she finished her tours I recall that she was not qualifying for VA care because she hadn't been disabled nor gone career, as she explained it). While it's fair to ask why this process is done on paper, it's worth-while to note that prior to just the last handful of years the portion of health records that had been digitized was nearly nil (up until about 5 years ago it was standard practice in most hospitals and private clinics to go digital-to-paper-to-digital). Essentially everyone's records were on paper, and all record transfers happened via fax (and still do, if a receiving physician is not inside the same hospital system on the same EMR as the referring physician).
source: work in hospital QI. I haven't worked in the VA, personally, so I am relating a lot of stuff second-hand, but the boundary between most hospitals' QI staff, the IHI, and the VA tends to be pretty porous, if you're in a healthcare capitol city - so I've heard a lot about the VA from co-workers that have worked there, or are currently engaged in ops research there.
There are also about 5 levels of bureaucracy to manage a single system within DOD. VA is amazing in comparison (it was really telling talking to doctors who had civilian jobs in the VA but were military reservists and working temporarily in military hospitals...)
Medical records aren't sufficient to make a determination of disability, though, which is mainly what this is about. Having the medical records helps, having personnel records helps.
Another problem is the VA for a long time was focused on long-term care for WW2/Korea elderly vets, and various mentally/physically disabled Vietnam vets. The population of seriously injured younger vets (who could still work, and have long and otherwise productive lives ahead of them, but need specific disability care) is a pretty new thing. That, and the huge number of PTSD/TBI/psych issues.
Medical records keeps track of medical procedure, patient history, what medications they're on, etc. In the private sphere, this is what a hospital would use.
What's probably at cause for the huge backlog is actually getting people into the system and approved for care. In the private sphere, this is what an insurance company would use.
Or this could be just your standard set of political exaggerations.
When designing (evolving, inventing) rules, regulations, policy, exceptions, limitations, audits, checks, balances, the bureaucracy's princes seem to ignore the bigger picture of what it's all become: a mess.
It's not just government systems either. There's a reason why people joke about their phone bills being difficult to understand. Same goes for most utilities.
Now, once you have an incoherent, contradictory, thousand page mishmash set of rules, every engineer worthy of the title is going to stay as far away as possible. So, the whole mess gets dumped on an army of coders from a big consultancy.
Perhaps instead of focusing on the useless software written to fit a pile of inane rules, people could focus on the economic and operational benefits of redoing the entire policy regime with a zen-shaped comb.
Software is an enabler. It's actually covered-up a lot of really awful decision-making. The US tax code could never have become what it is if developers hadn't been able to consistently keep up with crazy new patches to implement it: nobody could possibly manage it by hand.
More than the technical disgust of building a policy-laden system, marketing is terrifically difficult under regulation. Consumers get scarred by inconsistent regulatory rules and stop being willing to change processes because it's too frightening to possibly run afoul of policy.
This is, by the way, a great opportunity for organizing regulatory consultants. Such a service might be an important forerunner or partner to ventures trying to change heavily regulated industries.
Changing a massive medical system from paper-based to digital is "simple"?