If your respirator monitoring machine gets hacked, or your records can not be retrieved to show a drug sensitivity, you could die from complications in an ER. I know for a fact that the hospitals near me run Windows XP, connected to the public internet, on mission critical systems. If a system which controls a switchable lane freeway signal gets hacked, even with generic malware which accidentally causes a crash, a distracted driver could plow into another.
P.S. I'm facinated by your turn of phrase kettle of fish, here it would be a can of worms. Especially strange to me considering what I would think a kettle is [0][1][2][3][4]. Turns out there's such a thing as a fish kettle [5].
[0]: https://www.google.ca/search?tbm=isch&q=kettle [1]: https://www.google.co.uk/search?tbm=isch&q=kettle [2]: https://www.google.com.au/search?tbm=isch&q=kettle [3]: https://www.google.co.nz/search?tbm=isch&q=kettle [4]: https://www.google.co.in/search?tbm=isch&q=kettle [5]: http://www.phrases.org.uk/images/fish-kettle.jpg
They also found it was pretty hard to study general trends, due to an overwhelming number of factors and poor quality of source data.
If the effects of gun laws can be lost in noise among crime statistics, and the right of the people to keep and bear them shall not be infringed, it seems pretty outrageous to suggest that there is any reason to introduce equivalent laws which a) infringe the right of the people to keep and bear arms and b) have had no effect that can be detected in the statistics in any implemented form.
Vital records and the corresponding court records are an excellent, high-quality datasource. If your "life saving" laws can not be detected in vital statistics, then you have failed.