The CE mark endorses the specific device for sale in the EU not the manufacturer - if the manufacturers quality system/procedures are all that's verified, in all cases, then why not just endorse the manufacturer rather than type-test individual medical products when applicable?
It depends on the annex(s) that apply in the MDD what "standards" are expected of the manufacturer and their products - as I've said It's sort of a sliding scale, the higher the risk classification of the device, by virtue of what it does and how it does it, the more safeguards are built-in to protect the consumer.
At one extreme the manufacturer declares conformity based on the type of device being developed presenting a low-risk and at the other a certificate of conformity may only be issued by a notified body after a range of checks on design, manufacture, production, quality processes and samples of the product are tested, i.e. type-testing.
For example, the full quality assurance route in the MDD, under annex II, the manufacturer has to demonstrate that the specific design, manufacture and quality processes are in place for that product. In all cases, as far as I'm aware, there's a requirement for manufacturers to demonstrate good practice by following the relevant "standards" where necessary. Usually this is easier achieved through following existing manufacturing BE/EN/ISO, harmonised where possible, standards.
We expect this minimum standards of all medical electrical devices all of the time don't we? We look for conformance with every time a new item (with the CE mark) arrives at acceptance, eh? Why do you think most manufacturers tend to comply? Perhaps it's because there are systems of regulation and legislation in place - the regulatory framework surrounding CE perhaps - that expect it as well.
For relatively low-risk devices you may have a point, since there's more reliance on the manufacturer to classify the risk of a product properly thus apply the correct processes to obtain the CE mark for it. But for low risk devices, if they're crap, then when they fail or don't perform properly the threat to patients and users should be relatively insignificant. Fewer high-risk devices that are positively unsafe are likely to pass through a test-house are likely to be endorsed with a CE mark than if no system was in place, in my opinion.
