U.S. and Canadian Mobile Medical App Regulations: A Comparison
Are you planning to conduct research involving software in Canada or thinking of expanding your mobile medical application (MMA) research into Canada? If so, you should be aware of important distinctions between the regulatory frameworks there and in the United States.
Generally, software products that are intended for use in the diagnosis or treatment of disease constitute medical devices on both sides of the border.1 Yet, the extent to which each side requires a premarket submission may vary.
There is precious little guidance in Canada pertaining to the ever-growing market for medical applications. What if, for example, a device encourages treatment compliance in the management of chronic disease? Maybe the MMA includes patient-provider communication functionality.
What is the comparative regulatory landscape for such devices?
In the U.S., device studies typically require an Investigational Device Exemption (IDE), an applicable exemption from the IDE requirement (including most Class I and few Class II devices), or a nonsignificant risk (NSR) determination by an IRB.
However, the FDA has signaled that it intends to exercise enforcement discretion for MMAs that:
- Provide periodic educational information, reminders, or motivational guidance to patients.
- Perform simple calculations routinely used in clinical practice.
- Keep track of medications and provide user-configured reminders for improved medication adherence.
- Help patients document, show, or communicate to providers potential medical conditions.
- Allow a user to, collect, log, track and trend data from a device to eventually share with a heath care provider, or upload to an online (cloud) database, personal or electronic health record.2
Relevant guidance in Canada is not as descriptive. Though central IRBs may make an NSR determination as a matter of policy, such determinations are not part of the Canadian regulations. Instead, the Medical Device Regulations require submission of an Investigational Testing Authorization (ITA) for all Class II–IV devices.3 Thus, only Class I devices avoid premarket submission.4
The Minister of Health published a guidance document on the risk-based classification for in-vitro diagnostic (IVD) and non-IVD devices. (Read our analysis of IVD device regulations in precision medicine trials.)
Software is classified as part of the overall function of the device, and in the case of Rule 7 governing the classification of IVDs, the highest class of any one device in a combination determines the ultimate designation.
Like the regulations, the non-IVD guidance is essentially broken into three groups: invasive, non-invasive, and active devices.5 A large segment of MMAs under which the FDA would exercise enforcement discretion will come under the catch-all Rule 7.
Rule 7 establishes a default Class I designation for all other devices that are not intended either to act as a calibrator, tester, or quality control support to another medical device or to be connected to an active device that is classified as Class II, III, or IV. These two carve-outs to the default classification for devices, which otherwise do not fall within Rules 4–6 of noninvasive devices, are deemed Class II.
The Rules that explicitly identify software all fall within the active device group:
- Rule 8 (1): [A]n active device intended to emit ionizing radiation, including any device or software intended to control or monitor such a device or directly influence its performance, is classified as Class III.
- Rule 9 (1): [A]n active therapeutic device, including any dedicated software, intended to be used to administer or withdraw energy to or from the body is classified as Class II.
- Rule 10 (1): [A]n active diagnostic device, including any dedicated software, that supplies energy for the purpose of imaging or monitoring physiological processes is classified as Class II.
- Rule 11 (1): [A]n active device, including any dedicated software, intended to administer drugs, body fluids or other substances to the body or withdraw them from the body is classified as Class II.
The Minister of Health earlier published two guidance documents on the classification of less risky devices, focusing on the difference between Class I and II medical devices involving software. Of note, software that is intended to be used to view images or other real-time data as an adjunct to the monitoring device itself for the purpose of aiding in treatment or diagnosis of a patient would be considered Class I.
Yet, medical device software that is an adjunct to another medical device and is involved in data manipulation, data analysis, data editing, image generation, determination of measurements, identification of a region of interest in an image, or identification (by an alarm or alert) of results from a monitor that are outside of an established range is a Class II medical device if it either:
- Provides the only means and opportunity to capture or acquire data from a medical device for aiding directly in diagnosis or treatment of a patient.
- Replaces a diagnostic or treatment decision made by a physician.
As you can see, the regulatory status of mobile medical applications is less explicit in Canada. Still, it is fairly clear that most noninvasive MMAs, if they fall within the medical device regulations, will receive Class I designation in Canada.
If there are questions about these rules, it is always prudent to engage early with the responsible regulatory body or expert consultants to assess your particular situation.
1 See R.S.C., c. F-27, s. 2 (available at http://laws-lois.justice.gc.ca/eng/acts/f-27/); 21 U.S.C. § 321(h) (available at https://www.gpo.gov/fdsys/pkg/USCODE-2010-title21/html/USCODE-2010-title21-chap9-subchapII-sec321.htm).
3 See SOR/98-282.
5 There is also a section describing special rules.