Pre-emptive asthma risk notifications based on medicament device monitoring

1. A method comprising:

accessing a set of historical rescue inhaler usage events recorded for a population of patients within a region, each event of the set of historical rescue inhaler usage events previously received from one of: a client device, the rescue inhaler unit, or a medicament device sensor removably attached to the rescue inhaler unit and configured to monitor medicament usage of the rescue inhaler unit, wherein a historical rescue inhaler usage event is recorded when the rescue inhaler unit dispenses a rescue medication to one of the patients within the region;

determining a region-specific baseline risk threshold based on a value representing a percentage of the set of historical rescue inhaler usage events recorded for the population of patients within the region over a time period;

responsive to a triggering condition,

accessing a set of input values;

inputting the set of input values, and the region-specific baseline risk threshold into a machine-learned model to determine a risk score for the region, the machine-learned model trained using a training dataset to determine a set of parameter values, the training dataset comprising input values recorded during a historical rescue inhaler usage event in a region and labeled with a risk score determined for the region during the prior rescue inhaler usage event and the set of parameter values representing a relationship between each input value and the risk score determined for the region; and

generating, for presentation on a computing device of each patient in the region, a notification comprising an updated adherence schedule for dispensing rescue medication to the patient based on the risk score.

2. The method of claim 1 , further comprising:

receiving the set of historical rescue inhaler usage events from the client device, the medicament device sensor, or the rescue inhaler unit.

3. The method of claim 1 , wherein the region-specific baseline risk threshold is determined with respect to a prior time period, the prior time period including events from a window of time preceding a current time.

4. The method of claim 3 , wherein the region-specific baseline risk threshold is determined as a percentage of a sum of historical rescue inhaler usage events within the prior time period.

5. The method of claim 1 , wherein the region-specific baseline risk threshold is periodically determined.

6. The method of claim 1 , wherein the triggering condition includes a threshold change in a physical location of the client device.

7. The method of claim 1 , wherein the triggering condition includes a periodic conclusion of a time interval.

8. The method of claim 1 , wherein the triggering condition includes a threshold change in an input value of at least one of a plurality of parameters.

9. The method of claim 1 , wherein the set of parameter values are determined using a boosted gradient model.

10. The method of claim 1 , wherein the risk score is a numerical value representing a likelihood that a user in the region on a current day will experience a number of rescue inhaler usage events exceeding the threshold that day.

11. The method of claim 1 , wherein the set of parameters include at least one contextual data parameter.

12. The method of claim 1 , wherein the set of parameters include one or more of the following:

at least one current region parameter, further comprising:

a current latitude and longitude coordinate of the client devices of the plurality;

a geographical region;

a current date;

a difference in number of controller puffs taken by the patient and prescribed for the patient; and

a count of rescue puffs taken for a rescue inhaler usage event.

13. The method of claim 1 , wherein the set of parameters include one or more of the following:

at least one air pollutant parameter, further comprising:

a concentration of ozone molecules (O 3 );

a concentration of nitrogen dioxide molecules (NO 2 );

a concentration of sulfur dioxide molecules (SO 2 );

a concentration of carbon monoxide molecules (CO);

a concentration of carbon dioxide molecules (CO 2 );

particulate matter, 2.5 micrometers or less (PM 2.5 );

particulate matter, 1 micrometer or less (PM 1.0 ); and

air quality index (AQI).

14. The method of claim 1 , wherein the set of parameters include one or more of the following:

at least one weather parameter, further comprising:

temperature;

humidity;

wind speed;

wind direction;

station pressure;

precipitation;

thunderstorms; and

visibility.

15. The method of claim 1 , wherein the set of parameters include:

temperature,

relative humidity,

wind speed,

nitrogen dioxide (NO 2 );

sulfur dioxide (SO 2 );

particular matter 2.5 micrometers or less (PM 2.5 ); and

particulate matter 1 micrometers or less (PM 1.0 ).

16. The method of claim 1 , wherein the risk notification is presented at a first time that is different from a second time when the region-specific baseline risk threshold is determined.

17. The method of claim 1 , wherein the notification comprises informational content regarding the triggering condition.

18. The method of claim 1 , wherein the notification comprises informational content regarding at least one of:

the triggering event;

the region-specific baseline risk threshold; and

a risk categorization based on the regional risk score.

19. The method of claim 1 , wherein the risk notification further comprises informational content regarding an inhaler rescue usage event, wherein the informational content further comprises a subset of parameters responsible for a change in risk categorization compared to a previous time period.

20. The method of claim 19 , wherein the risk notification further comprises informational content regarding the rescue inhaler usage event, wherein the informational content further comprises a recommendation regarding how to prevent future rescue inhaler events and is based on the subset of parameters responsible for the change in risk categorization.

21. A non-transitory computer readable storage medium comprising computer program instructions that when executed by a computer processor cause the processor to:

access a set of historical rescue inhaler usage events recorded for a population of patients within a region, each event of the set of historical rescue inhaler usage events previously received from one of: a client device, the rescue inhaler unit, or a medicament device sensor removably attached to the rescue inhaler unit and configured to monitor medicament usage of the rescue inhaler unit, wherein a historical rescue inhaler usage event is recorded when the rescue inhaler unit dispenses a rescue medication to one of the patients within the region;

determine a region-specific baseline risk threshold based on a value representing a percentage of the set of historical rescue inhaler usage events recorded for the population of patients within the region over a period of time;

responsive to a triggering condition,

access a set of input values;

input the set of input values, and the region-specific baseline risk threshold into a machine-learned model to determine a risk score for the region, the machine-learned model trained using a training dataset to determine a set of parameter values, the training dataset comprising input values recorded during a historical rescue inhaler usage event in a region and labeled with a risk score determined for the region during the prior rescue inhaler usage event and the set of parameter values representing a relationship between each input value and the risk score determined for the region; and

generate, for presentation on a computing device of each patient in the region, a notification comprising an updated adherence schedule for dispensing rescue medication to the patient based on the risk score.