METHODS FOR DETERMINING RISK AND TREATING DISEASES AND CONDITIONS THAT CORRELATE TO WEATHER DATA
First Claim
Patent Images
1. A method for generating a model set of equations for predicting a risk of a subject who experiences adverse medical events associated with the weather, of experiencing a new-onset event (NOE), the method comprising:
- a) identifying a climate region of interest;
b) collecting daily mean barometric pressure (BP) data for a time frame and dividing the days of the time frame into at least upper, middle and lower quantile BP days to identify the upper quantile BP days;
c) collecting daily NOE data for a subject cohort consisting of subjects known to suffer from the adverse medical event for the time frame and calculating a daily incident rate (IR) of NOEs for each day of the time frame and dividing the days of the time frame into at least upper, middle and lower quantile IR-NOE days to identify the upper quantile IR-NOE (UQ-IR-NOE) days;
d) determining a relevant number of seasons based on an association between the upper IR-NOE quantile days identified in c) and the upper BP quantile days identified in b);
e) collecting hourly weather data for the time frame for a number of weather parameters and determining a set of weather variables;
f) employing a generalized linear regression analysis to generate a rank for each weather variable as a predictor of the UQ-IR-NOE days for each relevant season, for each BP quantile;
g) identifying a predictive equation using a forward stepwise approach,wherein the model comprises a set of one or more equations for predicting the risk of a subject experiencing a new-onset event (NOE) at the completion of step (g).
2 Assignments
0 Petitions
Accused Products
Abstract
Methods and model equations are provided for predicting a risk of a subject who experiences adverse medical events associated with the weather, of experiencing a new-onset event. Methods include a) identifying a climate region of interest; b) collecting daily mean barometric pressure (BP) data for a time frame and dividing the days of the time frame into at least upper, middle and lower quantile BP days to identify the upper quantile BP days.
-
Citations
26 Claims
-
1. A method for generating a model set of equations for predicting a risk of a subject who experiences adverse medical events associated with the weather, of experiencing a new-onset event (NOE), the method comprising:
-
a) identifying a climate region of interest; b) collecting daily mean barometric pressure (BP) data for a time frame and dividing the days of the time frame into at least upper, middle and lower quantile BP days to identify the upper quantile BP days; c) collecting daily NOE data for a subject cohort consisting of subjects known to suffer from the adverse medical event for the time frame and calculating a daily incident rate (IR) of NOEs for each day of the time frame and dividing the days of the time frame into at least upper, middle and lower quantile IR-NOE days to identify the upper quantile IR-NOE (UQ-IR-NOE) days; d) determining a relevant number of seasons based on an association between the upper IR-NOE quantile days identified in c) and the upper BP quantile days identified in b); e) collecting hourly weather data for the time frame for a number of weather parameters and determining a set of weather variables; f) employing a generalized linear regression analysis to generate a rank for each weather variable as a predictor of the UQ-IR-NOE days for each relevant season, for each BP quantile; g) identifying a predictive equation using a forward stepwise approach, wherein the model comprises a set of one or more equations for predicting the risk of a subject experiencing a new-onset event (NOE) at the completion of step (g). - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 25, 26)
wherein N is the number of subjects in the cohort eligible to have an NOH on a given day and is the denominator in the IR-NOH calculation, and ε
is an error term of GEE regression modeling.
-
-
17. A method of assessing a risk of a subject predisposed to experiencing weather-associated adverse events for experiencing a new onset event on a given day, the method comprising:
-
a) determining a climate region associated with the geographical location of the subject; b) determining the relevant season in which the given day falls; c) determining the projected mean BP for the day and identifying it as an upper, middle or lower quantile BP day; d) selecting an equation from the model set of equations according to claim 15 specific to the determined climate region, the determined relevant season and the determined BP quantile; and e) entering weather variable data for the given day into the selected equation to yield an assessment of the risk.
-
-
18. The method according to claim 17, wherein the weather-associated adverse event is associated with a condition selected from asthma, emphysema, depression, cardiovascular disease, arthritis, artherosclerosis, and diabetes.
-
19. The method according to claim 18, wherein the weather-associated adverse event is associated with artherosclerosis and comprises heart attack or stroke.
-
20. A method of assessing a risk of a subject residing in climate region Cfa and predisposed to experiencing weather-associated adverse events for experiencing a new onset migraine headache on a given day, the method comprising:
-
a) determining the relevant season in which the given day falls; b) determining whether the given day is an upper, middle or lower BP tertile day; c) selecting an equation from the model set of equations according to claim 16 specific to the determined relevant season and the determined BP tertile; d) entering weather variable data for the given day into the selected equation to yield an assessment of the risk.
-
-
21. The method according to claim 1, wherein the subject cohort is further controlled according to factors known to influence frequency of a medical condition precipitating the adverse event associated with the weather.
-
22. The method according to claim 21, wherein the factors are selected from race, gender, age, socio-economic status and combinations thereof.
-
23. A method for reducing the risk of a subject predisposed to experiencing weather-associated migraine headaches for experiencing a new onset migraine headache on a given day, the method comprising:
-
assessing the risk according to claim 20; treating the patient prophylactically to mitigate or avoid the migraine headache if the day;
is assessed as more likely than not to be a UT-IR-NOH day.
-
-
25. An article of manufacture comprising computer-readable code for implementing the method according to claim 16.
-
26. The article of manufacture according to claim 25 comprising a mobile application software product.
-
24. A method for increasing efficiency in a hospital staffing and resource commitment by predicting high admission days, a “
- high”
admission day being defined as a day falling in an upper quantile of the hospital'"'"'s daily admissions for a year, the method comprising;1) generating a model set of equations for predicting a risk of subjects who suffer from weather-associated medical conditions of being admitted to the hospital, wherein “
generating”
comprises the steps of;a) identifying the climate region in which the hospital is located; b) collecting daily mean barometric pressure (BP) data for the year and dividing the mean BP data into upper, lower and middle quantiles; c) collecting daily hospital admissions data for a subject cohort for the year, said subject cohort consisting of subjects known to suffer from a weather-associated medical condition and who have been admitted to a hospital at least once previously due to experiencing an adverse event associated with the condition, to calculate a daily admission rate (AR) for each day of the year and to determine an upper quantile of days associated with the AR; d) determining a number of relevant seasons based on regression analysis of the upper BP quantile days and the upper AR quantile days; e) collecting weather parameter data across the year; f) employing GEE modeling to generate a rank for each weather variable as a continuous predictor of the upper quantile AR days for each relevant season, for each BP quantile; g) identifying a best predictive single variable equation based on p-value and QIC fit of the first-ranked weather variable in each relevant season, for each BP quantile; h) adding the next-ranked weather variable to the identified equation from g) in each season, for each BP and determining if fit improves; i) repeating step h) until addition of the next-ranked weather variable fails to improve fit, wherein the model comprises a set of equations for predicting a risk of subjects who suffer from weather-associated medical conditions of being admitted to the hospital at the completion of step i); 2) employing the model to determine which days are likely to be upper quantile admission rate (UQ-AR) days; and 3) staffing the hospital and committing resources to the hospital on the basis of the determination in step
2).
- high”
Specification