Method and system for patient discharge optimization using improved probability

Abstract

A method (100) for determining a home care enhancement probability for an inpatient, comprising: (i) receiving (120) patient data from a medical record database; (ii) extracting (130) a set of promoted probability features from the received patient data; (iii) analyzing (140) the boosted probability feature set by a trained boosted probability algorithm (263) at a first point in time to determine a boosted probability of discharge of the hospitalized patient relative to continued hospitalization; (iv) comparing (150) the determined lifting probability to a predetermined lifting probability threshold to determine a recommendation, where the recommendation is to keep the inpatient hospitalized when the determined lifting probability is below the predetermined lifting probability threshold, and where the recommendation is to keep the inpatient hospitalized when the determined lifting probability is below the predetermined lifting probability threshold. When the determined lifting probability is higher than the predetermined lifting probability threshold, the recommendation is to discharge the inpatient; and (v) providing (160) the recommendation to the user via a user interface.

Description

Technical Field

[0001] The present disclosure generally relates to methods and systems for determining a home care uplift probability for a hospitalized patient. Background Art

[0002] Hospitalization can have serious complications for patients due to negative events such as hospital-acquired infections, falls, coma, and depression, which can make hospital stays longer. Complications can also occur due to side effects of hospital treatment. All of these hospitalization complications can lead to an increase of about 10% in total hospitalization costs.

[0003] As a non-limiting example, heart failure is a serious health problem affecting about 64.3 million people worldwide, with about 15 million cases in Europe and about 6.2 million cases in the United States. Heart failure is a chronic condition associated with decreased quality of life, frequent hospitalizations due to acute attacks of decompensation and high mortality. In the United States, the median length of stay for acute heart failure admissions is 2 to 6 days. However, in the EU, the average length of stay is even longer, with an average duration of 9.5 days. A total of 23% of patients will be readmitted within the first 30 days after discharge. About half of the patients will be admitted to the hospital at least once within a year after diagnosis, 20% of the patients will be readmitted within the same year, and more than 80% of the patients will be admitted within five years. Hospitalization for heart failure represents 1% to 2% of all hospitalizations, and heart failure is the most common diagnosis in hospitalized patients over 65 years old. Readmission may be caused by early discharge, insufficient patient education, insufficient follow-up after discharge, and the inherent hidden nature of the disease. This has an adverse impact on both patient burden and social healthcare costs. Excessive readmissions in the United States are penalized by Medicare, where the average penalty can be a 0.64% reduction in payments for each Medicare patient hospitalization. The penalties involved can be significant for hospitals.

[0004] Therefore, as a safe alternative to hospitalization due to acute episodes, hospital systems explore opportunities for virtualization of care. Instead of staying in the hospital, patients are discharged and receive care at home using digital virtual care support such as remote patient management (RPM). Therefore, hospitals aim to discharge patients earlier, but within safe boundaries and with post-discharge home care support in order to reduce inpatient costs and avoid hospital-acquired complications. However, these existing systems utilize static, outdated measurements and checklists when determining discharge and readmission probabilities.

[0005] US2020 / 411193A1 describes a decision support tool for discharging patients by predicting the probability of readmission for patients with pneumonia.

[0006] US2021 / 098090A1 describes identifying complex patients and predicting patient outcomes based on various factors including medical, socioeconomic, psychiatric, and behavioral.

[0007] CA2945134 A1 describes an overall hospital patient care and management system including a data storage device operable to receive and store patient data including clinical and non-clinical data. Summary of the invention

[0008] Therefore, there is a continuing need for systems and methods to more accurately predict the benefits of discharging an inpatient for home care. Various embodiments and implementations herein relate to methods and systems configured to determine a home care boost probability for an inpatient using a home care boost probability system. The system receives patient data from a medical record database and extracts a boost probability feature set from the data. The system then analyzes the boost probability feature set through a trained boost probability algorithm to determine a boost probability for the inpatient's discharge relative to continued hospitalization. The system compares the determined boost probability with a predetermined boost probability threshold to determine a recommendation and provides the recommendation to a user.

[0009] In general, in one aspect, a method for determining a home care promotion probability for an inpatient is provided. The method includes: (i) receiving patient data from a medical record database; (ii) extracting a promotion probability feature set from the received patient data; (iii) analyzing the promotion probability feature set by a trained promotion probability algorithm at a first time point to determine a promotion probability of discharge of the inpatient relative to continued hospitalization; (iv) comparing the determined promotion probability with a predetermined promotion probability threshold to determine a recommendation, wherein when the determined promotion probability is below the predetermined promotion probability threshold, the recommendation is to keep the inpatient hospitalized, and wherein when the determined promotion probability is above the predetermined promotion probability threshold, the recommendation is to discharge the inpatient; and (v) providing the recommendation to a user via a user interface.

[0010] According to an embodiment, the method further comprises: repeating at least the analyzing and comparing steps for a second time point; and generating an average promotion probability based on the determined promotion probabilities.

[0011] According to an embodiment, providing further comprises providing one or more of the following to a user via a user interface: patient information and the determined probability of boosting.

[0012] According to an embodiment, the first time point is triggered by a care event of the inpatient.

[0013] According to an embodiment, the first time point is a predetermined time point at a specific time during the hospitalization of the inpatient.

[0014] According to an embodiment, the recommendation is provided via a user interface of the clinical decision support system.

[0015] According to an embodiment, the recommendation is provided via a user interface of the mobile device.

[0016] According to an embodiment, the method further comprises: receiving a decision regarding the provided recommendation from a clinician via the user interface; and implementing the provided recommendation by the clinician. According to an embodiment, the provided recommendation is a recommendation to discharge the hospitalized patient.

[0017] According to another aspect, a system for determining a home care promotion probability for an inpatient is provided. The system includes: a trained promotion probability algorithm, which is trained to analyze a promotion probability feature set for the inpatient to determine the promotion probability of discharge of the inpatient relative to continued hospitalization; a user interface; and a processor. The processor is configured to: (i) receive patient data from a medical record database; (ii) extract a promotion probability feature set from the received patient data; (iii) analyze the promotion probability feature set by the trained promotion probability algorithm at a first time point to determine the promotion probability of discharge of the inpatient relative to continued hospitalization; (iv) compare the determined promotion probability with a predetermined promotion probability threshold to determine a recommendation, wherein when the determined promotion probability is lower than the predetermined promotion probability threshold, the recommendation is to keep the inpatient hospitalized, and wherein when the determined promotion probability is higher than the predetermined promotion probability threshold, the recommendation is to discharge the inpatient; and provide the recommendation to the user via the user interface.

[0018] It should be understood that all combinations of the foregoing concepts and the additional concepts discussed in more detail below (assuming such concepts are not mutually inconsistent) are considered part of the inventive subject matter disclosed herein. In particular, all combinations of the claimed subject matter appearing at the end of this disclosure are contemplated as part of the inventive subject matter disclosed herein. It should also be understood that terms explicitly adopted herein that may also appear in any disclosure incorporated by reference should be given the meaning most consistent with the specific concepts disclosed herein.

[0019] These and other aspects of the various embodiments will be apparent from and elucidated with reference to the embodiment(s) described hereinafter. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In the accompanying drawings, like reference numerals generally refer to the same parts throughout the different views. The accompanying drawings illustrate features and modes of implementing various embodiments and should not be construed as limiting other possible embodiments that fall within the scope of the appended claims. Moreover, the drawings are not necessarily drawn to scale, but emphasis is generally placed on illustrating the principles of the various embodiments.

[0021] Figure 1 is a flow chart of a method for determining a home care escalation probability according to an embodiment.

[0022] Figure 2 is a schematic representation of a home care boost probability system according to an embodiment.

[0023] Figure 3 is a flow chart of determining a time point for boosting probability according to an embodiment.

[0024] Figure 4 is a flow chart of a method for determining a home care escalation probability according to an embodiment.

[0025] Figure 5 is a flow chart of a method for training a boosting probability algorithm according to an embodiment.

[0026] Figure 6 is a graph of a time point for determining a lift probability according to an embodiment. DETAILED DESCRIPTION

[0027] The present disclosure describes various embodiments of systems and methods that are configured to generate and provide a lift probability for discharging an inpatient for home care. More generally, the applicant has recognized and understood that it would be beneficial to determine when to discharge an inpatient for managed home care. Therefore, a home care lift probability system receives patient data from a medical record database and extracts a lift probability feature set from the data. The system then analyzes the lift probability feature set by a trained lift probability algorithm to determine the lift probability of the inpatient's discharge relative to continued hospitalization. The system compares the determined lift probability with a predetermined lift probability threshold to determine a recommendation and provides the recommendation to the user. Note that in addition to discharging a patient for home care (such as managed home care), a clinician may also decide to transfer the patient to a skilled nursing facility or rehabilitation program, such as in the form of care virtualization through remote patient monitoring (RPM).

[0028] According to embodiments, in some non-limiting embodiments, the systems and methods described or otherwise contemplated herein may be implemented as commercial products for patient analysis or monitoring (such as The system and method may be implemented in an existing or future clinical decision support system (CDS), application, and device. However, the present disclosure is not limited to these devices or systems, and thus the disclosure and embodiments disclosed herein may encompass any device or system capable of generating and reporting a lift probability for a patient.

[0029] refer to Figure 1 , in one embodiment, is a flow chart of a method 100 for determining a home care promotion probability for a hospitalized patient using a home care promotion probability system 200. The methods described in conjunction with the figures are provided only as examples and should be understood not to limit the scope of the present disclosure. The home care promotion probability system may be any system described herein or otherwise contemplated. The home care promotion probability system may be a single system or a plurality of different systems.

[0030] At step 110 of the method, a home care promotion probability system 200 is provided. Figure 2 In the embodiment of the home care promotion probability system 200 depicted in FIG, for example, the system includes one or more of a processor 220, a memory 230, a user interface 240, a communication interface 250, and a storage device 260 interconnected via one or more system buses 212. It should be understood that Figure 2 Abstractions are formed in some aspects, and the actual organization of the components of the system 200 may be different and more complex than that illustrated. In addition, the home care promotion probability system 200 may be any system described or otherwise contemplated herein. Other elements and components of the home care promotion probability system 200 are disclosed and / or contemplated elsewhere herein.

[0031] At step 120 of the method, the home care promotion probability system receives or obtains patient data from the medical record database 270. Patient data can be any information about the patient that the home care promotion probability system can or may utilize to perform analysis as described herein or otherwise contemplated. According to an embodiment, the patient data includes one or more of the following: demographic information about the patient, diagnosis for the patient, the patient's medical history, and / or any other information. For example, demographic information can include information about the patient, such as name, age, body mass index (BMI), and any other demographic information. Diagnosis for the patient can be any information about the medical diagnosis (history and / or current) for the patient. The patient's medical history can be any historical admission or discharge information, historical treatment information, historical diagnosis information, historical examination or imaging information, and / or any other information.

[0032] Other non-limiting examples of patient data include: patient psycho-social variables regarding things such as social support, living situations, smoking, alcohol use, depression, anxiety; patient functional variables regarding things such as frailty, dementia, self-efficacy, and motivation; signs and symptoms of heart failure (HF), such as shortness of breath, fatigue, ankle, leg, and abdominal edema, exercise capacity, cough, and palpitations; the type and number of comorbidities (e.g., diabetes, COPD, etc.) that the patient has (e.g., as measured by the Charlson Comorbidity Index (CCI)); the patient's history of HF and other cardiac conditions, such as previous myocardial infarction (MI), and arrhythmias; medical utilization: number of previous hospitalizations or emergency department (ED) visits; preferably multiple The severity of heart failure assessed at the time of the study - NYHA class (New York Heart Association); echo or radiological assessment of the heart, such as ejection fraction level, left atrial volume index (LAV I), right atrial pressure (RAP), and pulmonary congestion; vital signs resulting from physical examination, such as blood pressure, heart rate, respiratory rate, weight, saturations, BMI; relevant blood tests, such as electrolytes (sodium, potassium), cholesterol, albumin, BUN, creatinine, and biomarkers, such as NTproBNP, ST2, hs-cTnT, and troponin; data on current hospitalization, such as days in the ICU, days in the ward, etc.; and drug therapy and dosages for diuretics, ARBs, ACE inhibitors, beta blockers, etc.

[0033] Patient data may be received or obtained from one or more different sources. According to an embodiment, patient data is received, retrieved, or otherwise obtained from an electronic medical record (EMR) database or system 270. The EMR database or system may be local or remote. The EMR database or system may be a component of the home care enhancement probability system, or may communicate locally and / or remotely with the home care enhancement probability system. The received patient data may be utilized immediately, or may be stored in a local or remote storage device for use in additional steps of the method.

[0034] Patient data may include patient information and medical records covering a period of time. For example, patient data may include patient information and medical records for the entire patient's hospital stay to date, a predetermined prior time, or any other time period. Patient data may be received or obtained continuously, or may be received or obtained periodically, such as once an hour, once a day, etc. Additionally or alternatively, patient data may be received or obtained in response to a triggering event (such as a patient care event, such as the availability of a new record, a transition to a new diagnosis or care routine, a move to a new location, etc.). Many other triggering events are possible.

[0035] At step 130 of the method, the home care lift probability system analyzes some or all of the received patient data to extract a lift probability feature set (optionally represented herein as X) for the patient. The lift probability feature set X may include any patient information that the home care lift probability system may or may utilize to perform an analysis as described herein or otherwise contemplated. The feature set X may include, for example, various EMR-based or hospital-based data elements captured at various moments from admission to discharge. Some data elements will change value when repeatedly captured over time, while other data elements will have constant values.

[0036] Each of the plurality of features in the lift probability feature set X can be identified or extracted using known methods for identifying and extracting features in patient data. According to an embodiment, the plurality of features used by the home care lift probability system for analysis can depend on the patient's diagnosis, treatment, demographics, and / or any other information specific to the patient. The parameters of the feature set X can be determined in part or in whole based on user input, or can be identified or determined in part or in whole based on received or obtained patient data. Once extracted, the feature set X can be utilized immediately, or can be stored in a local or remote storage device for use in further steps of the method.

[0037] At step 140 of the method, a trained lift probability algorithm of the home care lift probability system analyzes the lift probability feature set X at the first time point to determine the lift probability of discharge of the inpatient relative to continued hospitalization. The trained lift probability algorithm can be any algorithm or model configured to utilize the lift probability feature set X as input to generate the lift probability of discharge of the inpatient relative to continued hospitalization.

[0038] According to an embodiment, the trained lift probability algorithm analyzes the lift probability feature set X at any one or more of a plurality of different time points. For example, the generated lift probability of discharge of an inpatient relative to continued hospitalization, also known as a "benefit score," may be measured or determined at a number of inpatient moments from the time of admission to the actual discharge, such as Figure 3 This score, which can be provided as a service to clinical staff to recommend patient transitions, expresses the expected benefit (as a probability of improvement) of a patient being discharged with home care support versus continuing to be hospitalized. Figure 3 For example, there are initial family benefit score calculation (t1), discharge family benefit score calculation (t d ) and several intermediate household benefit score calculations (t2, t3, t4, t d-1 ). Many other combinations are possible.

[0039] According to an embodiment, during the patient's hospital stay from the date of admission to the end of the patient's hospital stay, the probability of promotion is calculated at a specific point (predetermined, random or in response to a triggering event) every day. The trend of this promotion will also provide information about the progression of the patient's health status. For example, is the patient becoming more stable over time? Or, how many days must the patient continue to stay in the hospital? This information can be provided to the clinician as described herein or otherwise contemplated.

[0040] refer to Figure 4 , in one embodiment, is a flow chart of a method 400 for generating and providing a clinician with an improved probability for discharge. According to an embodiment, after a patient is admitted to a hospital for acute heart failure (HF), at each evaluation time t n Calculate the patient's promotion probability P U . Assessment moments can be triggered by a transition to a different care setting (e.g., ED, ICU, ward) or by an update of an ordered measurement (e.g., lab test, vital sign) or medication or medical examination performed. Assessment moments can also follow a regular frequency (e.g., daily) independent of the entry of new patient level data into the EHR. Each assessment moment is a call to the CDS service and the data required for the calculation is obtained from the EHR. According to an embodiment, the probability of improvement is defined as the difference between the probability of success if the patient receives treatment (enrolled in the discharge RPM program) or does not receive treatment (control group - continued hospitalization). According to an embodiment, success can be measured for selected outcomes, such as avoidance of adverse events such as readmission or complications in acute care after discharge, as well as other selected or predetermined or identified outcomes.

[0041] According to an embodiment, the lift probability may be derived from a lift model defined using the following equation:

[0042] P U (X)∶=P T (Y=1|X)-P C (Y=1|X) Equation 1

[0043] Where Y=1 represents the desired (or positive, success) outcome of no adverse event (e.g., no readmission, no death), and Y=0 represents the corresponding classification of the undesirable (or negative, failure) outcome (e.g., readmission, death). T and P C denote the probability distribution of the outcome Y estimated on patients in the treatment (receiving the RPM procedure) and control groups, respectively; P C,T(Y=1|X) is the class probability of a positive outcome conditioned on the feature vector X. According to an embodiment, the boost model is constructed and validated in an initial trial or pilot phase that requires a limited number of patients to receive the RPM home care intervention after discharge from HF and compare them to a matched control group that did not receive the treatment - but remained in the hospital. The outcome variable is then evaluated after a defined period of time (e.g., 30 days, one year). According to another embodiment, an outcome can also be defined for a continuous variable L (e.g., number of non-hospital days survived after indexing admission) when its expected outcome value is greater than a predefined threshold M (e.g., the median of the sample).

[0044] Then the binary class variable Y of the boosting model can be obtained through the following discretization function:

[0045]

[0046] The lifting model can be constructed intuitively from two models, for example via logistic regression, with each estimating the probability P T and P C , and then subtract these to estimate the lift P U .

[0047] In another embodiment, the result variable can be transformed by the following function (see Table 1):

[0048]

[0049] And the lift probability can be calculated based on the following single model approach:

[0050] P U (X) = 2Pr(Z = 1|X) - 1 (Equation 4)

[0051] where Pr can be estimated, for example, by a logistic regression model with a transformed outcome variable Z. The model requires equal group sizes, however, this can always be achieved by reweighting the groups. The advantage of a single lift model approach is that it directly predicts lift, rather than differences as is the case with a dual model approach; a single model is also built from a larger sample size of both the treatment and control cohorts in the pilot phase. In dual model construction, the treatment model and the control model are built independently and focus on predicting outcomes separately. This separation may also result in having a different feature set X for each model. Each model can prioritize different features to have high predictive power to estimate the outcome, while ignoring features that best estimate lift across both models.

[0052] Table 1. Variable transformations.

[0053]

[0054] For patient p, we can evaluate i Calculate the probability of improvement Among them, the improvement model P U is obtained by using the current patient-specific feature vector x p,i To evaluate:

[0055]

[0056] Once in the evaluation interval [t1, ..., t n ] calculated a series of improvement probabilities for patients Choose an averaging function (e.g., moving average, weighted mean) to calculate the most recent evaluation time t n The average value A p,n If A p,n >θ (where θ≥0 is the threshold), the patient is recommended to be discharged and enrolled in the RPM home care program.

[0057] According to an embodiment, the value of θ may be determined based on resource constraints such as budget constraints, RPM staff availability, etc. In the case of θ>0, not all patients with a positive probability of promotion will be recommended for the RPM procedure, even though their expected benefit is positive. θ may take a fixed value defined at the beginning of the procedure, or may be a variable that changes continuously based on resource constraints. Methods for capacity management of RPM resource allocation based on manual determination or predictive modeling may be used to optimize thresholds and optimize capacity management based on available resources.

[0058] refer to Figure 5 , in one embodiment, is a flow chart of a method 500 for training a boost probability algorithm of a home care boost probability system 200. At step 510 of the method, the system receives or obtains a training data set including training data (such as historical patient data and information) for a plurality of patients. The training data may include inputs such as demographic information about the patient, a diagnosis for the patient, the patient's medical history, and / or any other information. For example, the demographic information may include information about the patient, such as name, age, body mass index (BMI), and any other demographic information. The diagnosis for the patient may be any information about a medical diagnosis (historical and / or current) for the patient. The patient's medical history may be any historical admission or discharge information, historical treatment information, historical diagnostic information, historical examination or imaging information, and / or any other information. The training data may be stored in one or more databases and / or received from one or more databases. The database may be a local and / or remote database. For example, the home care boost probability system may include a database of training data.

[0059] According to an embodiment, the home care promotion probability system may include a data preprocessor or similar component or algorithm configured to process the received training data. For example, the data preprocessor analyzes the training data to remove noise, bias, error and other potential problems. The data preprocessor can also analyze the input data to remove low-quality data. Many other forms of data preprocessing or data point identification and / or extraction are possible.

[0060] At step 520 of the method, the system trains a machine learning algorithm, which will be an algorithm for analyzing input information, as described or otherwise contemplated. The machine learning algorithm is trained using a training data set according to known methods for training machine learning algorithms. According to an embodiment, the algorithm is trained using a processed training data set to determine the increased probability of discharge relative to continued hospitalization of an inpatient using an increased probability feature set X, as described or otherwise contemplated herein.

[0061] At step 630 of the method, the trained boost probability algorithm of the home care boost probability system is stored for future use. Depending on the embodiment, the model may be stored in a local or remote storage device.

[0062] Return to Figure 1 In method 100 of the method, at step 150 of the method, the home care lift probability system compares the determined lift probability to a predetermined lift probability threshold to determine a recommendation. According to an embodiment, when the determined lift probability is below the predetermined lift probability threshold, the recommendation may be to keep the inpatient inpatient. According to an embodiment, when the determined lift probability is above the predetermined lift probability threshold, the recommendation may be to discharge the inpatient. Other recommendations are possible.

[0063] According to an embodiment, the threshold value may be determined by a clinician, or may be determined by the system or another system. For example, according to an embodiment, the value of the threshold value may be determined based on resource constraints such as budget constraints, availability of RPM staff. When the threshold value is greater than zero, as expected, not all patients with a positive probability of improvement will be recommended for the RPM procedure, even though their expected benefit is positive. The threshold value may be a fixed value defined at the beginning of the procedure, or may be a variable that changes continuously based on resource constraints. Methods for capacity management of RPM resource allocation based on manual determination or predictive modeling can be used to optimize the threshold value and optimize capacity management based on available resources.

[0064] According to an embodiment, the probability of promotion may be determined for a patient at more than one time point. Figure 3 In the non-limiting example shown in FIG. , there is an initial household benefit score calculation (t1), an emission household benefit score calculation (t d) and several intermediate household benefit score calculations (t2, t3, t4, t d-1 ). Many other combinations are possible.

[0065] According to an embodiment, for complex patients, the probability of promotion may not exceed a threshold at any time during the hospital stay. Since the patient cannot remain in the hospital indefinitely, in this case, a maximum length of stay may be included for which a clinical decision on a transition of care needs to be made. For example, rather than simply comparing the determined probability of promotion to a predetermined threshold and making a determination, the system may compare the determined probability of promotion to a predetermined threshold, and if the determined probability of promotion does not meet or exceed the threshold, the patient's current stay may be compared to a predetermined length of stay threshold, and if the current stay meets or exceeds the predetermined length of stay threshold, a recommendation may be provided.

[0066] At optional step 152 of the method, the system generates an average promotion probability based on two or more determined promotion probabilities. As described above, according to an embodiment, once the evaluation interval [t1, ..., t n ] calculated a series of lifting probabilities for the patient You can choose an average function (e.g., moving average, weighted mean) to calculate the most recent evaluation time t n The average value A p,n The system can then return to step 150 of the method and write the average value A p，n is compared with the threshold to generate a recommendation. p,n >θ (where θ ≥ 0), the patient is recommended to be discharged and enrolled in the RPM home care program. p,n <θ, it is recommended that the patient remain hospitalized.

[0067] At step 160 of the method, the determined recommendation and / or determined promotion probability for the patient is provided to the user via the user interface of the home care promotion probability system. The user can be any individual, although it will generally be a clinician involved in or responsible for the care of the patient. The determined recommendation and / or determined promotion probability can be provided to the user via any mechanism for displaying, visualizing or otherwise providing information via the user interface. According to an embodiment, information can be delivered to the user interface and / or another device by wired and / or wireless communication. For example, the system can deliver information to a mobile phone, a computer, a laptop computer, a wearable device and / or any other device configured to allow the display and / or other communications of the report. The user interface can be any device or system that allows to convey and / or receive information, and can include a display, a mouse and / or a keyboard for receiving user commands. According to an embodiment, the display can also include one or more of the following: the patient's name and / or treatment recommendation, and many other types of information.

[0068] According to embodiments, the determined recommendations and / or determined probability of improvement for the patient, optionally including other information, may be provided to the user via various other mechanisms, devices, and systems. For example, according to embodiments, the information may be provided via a clinical decision support (CDS) device, system, or application. The information may be provided in the form of a card representing the patient-specific recommendation, which is presented to the user within an app that can be launched from a CDS client (e.g., an EHR system). The information presented to the user may include the patient's n The probability of benefiting from enrollment in the nursing program is P U and its average function at various measurement moments, as well as the patient at time t d Recommendations for cases where discharge would benefit from enrollment in a nursing program, assuming the lift probability mean function curve exceeds a threshold level Q. Figure 6 , in one embodiment, is the time (t1, t2, ..., t d-1 ,t d ) on the curve graph, which is about the promotion probability P for the patient U (ie, the plotted data points) and the average lift probability, both of which are compared to a predetermined lift probability threshold level (θ). Figure 6 The example visualization in also shows the savings in time spent in an inpatient ward as the difference from the average or expected length of stay for that patient.

[0069] According to an embodiment, information may be provided to clinicians via a clinical decision support service app. For example, the system and method may be implemented as an AI-based RPM patient recommendation as a service solution that provides clinical decision support (CDS) to healthcare professionals or hospital staff who identify patients to be enrolled in a post-discharge care program. The system provides recommendations for patient transitions displayed via a CDS service app for clinicians. The CDS service app will then display a recommendation, such as "discharge the patient for an RPM home care program." The user can then accept, reject, or overrule the recommendation displayed by the CDS service app. When the app user confirms the decision, the patient may be scheduled to receive a transition service, which, depending on the confirmed decision, may invoke a service for discharging the patient to begin the onboarding process into a care management program that includes RPM services.

[0070] According to an embodiment, the CDS service app user may be a healthcare professional or staff member involved in reviewing a patient's readiness for discharge, or a patient flow coordinator at a clinical command center (CLOC) or hospital admission and discharge office. The CDS service app may be installed on a smartphone, tablet, laptop, etc. The app may also be launched from within a CDS client (such as an EHR system) as a CDS service request using interoperable standards (such as RESTful APIs) and interaction patterns with connected data systems, such as described by CDS hooks. CDS services may be initiated by an app user or activated by an automatic trigger, such as a new data entry that invokes an algorithm and produces an updated output.

[0071] At the optional step 170 of the method, the home care promotion probability system receives a decision about the recommendation provided from the clinic via a user interface. For example, a clinician or other decision maker uses the displayed recommendation to make a care decision for the patient. The decision can be an input provided to the system by the clinician via the user interface, such as a selection of the recommendation, a rejection of the recommendation, a request for more information, and / or any other input. The input can be provided via any mechanism for providing information via a user interface. The user interface can be a user interface of the system, a mobile computing device, or any other user interface, and the input can be transmitted to the home care promotion probability system via wired and / or wireless communication. Providing a decision or input about the recommendation can trigger a series of events that lead to the implementation of the recommendation, as described herein or otherwise contemplated.

[0072] exist Figure 1At optional step 180 of method 100 depicted in , the recommendation provided may be implemented, such as by a clinician and / or by a care system of a hospital or other care setting. For example, an implementation may include a prescription, an order, additional testing, and / or another implementation. According to an embodiment in which the recommendation is to keep an inpatient hospitalized when the determined probability of lift is below a predetermined probability of lift threshold, the implementation of the recommendation may be an order to keep the patient hospitalized. According to an embodiment in which the recommendation is to discharge an inpatient when the determined probability of lift is above a predetermined probability of lift threshold, the implementation of the recommendation may be an order to discharge the patient or to prepare the patient for discharge. Many other implementations are possible.

[0073] refer to Figure 2 is a schematic representation of a home care enhancement probability system 200. The system 200 may be any system described or otherwise contemplated herein and may include any components described or otherwise contemplated herein. It should be understood that Figure 2 An abstraction is constituted in some aspects, and the actual organization of the components of system 200 may be different and more complex than illustrated.

[0074] According to an embodiment, the system 200 includes a processor 220 capable of running instructions stored in the memory 230 or the storage device 260 or otherwise processing data to, for example, perform one or more steps of the method. The processor 220 may be formed by one or more modules. The processor 220 may take any suitable form, including but not limited to a microprocessor, a microcontroller, multiple microcontrollers, a circuit, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a single processor, or multiple processors.

[0075] Memory 230 may take any suitable form, including non-volatile memory and / or RAM. Memory 230 may include various memories, such as L1, L2 or L3 cache or system memory. In this way, memory 230 may include static random access memory (SRAM), dynamic RAM (DRAM), flash memory, read-only memory (ROM) or other similar memory devices. The memory may store an operating system, etc. RAM is used by the processor for temporary storage of data. According to an embodiment, the operating system may include code that controls the operation of one or more components of system 200 when run by the processor. It will be apparent that in an embodiment where the processor implements one or more functions described herein in hardware, software described as corresponding to such functions in other embodiments may be omitted.

[0076] The user interface 240 may include one or more devices for realizing communication with the user. The user interface may be any device or system that allows to convey and / or receive information, and may include a display, a mouse and / or a keyboard for receiving user commands. In certain embodiments, the user interface 240 may include a command line interface or a graphical user interface that may be presented to a remote terminal via a communication interface 250. The user interface may be located together with one or more other components of the system, or may be located away from the system and communicated via a wired and / or wireless communication network.

[0077] The communication interface 250 may include one or more devices for implementing communication with other hardware devices. For example, the communication interface 250 may include a network interface card (NIC) configured to communicate according to the Ethernet protocol. In addition, the communication interface 250 may implement a TCP / IP stack for communicating according to the TCP / IP protocol. Various alternative or additional hardware or configurations of the communication interface 250 will be apparent.

[0078] The storage device 260 may include one or more machine-readable storage media, such as a read-only memory (ROM), a random access memory (RAM), a magnetic disk storage medium, an optical storage medium, a flash memory device, or a similar storage medium. In various embodiments, the storage device 260 may store instructions for execution by the processor 220 or data that the processor 220 may operate on. For example, the storage device 260 may store an operating system 261 for controlling various operations of the system 200.

[0079] It will be apparent that various information described as being stored in storage device 260 may additionally or alternatively be stored in memory 230. In this regard, memory 230 may also be considered to constitute a storage device, and storage device 260 may be considered to be a memory. Various other arrangements will be apparent. In addition, both memory 230 and storage device 260 may be considered to be non-transitory machine-readable media. As used herein, the term non-transitory will be understood to exclude transient signals, but include all forms of storage devices, including both volatile memory and non-volatile memory.

[0080] Although system 200 is shown as including one of each described component, various components may be replicated in various embodiments. For example, processor 220 may include multiple microprocessors configured to independently perform the methods described herein, or configured to perform the steps or subroutines of the methods described herein, so that multiple processors collaborate to implement the functions described herein. In addition, in the case where one or more components of system 200 are implemented in a cloud computing system, various hardware components may belong to separate physical systems. For example, processor 220 may include a first processor in a first server and a second processor in a second server. Many other variations and configurations are possible.

[0081] According to an embodiment, the electronic medical record system 270 is an electronic medical record database from which information about a patient, including patient data and / or training data, can be obtained or received. The electronic medical record database can be a local or remote database and communicate directly and / or indirectly with the home care improvement probability system 200. Therefore, according to an embodiment, the home care improvement probability system includes an electronic medical record database or system 270.

[0082] According to an embodiment, the storage device 260 of the system 200 may store one or more algorithms, modules and / or instructions to perform one or more functions or steps of the methods described or otherwise contemplated herein. For example, the system may include feature extraction instructions 262, trained boost probability algorithms 263, and / or reporting instructions 264, as well as other instructions or data.

[0083] According to an embodiment, feature extraction instructions 262 guide the system to analyze some or all of the received patient data to extract the patient's promotion probability feature set X. The promotion probability feature set X may include any patient information that the home care promotion probability system may or may utilize to perform analysis as described herein or otherwise contemplated. Feature set X may include, for example, various EMR-based or hospital-based data elements captured at various moments from admission to discharge. Some data elements will change values ​​when they are repeatedly captured over time, while other data elements will have constant values. Each of the multiple features in the promotion probability feature set X can be identified or extracted using known methods for identifying and extracting features in patient data. According to an embodiment, the multiple features used by the home care promotion probability system for analysis may depend on the patient's diagnosis, treatment, demographics, and / or any other information specific to the patient. The parameters of feature set X may be determined in part or in whole based on user input, or may be identified or determined in part or in whole based on received or obtained patient data. Once extracted, feature set X may be immediately utilized, or may be stored in a local or remote storage device for additional steps of the method.

[0084] According to an embodiment, the trained lift probability algorithm 263 is trained to analyze the lift probability feature set X to determine the lift probability of discharge of an inpatient relative to continued hospitalization. The trained lift probability algorithm can be any algorithm or model configured to use the lift probability feature set X as input to generate the lift probability of discharge of an inpatient relative to continued hospitalization. According to an embodiment, the trained lift probability algorithm analyzes the lift probability feature set X at any one or more of a plurality of different time points. Figure 4 is a method for determining a boost probability of discharge of an inpatient using a trained boost probability algorithm 263, and reference is made to Figure 5 It is a method used to train the boost probability algorithm 263.

[0085] According to an embodiment, the report instruction 264 guides the system to generate the determined recommendation and / or the determined probability of promotion for the patient and provides it to the user via the user interface. The user can be any individual, although it is usually a clinician who participates in or is responsible for patient care. The determined recommendation and / or the determined probability of promotion can be provided to the user via any mechanism for displaying, visualizing or otherwise providing information via the user interface. According to an embodiment, information can be delivered to the user interface and / or another device by wired and / or wireless communication. For example, the system can deliver information to a mobile phone, a computer, a laptop computer, a wearable device and / or any other device configured to allow the display and / or other communications of the report. The user interface can be any device or system that allows to convey and / or receive information, and can include a display, a mouse and / or a keyboard for receiving user commands. According to an embodiment, the display can also include one or more of the following: the patient's name and / or treatment recommendation, and many other types of information.

[0086] According to an embodiment, the determined recommendation and / or determined probability of promotion for the patient can be provided to the user via various other mechanisms, devices and systems, optionally including other information. For example, according to an embodiment, information can be provided via a clinical decision support (CDS) device, system or application. The information can be provided in the form of a card representing a patient-specific recommendation, which is presented to the user in an app that can be started from a CDS client (e.g., an EHR system). According to an embodiment, information can be provided to a clinician via a clinical decision support service app. For example, the system and method can be implemented as an AI-based RPM patient recommendation, which, as a service solution, provides clinical decision support (CDS) for healthcare professionals or hospital staff who identify patients to be registered in a post-discharge care program. The system provides recommendations for patient transitions displayed via a CDS service app for clinicians. Then, the CDS service app program will display recommendations, such as "discharge the patient for an RPM home care program". Then, the user can accept, reject or overrule the recommendations displayed by the CDS service app. When the app user confirms the decision, the patient can be scheduled to receive the transition service, which, depending on the confirmed decision, can call the service for discharging the patient to begin the onboarding process into the care management program including the RPM service. According to an embodiment, the CDS service app user can be a healthcare professional or staff member involved in reviewing the patient's discharge readiness, or a patient flow coordinator in a clinical command center (CLOC) or hospital admission and discharge office. The CDS service app can be installed on a smartphone, tablet computer, laptop computer, etc. The app can also be launched from within a CDS client (such as an EHR system) as a CDS service request using interoperable standards (such as RESTful APIs) and interaction modes with connected data systems, such as described by CDS hooks. The CDS service can be initiated by the app user or activated by an automatic trigger, such as calling an algorithm and generating a new data entry with updated output.

[0087] Thus, in the context of the disclosure herein, aspects of the embodiments may take the form of a computer program product embodied in one or more non-transitory computer-readable media having computer-readable program code embodied thereon. Thus, according to one embodiment is a non-transitory computer-readable storage medium comprising computer program code instructions, which when executed by a processor enable the processor to perform a method comprising: (i) receiving patient data from a medical record database; (ii) extracting a lift probability feature set from the received patient data; (iii) analyzing the lift probability feature set by a trained lift probability algorithm at a first point in time to determine a lift probability of discharge of the inpatient relative to continued hospitalization; (iv) comparing the determined lift probability with a predetermined lift probability threshold to determine a recommendation, wherein when the determined lift probability is below the predetermined lift probability threshold, the recommendation is to keep the inpatient hospitalized, and wherein when the determined lift probability is above the predetermined lift probability threshold, the recommendation is to discharge the inpatient; and (v) providing the recommendation to a user via a user interface. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server.

[0088] According to an embodiment, the home care promotion probability system is configured to process thousands or millions of data points in the input data for training the system, and to process and analyze the received patient data. For example, the use of automated processes such as feature recognition and extraction and subsequent training to generate a functional and skilled trained system requires processing millions of data points from the input data and the generated features. This may require millions or billions of calculations to generate a novel trained system from these millions of data points and millions or billions of calculations. Therefore, the trained system is novel and different based on the input data and parameters of the machine learning algorithm, and thus improves the functionality of the home care promotion probability system. The generation of functional and skilled trained systems includes a large number of calculations and analysis processes that the human brain cannot complete in a lifetime or multiple lifetimes. By providing improved patient analysis, this novel home care promotion probability system has a huge positive impact on patient analysis and care compared to prior art systems.

[0089] There are no tools available to physicians that provide a continuous assessment of risk and enable them to make the correct treatment choice at the earliest possible moment. Currently, physicians assess the clinical status of patients in a non-continuous manner and make care decisions based on that assessment. They also do not have quantitative information on the benefits of different care models for patients. The methods and systems described or otherwise contemplated herein provide such a system and the necessary quantitative information.

[0090] Among other improvements, the home care boost probability system allows hospitals to gain efficiencies in healthcare inpatient resource utilization by enabling patients to be discharged earlier with ongoing coordinated care at home. Expensive inpatient days are replaced with days spent in lower acuity settings. Thus, the methods and systems described herein reduce the average length of stay for the entire patient population.

[0091] The method and system also improves the quality of life and patient satisfaction for patients and relatives by experiencing more care from the comfort of home. The risk of hospital-acquired complications for elderly patients is reduced. For patients with acute heart failure, the risk of experiencing the side effects of hospitalization treatment is reduced. These complications often prolong hospital stays or even lead to in-hospital deaths.

[0092] The method and system are optimized for the desired outcomes of readmission-free and death-free or number of non-hospital days alive after the index admission to reduce the associated risk. Any remaining risk of readmission or death is mitigated by providing a safe patient discharge moment, adequate patient education, and post-discharge follow-up for home care.

[0093] It should be understood that all combinations of the foregoing concepts and the additional concepts discussed in more detail below (assuming such concepts are not mutually inconsistent) are considered part of the inventive subject matter disclosed herein. In particular, all combinations of the claimed subject matter appearing at the end of this disclosure are contemplated as part of the inventive subject matter disclosed herein. It should also be understood that terms explicitly adopted herein that may also appear in any disclosure incorporated by reference should be given the meaning most consistent with the specific concepts disclosed herein.

[0094] All definitions, as defined and used herein, should be understood to control over dictionary definitions, definitions in documents incorporated by reference, and / or ordinary meanings of the defined terms.

[0095] Unless expressly indicated to the contrary, the terms "a" and "an" as used in the specification and claims herein should be understood to mean "at least one."

[0096] As used in the specification and claims herein, the phrase "and / or" should be understood to mean "either or both" of the elements so combined, i.e., elements that are present in combination in some cases and separately in other cases. Multiple elements listed with "and / or" should be interpreted in the same manner, i.e., "one or more" of the elements so combined. In addition to the elements specifically identified by the "and / or" clause, other elements may optionally be present, whether related or unrelated to those elements specifically identified.

[0097] The phrase "at least one" as used in the specification and claims herein referring to a list of one or more elements should be understood to mean at least one element selected from any one or more elements in the list of elements, but not necessarily including at least one element of each and every element specifically listed in the list of elements, and not excluding any combination of elements in the list of elements. This definition also allows that elements may optionally be present in addition to the elements specifically identified in the list of elements to which the phrase "at least one" refers, whether related or unrelated to those elements specifically identified.

[0098] As used herein, although the terms first, second, third, etc. may be used to describe various elements or components in this article, these elements or components should not be limited by these terms. These terms are only used to distinguish one element or component from another element or component. Therefore, without departing from the teaching of the present invention, the first element or component discussed below may be referred to as the second element or component.

[0099] Unless otherwise specified, when an element or component is referred to as being "connected to," "coupled to," or "adjacent to" another element or component, it should be understood that the element or component may be directly connected or coupled to the other element or component, or there may be intervening elements or components. That is, these terms and similar terms cover situations where one or more intervening elements or components may be used to connect two elements or components. However, when an element or component is referred to as being "directly connected" to another element or component, this only covers situations where two elements or components are connected to each other without any intervening or intervening elements or components.

[0100] In the claims and the preceding description, all transitional phrases, such as "comprises," "comprising," "carrying," "having," "containing," "involving," "holding," "including," etc., should be understood as open-ended, i.e., meaning including but not limited to. Only the transitional phrases "consisting of" and "consisting essentially of" should be closed or semi-closed transitional phrases, respectively.

[0101] It should also be understood that in any method claimed herein that includes more than one step or action, the order of the steps or actions of the method is not necessarily limited to the order in which the steps or actions of the method are listed unless explicitly stated to the contrary.

[0102] The above examples of the described subject matter can be implemented in any of a variety of ways. For example, some aspects can be implemented using hardware, software, or a combination thereof. When any aspect is implemented at least in part in software, the software code can be run on any suitable processor or set of processors, whether provided in a single device or computer or distributed among multiple devices / computers.

[0103] The present disclosure may be implemented as a system, method and / or computer program product at any possible level of technical detail integration. The computer program product may include a computer-readable storage medium (or media) having computer-readable program instructions thereon for causing a processor to execute various aspects of the present disclosure.

[0104] Computer readable storage medium can be a tangible device that can maintain and store instructions for use by an instruction execution device. Computer readable storage medium can be, for example, but not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. A non-exhaustive list of more specific examples of computer readable storage medium includes the following: a portable computer floppy disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disc (DVD), a memory stick, a floppy disk, a mechanical encoding device (such as a punch card or a pull-up structure in a groove with instructions recorded thereon) and any suitable combination of the foregoing. As used herein, a computer readable storage medium will not be interpreted as a transient signal itself, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagated by a waveguide or other transmission medium (e.g., a light pulse passing through an optical fiber cable) or an electrical signal transmitted by wiring.

[0105] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to a corresponding computing / processing device or downloaded to an external computer or external storage device via a network (e.g., the Internet, a local area network, a wide area network, and / or a wireless network). The network can include copper transmission cables, optical transmission fibers, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in the corresponding computing / processing device.

[0106] The computer-readable program instructions for performing the operations of the present disclosure may be assembler instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, configuration data for an integrated circuit, or object code written in either source code or in any combination of the following: one or more programming languages ​​(including object-oriented programming languages ​​(such as Smalltalk, C++, etc.) and procedural programming languages ​​(such as "C" programming language or similar programming languages)). The computer-readable program instructions may be executed entirely on a user's computer, partially on a user's computer, partially on a user's computer and partially on a remote computer, or entirely on a remote computer or server as a stand-alone software package. In the latter scenario, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or a connection may be made to an external computer (e.g., via the Internet using an Internet service provider). In some examples, electronic circuits including, for example, programmable logic circuits, field programmable gate arrays (FPGAs), or programmable logic arrays (PLAs) may execute computer-readable program instructions by utilizing state information of the computer-readable program instructions to personalize the electronic circuits so as to perform various aspects of the present disclosure.

[0107] Various aspects of the present disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems) and computer program products according to examples of the present disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams and combinations of blocks in the flowchart illustrations and / or block diagrams can be implemented by computer-readable program instructions.

[0108] Computer-readable program instructions may be provided to a processor of a special-purpose computer or other programmable data processing device to produce a machine, such that instructions executed by the processor of the computer or other programmable data processing device create a module for implementing the functions / actions specified in the flowchart and / or one or more block diagram blocks. These computer-readable program instructions may also be stored in a computer-readable storage medium, which may direct a computer, a programmable data processing device, and / or other equipment to operate in a particular manner, such that a computer-readable storage medium having instructions stored therein includes an article of manufacture containing instructions for implementing aspects of the functions / actions specified in the flowchart and / or block diagram or block.

[0109] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operating steps to be executed on the computer, other programmable apparatus, or other device to produce a computer-implemented process, so that the instructions executed on the computer, other programmable apparatus, or other device implement the functions / actions specified in the flowchart and / or one or more block diagram blocks.

[0110] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functions and operations of possible implementations of the systems, methods and computer program products according to various examples of the present disclosure. In this respect, each frame in the flowchart or block diagram can represent a module, segment or part of an instruction, which includes one or more executable instructions for implementing (one or more) specified logical functions. In some alternative embodiments, the functions indicated in the frame can occur out of the order indicated in the accompanying drawings. For example, the two frames shown in succession can actually be executed substantially simultaneously, or the frames can sometimes be executed in reverse order, depending on the functionality involved. It will also be noted that each frame and the combination of frames in the frame diagram and / or the flow chart diagram can be implemented by a system based on dedicated hardware that performs a specified function or action or performs a combination of dedicated hardware and computer instructions.

[0111] Other implementations are within the scope of the following claims and other claims to which the applicant may base his claims.

[0112] Although several inventive embodiments have been described and illustrated herein, it will be readily apparent to those of ordinary skill in the art that various other modules and / or structures for performing functions and / or obtaining results and / or one or more of the advantages described herein will be considered to be within the scope of the inventive embodiments described herein, and each of such variations and / or modifications is considered to be within the scope of the inventive embodiments described herein. More generally, it will be readily understood by those skilled in the art that all parameters, dimensions, materials, and configurations described herein are intended to be exemplary, and that actual parameters, dimensions, materials, and / or configurations will depend on one or more specific applications using the invention teachings. Those skilled in the art will recognize or be able to determine many equivalents of the specific inventive embodiments described herein using experimental means that do not exceed conventional means. Therefore, it should be understood that the foregoing embodiments are presented only by way of example, and within the scope of the claims and their equivalents, the inventive embodiments may be practiced in a manner different from that specifically described and claimed. The inventive embodiments of the present disclosure relate to each individual feature, system, article, material, kit, and / or method described herein. In addition, if such features, systems, articles, materials, kits, and / or methods do not contradict each other, any combination of two or more such features, systems, articles, materials, kits, and / or methods is included within the scope of the invention of the present disclosure.

Claims

1. A method (100) for determining a probability of home care promotion for a hospitalized patient, comprising: receiving (120) patient data from a medical records database; extracting (130) a set of boost probability features from the received patient data; analyzing (140) the lift probability feature set at a first time point using a trained lift probability algorithm (263) to determine a lift probability of discharge relative to continued hospitalization of the inpatient; comparing the determined probability of promotion to a predetermined probability of promotion threshold (150) to determine a recommendation, wherein the recommendation is to keep the inpatient hospitalized when the determined probability of promotion is below the predetermined probability of promotion threshold, and wherein the recommendation is to discharge the inpatient hospitalized when the determined probability of promotion is above the predetermined probability of promotion threshold; The recommendation is provided (160) to a user via a user interface.

2. The method according to claim 1, further comprising: repeating at least the analyzing and comparing steps for a second time point; and An average boost probability is generated (152) based on the determined boost probabilities.

3. The method according to claim 1, wherein: Providing also includes providing, via the user interface, one or more of: patient information and the determined probability of escalation to the user.

4. The method according to claim 1, wherein: The first time point is triggered by a care event for the inpatient.

5. The method according to claim 1, wherein: The first time point is a predetermined time point at a specific time during the hospitalization of the inpatient.

6. The method according to claim 1, wherein: The recommendation is provided via a user interface of the clinical decision support system.

7. The method according to claim 1, wherein: The recommendation is provided via a user interface of the mobile device.

8. The method according to claim 1, further comprising the steps of: receiving (170) from a clinician via the user interface a decision regarding the provided recommendation; and The provided recommendation is implemented (180) by the clinician.

9. The method according to claim 8, wherein: The recommendation provided is a recommendation to discharge the hospitalized patient.

10. A system (200) for determining a probability of home care promotion for a hospitalized patient, comprising: a trained promotion probability algorithm (263) trained to analyze a promotion probability feature set for the hospitalized patient to determine a promotion probability of discharge relative to continued hospitalization for the hospitalized patient; User interface (240); as well as A processor (220) configured to: (i) receive patient data from a medical records database (270); (ii) extracting a set of features that enhance probability from the received patient data; (iii) analyzing the lift probability feature set at a first time point by a trained lift probability algorithm to determine a lift probability of discharge of the inpatient relative to continued hospitalization; (iv) comparing the determined lift probability with a predetermined lift probability threshold to determine a recommendation, wherein when the determined lift probability is below the predetermined lift probability threshold, the recommendation is to keep the inpatient hospitalized, and wherein when the determined lift probability is above the predetermined lift probability threshold, the recommendation is to discharge the inpatient; and providing the recommendation to a user via the user interface.

11. The system of claim 10, further comprising a medical records database (270) comprising the patient data.

12. The system according to claim 10, wherein: The processor is further configured to repeat at least the analyzing and comparing steps for a second time point; and is further configured to generate an average boost probability based on the determined boost probabilities.

13. The system according to claim 10, wherein: Providing also includes providing, via the user interface, one or more of: patient information and the determined probability of escalation to the user.

14. The system according to claim 10, wherein: The recommendation is provided via a user interface of the clinical decision support system.

15. The system according to claim 10, wherein: The recommendation is provided via a user interface of the mobile device.

Images