Discharge planning is a difficult process for physicians and hospital professionals. Discharge planning may be especially complicated for patients suffering from certain diseases and/or conditions. For example, managing a patient suffering from acute decompensated heart failure (ADHF) can be complex because of the different etiology and many co-morbidities such as renal dysfunction, COPD, hypertension, diabetes, sleep apnea, etc. Discharge planning is further complicated by the fact that there is currently no objective measurement for determining whether a patient is ready to be discharged from the hospital. A patient that is discharged too early may experience inadequate symptom relief and may require readmission to the hospital, resulting in increased costs. Unmet patient needs are not systematically identified prior to a discharge decisions and are thus not proactively addressed. In addition, current discharge planning tools cannot predict a patient's readiness for discharge based on a particular treatment or treatment modification. Thus, it is impossible to estimate factors such as a patient's currently projected length of stay and the potential for a reduction, risk for readmission and total medical costs, which makes it difficult for the hospital to prepare and plan accordingly.
A method of patient discharge planning including evaluating a patient record including patient data parameters of a patient, predicting a change in the patient record for all possible treatment options, generating a discharge recommendation based on at least one of the patient record and the predicted change in the patient record; and displaying the discharge recommendation to a user.
A system for discharge planning having a memory storing a patient record including patient data parameters for a patient and a population database including patient data for all patients. The system further includes a processor evaluating the patient record, predicting a change in the patient record and generating a discharge recommendation based on at least one of the patient record and the predicted change in the patient record and a display displaying the discharge recommendation.
A non-transitory computer-readable storage medium including a set of instructions executable by a processor. The set of instructions operable to evaluate a patient record including patient data parameters of a patient, predict a change in the patient record for all possible treatment options, generate a discharge recommendation indicating whether the patient is ready for discharge with respect to the patient record and display the discharge recommendation to a user.
The exemplary embodiments may be further understood with reference to the following description and the appended drawings wherein like elements are referred to with the same reference numerals. The exemplary embodiments relate to a system and method for predictive discharge planning for a patient that has been admitted to the hospital. In particular, the exemplary embodiments provide a system and method for generating recommendations regarding whether a patient should be discharged and whether a patient's current treatment plan should be modified. The system and methods of the exemplary embodiments may also predict other variable such as a patient's currently projected length of stay and the potential for a reduction, a risk-of-readmission index and total costs associated with the patient's care so that the patient's discharge may be planned and optimized by taking multiple factors into consideration. Although the exemplary embodiments are specifically described in regard to a patient having acute decompensated heart failure (ADHF), it will be understood by those of skill in the art that the system and method of the present invention may be used for patients having any of a variety of diseases or conditions such as renal dysfunction, COPD and other chronic conditions.
As shown in
The patient record 110 includes patient data such as patient identification (e.g., name, age, gender), factors associated with biophysical health (e.g., reason for admission, vitals, test results, medical history and co-morbidities), factors associated with mental health, factors associated with daily living and factors associated with personal, community and healthcare environments.
The set of discharge criteria 120 includes criteria that are used to assess whether a patient is ready for discharge. The discharge criteria may be specific to the patient's disease or condition. For example, the discharge criteria for a patient suffering from ADHF includes criteria such as whether exacerbating factors have been addressed, achievement of near-optimal pharmacological therapy (or at least successful initiation of pharmacological therapy and plan for up-titration), stability of oral medication regimen, etc.
If no identified patient data parameters are missing, the method 200 skips steps 240 and 250, moving directly to a step 260. In the step 260, baseline and cut-off values for evaluation flags are provided. The evaluation flags are used to determine whether each of the identified patient data parameters fall within a normal (e.g., clinically acceptable rather than a normal distribution), close-to-normal (e.g., borderline) or abnormal (e.g., clinically unacceptable) range. As shown in
In a step 270, the evaluation manager 114 calculates a flag for each of the identified patient data parameters using the baseline and cut-off values provided in the step 260. The evaluation manager 114 determines whether values of each of the identified patient data parameters falls within the normal, close-to-normal or abnormal range on a given day. Since values of the identified parameters are available for current and previous days, flags are assigned for each of the available days. Flags may also be similarly predicted for future days based on predicted patient data, as will be further described below in regard to the method 400 described with reference to
The evaluation manager 114 is also used to evaluate whether the patient record 110 satisfies the discharge criteria 120 according to a method 300, as shown in
Once the necessary patient data has been identified, the evaluation manager 114, in a step 320, generates a discharge criteria score for each of the discharge criteria in the set of discharge criteria 120 on a given day using a discharge criteria evaluation algorithm. The discharge criteria evaluation algorithm evaluates the flag, as calculated in the step 270 using the method 200 described above, for each of the corresponding patient data parameters of the discharge criteria to determine the discharge criteria score. The discharge criteria score may indicate whether each of the discharge criteria is considered satisfied, somewhat satisfied or unsatisfied. Similarly to the evaluation flags described above in regard to the method 200, the satisfied discharge criteria may be represented by a green color (or a full pie-chart), the somewhat satisfied criteria may be represented by a yellow color (or a partially-filled pie chart) and the unsatisfied criteria may be represented by a red color (or an empty pie-chart). It will be understood by those of skill in the art that the discharge criteria may be displayed using other scoring methods besides the green, yellow and red color codes. For example, the scores may be represented using any predetermined color code, graphical representation, using descriptive terms such as “satisfied”, “somewhat satisfied” and “not satisfied,” numerical values, which may fall within defined ranges indicating a level of satisfaction, or any combination thereof. In an alternative embodiment, only the current value and the recent trend would be displayed using, for example, up, sideways and down arrows, instead of the history of scores.
The discharge criteria evaluation function may be defined as shown in
In a step 330, the individual discharge criteria scores are used to generate a discharge score indicating whether the patient is ready to be discharged. The discharge score indicates a patient response to treatment and a level of readiness to be discharged. As shown in
It will be understood by those of skill in the art that similarly to the discharge criteria scores, the discharge score may be indicated using any of a variety of display methods such as, for example, color codes, graphical representations, descriptive terms, numerical values falling within defined ranges of discharge readiness or any combination thereof. The discharge criteria scores generated in step 320 and the discharge score generated in step 330 for each of the previous and current days are displayed on the display 106, in a step 340, as shown in
As shown in
In a step 430, the predictions manager 116 uses a prediction model, which considers both the calculated change under the current treatment along with treatment results stored in the population database 112 to predict future changes in each patient parameter for any particular treatment. Thus, the predictions for any particular treatment may be based on both the current treatment of the patient and other treatments based on treatment results from the population database 112. The predictions model is based on techniques for extracting patterns from the population database 112 such as, for example, multi-vector, machine learning or cluster analysis. The predictions model can also be extended to predict a readmission probability index along with a mortality probability index and/or the Charlson co-morbidity index for each of the calculated and predicted changes of the patient data parameter based on the population database 112, in a step 440. As shown in
As shown in
Where the current discharge score is determined to be satisfied, the method 500 proceeds to a step 540, in which the decisions manager 118 recommends that the patient be discharged. The recommendation may, for example, be displayed on the display 106 as “Ready to Discharge Now.” As will be understood by those of skill in the art, however, the readiness for discharge may be indicated to the user in any of a variety of ways so long as it clear to the user that the decisions manager 118 recommends that the patient be discharged, i.e., the patient has been stabilized under the current treatment. Where the current discharge score is not satisfactory in the step 530, the method 500 proceeds to a step 550, in which the decisions manager 118 evaluates whether modifications in the current treatment could potentially increase the patient's readiness for discharge. The treatment evaluation may be following a treatment evaluation method 600, as will be described in greater detail below in reference to
In a step 560, the processor 102 determines whether a treatment modification has been made based on the treatment evaluation of step 550. If a treatment modification has not been made, the patient should remain in the hospital under the current treatment for further observation and evaluation. Thus, in a step 570, the decisions manager will recommend that the patient is not ready to be discharged. This discharge recommendation may be displayed on the display 106 as “Not Ready for Discharge.” As will be understood by those of skill in the art, however, the recommendation may be indicated in any of a variety of ways so long as it is clear to the user that the decisions manager 118 recommends that the patient not be discharged. If it is determined in the step 550 that a treatment modification has been made, the method 500 proceeds from the step 560 to a step 580, in which the processor 102 determines whether the modified treatment includes an out-patient component. Where the modified treatment is determined to include an out-patient component, the decisions manager 118 may recommend that the patient be discharged with the out-patient treatment, in a step 590. Where the modified treatment does not include an out-patient component, the method 500 reverts to the step 570, recommending that the patient not be discharged. It will be understood by those of skill in the art that where the decisions manager 118 does not recommend that the patient be discharged, the method 500 may revert back to the step 510 such that any new patient data will be re-evaluated to determine the patient's readiness for discharge.
As described above, if it is determined that the discharge score did not qualify for a recommendation of discharge (e.g., where the discharge score is not green), the method 500 may evaluate whether a treatment should be changed, using the method 600. As shown in
In the step 620, the decisions manager 118 generates a list of possible in-hospital treatment options, as shown in
Based on these predicted values a number of additional variables are also calculated. For example, the method 600 calculates variables such as predicted days until discharge (D2D), length of stay (LoS), readmission probability index (RIndex) and total medical cost (Total Cost), as shown in
D2D=(First Day DScorepre=green)−(Current Day); 1)
Length-of-Stay (LoS)=Current Day+D2D; 2)
Readmission probability Index (Rlndex)=30-days post-discharge risk of re-admission calculated by the Predictions Manager; 3)
and
Total Medical Cost=ΣCost(Tx@Day dk), k=1, . . . , LoS. 4)
These variables are well-established outcomes that can be used to guide the treatment decisions, as described in a step 650. These variables also aid in hospital resource planning. For example, a predicted length of stay permits the hospital to predict bed availability, availability of physicians and nurses on the medical ward during day/night shifts, patients schedule of the discharge planner nurse who will prepare the patient for discharge, etc. These variables are also used to plan for out-of-hospital resources such as availability of out-patient services, telehealth services, long term condition care provided by a community nurse, palliative care, etc. Although the exemplary embodiment describes specific variable above, it will be understood by those of skill in the art that the method 600 may also include the prediction and/or calculation of other desired variables.
In the step 650, the decisions manager 118 generates a treatment recommendation that optimizes a selected outcome or a combination thereof. The decisions manager 118 may recommend a treatment based upon predetermined recommendation requirements such as, for example, guideline-conforming care, a minimum predicted length of stay, a minimum rate of readmission an/or a reduced total cost. The treatment decision recommendations may be, for example, to keep the current treatment (e.g., “Keep CurTx”), modify the current treatment to include an out-hospital treatment (e.g., “Consider Modifying CurTx into In-Out Tx2”)or modify the current treatment to a different in-hospital treatment (e.g., “Consider Modifying CurTx into InTx1”). It will be understood by those of skill in the art that these recommendations may be displayed on the display 106 as described above or in any of a variety of ways so long as the recommended treatment option is made clear to the user. The treatment decision recommendation may also include treatment adaptations actions that may be displayed as an alert to the user. The alerts may include, for example, suggestions for medication changes, new lab orders, scheduling follow-up visits, planning home visits, etc.
It is noted that the claims may include reference signs/numerals in accordance with PCT Rule 6.2(b). However, the present claims should not be considered to be limited to the exemplary embodiments corresponding to the reference signs/numerals.
Those skilled in the art will understand that the above-described exemplary embodiments may be implemented in any number of manners, including, as a separate software module, as a combination of hardware and software, etc. For example, the evaluation manager 114, the predictions manager 116 and the decisions manager 118 may be a program containing lines of code that, when compiled, may be executed on a processor.
It will be apparent to those skilled in the art that various modifications may be made to the disclosed exemplary embodiments and methods and alternatives without departing from the spirit or scope of the disclosure. Thus, it is intended that the present disclosure cover modifications and variations provided that they come within the scope of the appended claims and their equivalents.
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/IB12/50474 | 2/1/2012 | WO | 00 | 8/5/2013 |
Number | Date | Country | |
---|---|---|---|
61439586 | Feb 2011 | US |