A system and method for estimating blood concentrations of lithium

Abstract

The invention relates to a system and a method for estimating blood concentrations of lithium 12-hours after a lithium dose has been received or taken by a patient, but when the blood test has been taken at a different time than 12 hours after the last lithium dose. When a patient has received a lithium dose, it is registered and saved within a database. A lithium level registration unit configured to receive an indication of a determined patient's lithium level and the time the patient's lithium level was determined. When a patient's lithium level is registered then a processing unit is configured to obtain the patient's lithium level, retrieve the corresponding dose record from the database, perform estimation calculation to calculate an estimated 12-hours blood concentration of lithium (eLi12) applying the patient's lithium level, the time the patient's lithium level was determined, and the time the patient received the lithium dose, and output the estimated 12-hours blood concentration of lithium (eLi12).

Description

[0001] 83894PC01

[0002] 1

[0003] A SYSTEM AND METHOD FOR ESTIMATING BLOOD CONCENTRATIONS OF LITHIUM

[0004] FIELD OF THE INVENTION

[0005] The present invention relates to a system and a method for estimating blood concentrations of lithium.

[0006] BACKGROUND OF THE INVENTION

[0007] Lithium plays an important role in the treatment of bipolar disorder, with approximately 2.4 million prescriptions sold annually in the United States. Maintaining blood lithium levels within a narrow therapeutic range — typically 0.6 to 1.2 mEq / L (milliequivalents per liter) — is essential for effective treatment.

[0008] Clinicians rely on estimating lithium levels, usually by measuring them approximately 12 hours after the last dose. Based on the 12-hour trough level, clinicians adjust the dose to avoid subtherapeutic levels (< 0.6 mEq / L) or toxic levels (> 1.2 mEq / L).

[0009] However, many patients inadvertently have their lithium levels measured either before or after the standard 12-hour window, potentially leading to inaccuracies in measurements.

[0010] For decades, lithium has been a standard mood-stabilizing treatment for bipolar disorder. Treatment doses are determined based on clinical efficacy and lithium levels. These Lithium levels correlate with clinical outcomes and can be measured accurately. Guidelines recommend monitoring Lithium levels during dose adjustments and regularly when patients are on stable doses. Beyond treatment decisions, monitoring Lithium levels helps prevent adverse effects associated with high levels. Lithium's impact on the kidneys and thyroid gland is well-known, given its narrow therapeutic window.

[0011] The current standard practice involves measuring Lithium levels 12 hours after the patient's last lithium dose, with guidelines suggesting a window between 8 and 14 hours. Ideally, patients should receive their lithium dose in the evening and undergo blood tests the following morning before any morning dose. However, 83894PC01

[0012] 2 adherence to lithium treatment is often suboptimal. Consequently, patients may not adhere to the 12-hour requirement, leading to a time lag between the last lithium dose and Lithium measurement. This discrepancy could result in clinical challenges — for instance, falsely low Lithium levels (e.g., interpreting an 18-hour level as the 12-hour level) leading to unnecessary dose adjustments and increased risk of side effects.

[0013] One method has been proposed by ALDA ET AL: "Method for prediction of serum lithium levels", BIOLOGICAL PSYCHIATRY, ELSEVIER, AMSTERDAM, NL, vol. 24, no. 2, 1 June 1988 (1988-06-01), pages 218-224. It discloses a new computer- assisted method for predicting lithium levels and consulting dosage of lithium. The method is based on a mathematical model describing elimination of lithium after lithium treatment of any duration and regularity. From two values of serum concentration of lithium obtained during a single day, parameters of the model are computed and used for prediction. In an evaluation study involving 20 inpatients, the results demonstrated a high correlation between predicted and observed levels (r = 0.80) and revealed no systematic error of prediction. The elimination rate of lithium in these patients was unrelated to age or to duration of previous lithium therapy. However, the clear limitation of this study is that it requires 2 blood tests on the same day, which for practical reasons is not possible to implement in everyday clinical practice. For >60 years, it has been gold standard to measure the lithium blood concentration in one morning blood test.

[0014] Therefore, an improved system and method for more efficient and reliable estimation of lithium levels without changing the present workflow for patients and clinicians would be advantageous.

[0015] OBJECT OF THE INVENTION

[0016] It is an object of the present invention to provide an improved system and method to estimate lithium levels in patients.

[0017] It is a further object of the present invention to provide an alternative to the prior art. 83894PC01

[0018] 3

[0019] In particular, it may be seen as an object of the present invention to provide a system and method that solves the above-mentioned problems of the prior art with patients not getting blood tests taken exactly 12 hours after the last lithium dose.

[0020] SUMMARY OF THE INVENTION

[0021] Thus, the above-described object and several other objects are intended to be obtained in a first aspect of the invention by providing a system to estimate 12- hours blood concentration of lithium comprising:

[0022] - a processing unit,

[0023] - a database comprising dose records of the respective times when patients received a lithium dose and the amount of the dose,

[0024] - one, or more, dose registration unit(s) configured to o register a dose record comprising the time a patient received a lithium dose and the amount of the dose, o assign a unique patient ID to the dose record, and o transmit said dose record to the processing unit to be stored within the database,

[0025] - a lithium level registration unit configured to receive an indication of a patient's lithium level, and the time the patient's lithium level was determined; wherein the processing unit is configured to:

[0026] - obtain the patient's lithium level from the lithium level registration unit, retrieve the corresponding dose record from the database, perform estimation calculation to calculate an estimated 12-hours blood concentration of lithium (eLiiz) applying the patient's lithium level, the time the patient's lithium level was determined, and the time the patient received the lithium dose, and

[0027] - output the estimated 12-hours blood concentration of lithium (el_ii2) . 83894PC01

[0028] 4

[0029] This invention presents a new concept to estimate the 12-hours blood concentration of lithium independent of when the lithium blood test is taken, which has the potential to improve real-world lithium treatment.

[0030] The term "lithium level" is synonymous with "blood concentration of lithium". Throughout this document, "lithium level" refers to the blood concentration of lithium. The phrase "lithium level" is frequently used for simplicity. Therefore, the term "estimated 12-hour lithium level" is synonymous with "estimated 12-hour blood concentration of lithium", and the term "patient's lithium level" is synonymous with "the patient's blood concentration of lithium".

[0031] The invention has actionable, clinical implications for several reasons. First, both research and clinical experience indicate that patients may have limited compliance with the 12-hour blood tests. This creates a time-lag between the last lithium dose and the actual lithium measurement, which may be due to missed doses, waiting time, logistical challenges, or insufficient knowledge about the 12- hour level.

[0032] Even slightly higher lithium levels can increase the long-term risk for potentially severe side effects and even kidney damage. The importance of estimating a 12- hour lithium level is further emphasized as patients often receive lithium for years or decades and often are followed in non-specialized settings without expert knowledge on the 12-hour levels.

[0033] By only requiring the time in hours since the last lithium dose, the invention provides an estimated 12-hour blood concentration of lithium (eLii?), that could help clinicians to better evaluate the measured lithium level. Furthermore, this gives an entirely new flexibility to the patient, as the lithium blood test can be taken at other time points than 12 hours. In addition, the estimated 12-hour blood concentration of lithium (eLii?) may support the use of lithium for patients where clinicians else would prefer other drugs without the need for specific blood test timing. Finally, as the time since the last lithium dose is the only required but hence very important variable, it is reassuring that the concept of the invention was supported in trials where patients reported this information and where we specifically registered the time of lithium intake. 83894PC01

[0034] 5

[0035] An advantage of the invention is that it requires only a single determination of the patient's lithium level, meaning only one blood test at a time point chosen by the lithium-treated patient, thereby enabling practical clinical implementation and simplifying the procedure for both patients and healthcare providers. This is unlike the Alda et al. article cited in the background of the invention, which requires two values of the patient's lithium level obtained during a single day and hence that patients to show up at the clinic twice a day, which is not practical possible for patients.

[0036] Furthermore, the timing of the blood sample collection does not need to be precisely 12 hours after dosing, offering greater flexibility in clinical settings.

[0037] The dose registration unit is a device used to register the dose received by a patient and the time the dose was received. It should be understood that when a patient receives a lithium dose, it is usually the patient themselves who takes it. The dose registration unit may be a computer or a mobile phone.

[0038] The dose registration unit may be a patient registration unit applied by the patient to register when a dose was received or taken. Then the patient registers the time and dose on the dose registration unit and the registration unit transmits a dose record comprising the dose, time, and patient ID to the database where the dose record is stored.

[0039] Alternatively, the dose registration unit may be a medical staff registration unit, in this case it is a person from the medical staff that is entering the dose, time and patient ID into the registration unit and the registration unit transmits a dose record comprising the dose, time and patient ID to the database where the dose record is stored.

[0040] The system estimates the 12-hour blood concentration of lithium 12 hours after the lithium dose was taken by the patient. However, the system may also be used to estimate the blood concentration of lithium at other times, so the expression "estimated 12-hour blood concentration of lithium" may be understood as the estimated blood concentration of lithium after for instance 8 hours, 10 hours, 14 83894PC01

[0041] 6 hours, 16 hours or any other suitable time after the lithium was received by the patient. The expression "estimated 12-hour blood concentration of lithium" is used, because 12 hours is the timespan normally used in practice.

[0042] The database comprises a collection of dose records, each dose record comprising a patient ID, dose, and the time the dose was received.

[0043] When an indication of a patient's lithium level is received by the lithium level registration unit, the patient's lithium level, the time it was determined, and the patient ID is transmitted to the processing unit, which then finds the last dose record for the given patient and estimates the 12-hours blood concentration of lithium for the given patient. The patient's lithium level usually is determined by being measured using standard techniques.

[0044] The processing unit may be a computer which is receiving dose records from the dose registration units and stores the dose records within the database. The processing unit may be the dose registration unit.

[0045] The processing unit may further receive an indication of a patient's lithium level and the time the patient's lithium level was determined from the lithium level registration unit and then estimate the patients 12-hours blood concentration of lithium by finding the corresponding dose record for the given patient within the database and calculate how much the time difference between when the lithium dose was received and when the patient's lithium level was determined differs from 12 hours and then use the time difference and the indication of a patient's lithium level to determine the estimated 12-hours blood concentration of lithium for the patient.

[0046] The processing unit may be the same unit as the lithium level registration unit, so the same unit may be used both as the processing unit and the lithium level registration unit.

[0047] The corresponding dose record for the given patient is the last record stored before the indication of a patient's lithium level was determined. 83894PC01

[0048] 7

[0049] The processing unit may also store a lithium level record within the database with information about the patient ID and the patient's lithium level and the time the patient's lithium level was determined for later use.

[0050] When the estimation calculation to calculate an estimated 12-hours blood concentration of lithium has been performed the processing unit may output the estimated 12-hours blood concentration of lithium for the given patient. The output may be presented on a screen or printed on paper. Further the output may be stored within the database.

[0051] When the processing unit outputs the estimated 12-hours blood concentration of lithium it may also be informing the clinician or doctor and guiding on clinical decision making on the individual lithium dose.

[0052] The output, which is an estimated 12-hours blood concentration of lithium may be used be a doctor to adjust the dose of lithium the patient is receiving.

[0053] Regarding "a processing unit", it is to be understood that the processing unit may be one, or more, processing units, which may be computers, so it may be different processing units used to store data and calculate the estimated 12-hours blood concentration of lithium.

[0054] The dose registration unit may be a computer or a mobile phone. Especially the patient's registration unit may preferably be a mobile phone. Also, the lithium level registration unit may be a computer or a mobile phone.

[0055] The database may comprise a central database and / or several sub-databases wherein each sub-database contains specific sets of data. These sub-databases might be linked to different units or components within the system. Despite being stored in separate locations, the data from these sub-databases may be accessed and used together for tasks such as making estimations and performing data analysis. 83894PC01

[0056] 8

[0057] According to an embodiment, the patient's lithium level applied in the estimation calculation of the 12-hours blood concentration of lithium, eLii2, is obtained from a single blood test.

[0058] In the usual clinical practice, where a patient comes to an outpatient clinic to have a blood test taken and leaves again afterwards, it is not practically possible to take more than one test, and therefore it is essential that only a single blood sample is needed to determine the 12-hours blood concentration of lithium (eLii?).

[0059] According to an embodiment the lithium level registration unit further is configured to receive a patient's creatinine level. The estimation calculation to calculate an estimated 12-hours blood concentration of lithium (eLii?) is further applying the patient's creatinine level, and the patient's creatinine level is obtained from the same single blood test as from which the patient's lithium level is obtained.

[0060] In addition to using the lithium concentration measured a number of hours after lithium intake, the eLil2 equation can include the patient's kidney function, via creatinine concentration, to further personalize the eLil2 estimation. This is important as lithium is entirely excreted through the kidneys, why an adaptation of the eLil2 depending on the kidney function will provide even more accurate estimations.

[0061] The patient's creatinine level may be used to adjust the calculation of the estimated 12-hours blood concentration of lithium (eLii?). The patient's creatinine level may be obtained from the same blood sample as the patient's lithium level is determined, so no additional blood sample is required. For >60 years, creatinine has always been measured together with the lithium blood concentration, as kidney function represents the most important marker for clinical safety during lithium treatment.

[0062] According to an embodiment the estimation calculation of the 12-hours blood concentration of lithium (eLii?) is performed by applying a declining function. 83894PC01

[0063] 9

[0064] The declining function is a mathematical model that represents the rate at which the lithium level decreases in the body. This function is based on the understanding that the concentration of lithium decreases over time as lithium is excreted from the body.

[0065] The declining function takes into account various factors such as the initial lithium level, the rate of lithium metabolism in the body, and the time elapsed since the last dose. By applying this function, the system can accurately estimate the blood concentration of lithium at the end of the 12-hour period.

[0066] According to an embodiment the estimation calculation of the 12-hours blood concentration of lithium (eLii?) is performed by applying a linear model, an exponential model, a power law model or a logarithmic decay model.

[0067] Different mathematical models may be used for the estimation calculation of the estimated 12-hours blood concentration of lithium, denoted as eLii2. The system can perform the estimation calculation by applying one of several models: a linear model, an exponential model, a power law model, a logarithmic decay model, or a double logarithmic decay model.

[0068] The linear model assumes a constant rate of decrease in the lithium level over the 12-hour period. The exponential model represents a rapid initial decrease in the lithium level, followed by a slower rate of decrease. The power law model describes a relationship between the lithium level and time that is proportional to a power n. The logarithmic decay model assumes a rapid initial decrease that slows down over time, while the double logarithmic decay model represents a more complex decay process that may better fit certain physiological conditions.

[0069] Each of these models has its own strengths and can be chosen based on the specific characteristics of the patient's lithium metabolism and the clinical scenario. By providing multiple model options, the system offers flexibility and allows for more accurate and personalized estimation of the blood concentration of lithium. 83894PC01

[0070] 10

[0071] According to an embodiment the estimated 12-hour blood concentration of lithium eLi 12 is determined from a function f(Li, h), wherein Li represents the actual determined patient's lithium level and h is the number of hours between the patient received the lithium dose and the time the patient's lithium level was determined.

[0072] A specific function, denoted as f(Li, h), is applied for determining the estimated 12-hour blood concentration of lithium, eLii2.

[0073] In this function, Li represents the actual determined patient's lithium level and h is the number of hours between the patient received the lithium dose and the time the patient's lithium level was determined. The function f(Li, h) takes these two parameters and applies a mathematical model to estimate the blood concentration of lithium at the 12-hour mark.

[0074] This function could be any of the models described above, such as a linear model, an exponential model, a power law model, a logarithmic decay model, or a double logarithmic decay model. The choice of model would depend on the specific characteristics of the patient's lithium metabolism and the clinical scenario.

[0075] By using the function f(Li, h), the system can provide a real-time estimate of the blood concentration of lithium at the 12-hour mark based on the most recent lithium level measurement and the time between the patient received the lithium dose and the time the patient's lithium level was determined. This feature allows for more precise and timely monitoring of lithium levels, thereby improving the safety and efficacy of lithium therapy.

[0076] According to an embodiment the estimation calculation of the estimated 12-hours blood concentration of lithium (eLii?) is performed according to the following equation: eLii2 = (Li1 / 2+ p * (12 - h))2, wherein is the coefficient, h is the difference between the time a patient's lithium level was determined and the time where the lithium dose was received by the patient, eLii2 is the estimated 12-hours blood concentration of lithium 83894PC01

[0077] 11 determined from the patient's lithium level, and Li is the actual determined patient's lithium level after h hours.

[0078] A specific equation for the estimation calculation of the 12-hours blood concentration of lithium, denoted as eLii? is: eLii? = (Li1 / 2+ p * (12 - h))2.

[0079] In this equation, is a coefficient that represents the rate of lithium metabolism in the body. The variable h is the difference between the time a patient's lithium level was determined and the time when the patient received the lithium dose. The estimated 12-hours blood concentration of lithium (eLii2) is determined from the actual determined patient's lithium level and the difference (h) between the time a patient's lithium level was determined and the time where the lithium dose was received. eLii2 represents the estimated blood concentration of lithium at the 12-hour mark. Li is the actual determined patient's lithium level. Li represents the most recent measurement of the lithium level in the patient's blood.

[0080] By using this equation, the system can provide a precise estimate of the patient's blood concentration of lithium at the 12-hour mark based on the most recent measurement of the patient's lithium level, the time elapsed since that measurement, and the rate of lithium metabolism in the body.

[0081] According to an embodiment the coefficient p is between -0.02 and -0.006.

[0082] The coefficient p represents the rate of lithium metabolism in the body. The range between -0.02 and -0.006 reflects the typical rates of lithium metabolism observed in patients undergoing lithium therapy.

[0083] According to an embodiment the estimation calculation of the estimated 12-hours blood concentration of lithium (eLii2) is performed according to the following equation: eLii2 = Li * exp(P * (h-12)), wherein p is the coefficient, h is the difference between the time a patient's lithium level was determined and the time where the lithium dose was received by the patient, eLii2 is the estimated 12-hours blood concentration of lithium 83894PC01

[0084] 12 determined from the patient's lithium level, and Li is the actual determined patient's lithium level after h hours.

[0085] An alternative equation for the estimation calculation of the 12-hours blood concentration of lithium is eLii2 = Li*exp(P*(h-12)). eLii? is the estimated 12-hour blood concentration of lithium. Li is the actual measured patient's lithium level at the patient-reported hour (h). The term exp(P * (h-12)) represents an exponential function, where p is a coefficient representing the rate of lithium metabolism in the body and h is the difference between the time a patient's lithium level was determined and the time where the patient received the lithium dose.

[0086] The exponential function models the rate at which the lithium level decreases over time, taking into account the determined lithium level (Li), the rate of lithium metabolism in the body (P), and the time elapsed since the last dose (h-12).

[0087] According to an embodiment the coefficient p is between -0.065 and -0.019.

[0088] The coefficient p represents the rate of lithium metabolism in the body. In the above-mentioned alternative equation, the range of p is between -0.065 and - 0.019.

[0089] In yet another alternative the equation to estimate the estimated 12-hours blood concentration of lithium (eLii2) may be eLii2 = exp(log(Li) + p * (h-12)) applying a log transform.

[0090] According to an embodiment the estimation calculation of the estimated 12-hours blood concentration of lithium (eLii2) is performed according to the following equation: eLii2 = exp(log(Li) + p * (h-12)), wherein p is the coefficient, h is the difference between the time a patient's lithium level was determined and the time where the lithium dose was received by the patient, eLii2 is the estimated 12-hours blood concentration of lithium 83894PC01

[0091] 13 determined from the patient's lithium level, and Li is the actual determined patient's lithium level after h hours.

[0092] In this alternative embodiment both eLii? and Li are transformed using a logarithmic function.

[0093] The logarithmic transformation is a mathematical operation that can help to linearize the relationship between eLii? and Li, making the estimation calculation more robust to variations in the data. It can also help to stabilize the variance and make the data more normally distributed, which can improve the performance of the estimation calculation.

[0094] Furthermore, this embodiment specifies a range for the coefficient p, which is between -0.025 and -0.006.

[0095] In yet another alternative the equation to calculate the estimated 12-hours blood concentration of lithium eLii2 may be a power transform where eLii2 may be determined from the equation eLH2p= Lip+ * h - 12).

[0096] According to an embodiment the estimation calculation of the estimated 12-hours blood concentration of lithium (eLii2) is performed according to the following equation: eLH2p= Lip+ * (7i - 12), wherein is the coefficient, h is the difference between the time a patient's lithium level was determined and the time where the lithium dose was received by the patient, eLii2 is the estimated 12-hours blood concentration of lithium determined from the patient's lithium level, and Li is the actual determined patient's lithium level after h hours.

[0097] In this alternative embodiment, both eLii2 and Li are transformed using a power function. Specifically, the estimated 12-hour lithium level eLii2 is determined using a power transform, and Li is applied in the power transform.

[0098] Power transformation is a mathematical operation that raises a variable to a certain power. This transformation can help to linearize the relationship between 83894PC01

[0099] 14 eLiiz and Li, especially when the relationship exhibits a non-linear pattern. It can also help to stabilize the variance and make the data more normally distributed, which can improve the performance of the estimation calculation.

[0100] Furthermore, this embodiment specifies a range for the power transform p, which is between 0.1 and 1.6. Additionally, claim 12 specifies a range for the coefficient P, which is between -0.04 and -0.008.

[0101] According to an embodiment, the coefficient is a fixed selected value. The fixed coefficient p is used to analyse all blood samples for all patients.

[0102] According to an embodiment, the lithium level registration unit further is configured to receive a patient's creatinine level and wherein the coefficient p is determined based on a patient's creatinine level.

[0103] The coefficient p may be determined based on a patient's creatinine level. The creatinine level can be determined from the same blood sample from which the patient's lithium level is determined, so no extra tests of the patient is needed.

[0104] The coefficient p may be determined from a table in the database.

[0105] Alternatively, the coefficient p may be calculated using an equation where p = f(c), where c is the patient's creatinine level.

[0106] The equation may be of the form p = Po + x * (c - co), where Po is a selected initial p value at the selected creatinine level co, and x is a constant multiplied by the difference between the determined patient's creatinine level c and the selected creatinine level co. The x-value is selected such that it changes the beta from e.g. 0.019 to 0.012 or to 0.026 depending on the difference between c and co. co may for instance be 73 pmol / L, such that p may increase or fall depending on whether c is larger than or smaller than 73 pmol / L.

[0107] According to an embodiment, the database comprises a table wherein the coefficient p can be obtained based on the patient's creatinine level. 83894PC01

[0108] 15

[0109] The database may comprise a table wherein the patient's creatinine level is used to find the coefficient p.

[0110] According to an embodiment the dose registration unit is a patient registration unit configured to receive, from the patient, the dose record comprising the time the patient reports to have received the last lithium dose, and to transmit said dose record to be stored within the database.

[0111] The dose registration unit may be a computer, or it may be a smartphone or any other device making it possible for the patient to register when he / she has received a lithium dose. When a patient uses the dose registration unit it is called a patient registration unit. A program or an app may be installed on the patient registration unit making it easy for the patient to register when he / she has received a lithium dose. The patient registration unit then is configured to transmit the dose record to a processing unit, which may be a central computer which then stores the dose record within the database.

[0112] According to an embodiment the patient registration unit is a smartphone wherein the patient enters the time that the patient received the last lithium dose via an app.

[0113] The app may be set up to know the patient ID and the size of the dose the patient is receiving, therefore the app may be quite simple so the patient may only have to push a button when a dose has been received and then the app obtain the time from the smartphone and transmits the dose size, patient ID and time in the dose record to a processing unit which stores the dose record within the database.

[0114] According to an embodiment the system is configured to send reminder notifications via the app to the patient in relation to the time when the patient should receive the lithium dose.

[0115] The app may also be configured to send a reminder notification to the patient when the time has come when the patient is supposed to receive the dose. For instance, the central processing unit may send a reminder to the app to remind 83894PC01

[0116] 16 the patient to take the lithium dose 12 hours before it is scheduled for the patient to have a blood test taken.

[0117] According to an embodiment the dose registration unit is a medical staff registration unit configured to receive, from a medical staff member, the dose record comprising the time the patient has received the last lithium dose, and to transmit said dose record to the system to be stored within the database.

[0118] It is also possible that the dose registration unit is located at a hospital or clinic where the patient may be present, and then a person from the medical staff is giving the dose to the patient and then the person from the medical staff may use the dose registration unit to register the dose record, which in this case may be called a medical staff registration unit.

[0119] According to an embodiment the system is further configured to send a reminder notification via the medical staff registration unit to the medical staff member at a predetermined amount of time after the inputted time the patient has received the last lithium dose.

[0120] When the patient is at the hospital or clinic the system may send a reminder to the medical staff registration unit, so that the medical staff can administer the dose to the patient.

[0121] According to an embodiment the dose record is stored within the database marked with the patient ID and the system applies the patient ID to retrieve the last dose record from the database.

[0122] In the database the dose record is stored and when the dose record is retrieved, it is retrieved by entering the patient ID in a processing unit, like a computer or other device, which may access the database to retrieve the dose record for the patient. Typically, it will be the last registered dose record which may be retrieved to be used to estimate a 12-hour blood concentration of lithium.

[0123] According to an embodiment the lithium level registration unit is configured to transmit the patient's blood concentration of lithium to the processing unit. 83894PC01

[0124] 17

[0125] When a person from the medical staff obtains a patient's lithium level, the staff member will enter the lithium level into a lithium level registration unit, which may be a computer, a smart phone or any other suitable device, and from the lithium level registration unit the patient's lithium level is transmitted to the central processing unit where the corresponding dose record is retrieved from the database and the estimated 12-hour blood concentration of lithium is calculated.

[0126] According to an embodiment the dose registration unit is a medicine container, configured to generate the dose record comprising the time when the medicine container has dispensed a dose of lithium to the patient, and to transmit said dose record to the system to be stored within the database.

[0127] The patient may receive the dose from a machine container. In this case the machine container may be the dose registration unit and m automatically record and transmit the dose record to the processing unit for storing within the database when the dose has been administrated.

[0128] According to an embodiment the system is further configured to create a prioritization of patients within said database to have a blood sample taken applying i) whether the dose record was received from a specific dose registration unit, and ii) the reported time since the last lithium dose was received by the patients.

[0129] In a hospital or clinic where a medical staff member is taking care of administering doses to several patients, the system may be able to prioritize the order in which patients are having a blood sample taken by retrieving information from the database about when the patients last received a lithium dose so that the patient may have the blood sample taken as close to the 12 hour mark as possible. Using the information about which dose registration unit was used to register the last dose to indicate whether the patient is at the hospital or clinic or not. The dose registration unit may be additional information stored in the dose record. 83894PC01

[0130] 18

[0131] The system of the first aspect of the invention only uses the determined patient's lithium level, the time between the taking a dose and determining the patient's lithium level, the coefficient p, and optionally the patient's creatinine level determined in the same blood sample as the patient's lithium level to adjust the coefficient , to estimate the 12-hours blood concentration of lithium. This is a simple but efficient method to determine the 12-hours blood concentration of lithium.

[0132] In a second aspect, the invention relates to a computer-implemented method for estimating 12-hours blood concentration of lithium, wherein the method comprises

[0133] - registering by a dose registration unit, a dose record comprising a time a patient received a lithium dose and the amount of the dose,

[0134] - assigning a unique patient ID to the dose record,

[0135] - transmitting from the dose registration unit the dose record to a processing unit to be stored within a database,

[0136] - storing the dose record within the database,

[0137] - obtaining a patient's lithium level and the time the patient's lithium level was determined from a lithium level registration unit,

[0138] - performing an estimation calculation, applying the patient's lithium level, the time the patient's lithium level was determined, and the time the patient received the lithium dose, to calculate an estimated 12-hours blood concentration of lithium (eLii?), and

[0139] - outputting the estimated 12-hours blood concentration of lithium (eLii?).

[0140] The first and second aspect of the present invention may each be combined with any of the other aspects. These and other aspects of the invention will be apparent from and elucidated with reference to the embodiments described hereinafter.

[0141] BRIEF DESCRIPTION OF THE FIGURES

[0142] The system and method for estimating blood concentration of lithium according to the invention will now be described in more detail with regard to the accompanying figures. The figures show one way of implementing the present 83894PC01

[0143] 19 invention and is not to be construed as being limiting to other possible embodiments falling within the scope of the attached claim set.

[0144] Fig. 1 illustrates the system of the invention.

[0145] Fig. 2 illustrates the dose record.

[0146] Fig. 3 illustrates the lithium level record.

[0147] Fig. 4 illustrates the output 42. The output comprises as a minimum the patient ID and the determined 12-hour blood concentration of lithium.

[0148] Fig. 5 shows a plot based on data from a Bipolar CHOICE Discovery trial, plotting the association between the time in hours since the last lithium dose and the lithium level among 122 patients

[0149] Fig. 6 shows data from the Aarhus eLii? proof-of-concept trial plotting the individual-level actual measured patient's lithium level and the elii2 level based on the time in hours since the last lithium dose.

[0150] Fig. 7 shows Table 4: Findings from the proof-of-concept trial conducted in Aarhus, Denmark, to test the final accuracy of the first developed eLii2 equation eLii2 = (Li1 / 2+ p * (12 - h))2.

[0151] Fig. 8 shows Table 5: Findings from the proof-of-concept trial conducted in Aarhus, Denmark, to test the final accuracy of the second developed eLii2 equation eLii2 = Li * exp(P * (h-12)).

[0152] Fig. 9 is a flow-chart of the method according to the invention.

[0153] Fig. 10 is a diagram showing the change of the measured lithium level, multiplied by 100, over a period of 24 hours after the dose was taken based on different creatinine values at time zero.

[0154] DETAILED DESCRIPTION OF AN EMBODIMENT

[0155] Fig. 1 illustrates the system of the invention. The system comprises a processing unit 10, dose registration units 12, 14, 15, which may be a patient registration unit 12, a medical staff registration unit 14, or a medicine container 15. The dose registration units are used to enter data about when a dose is received by the patient. A dose record comprising the size of the dose, the time the dose was received by the patient and the patient ID is transmitted, possible wireless, to the processing unit, which is then storing the dose record within the database 20. A 83894PC01

[0156] 20 lithium level registration unit 16 is used to enter the patient's lithium level into the processing unit. When receiving a patient's lithium level, the processing unit 10 retrieves the corresponding dose record within the database for the given patient ID. The corresponding dose record is the last dose record stored within the database before the time of the determination of the patient's lithium level. The processing unit is then calculating the estimated 12-hour blood concentration of lithium for the patient and is outputting the estimated 12-hour blood concentration of lithium on the output unit 18. The output may be displayed on a screen or printed on paper. Further the output may be stored within the database.

[0157] Fig. 2 illustrates the dose record 22. A dose record may be stored within the database 20 for each for each lithium dose registered to be received by a patient. The dose record comprises the patient ID 24, the time 26 the dose was received and the size of the dose 28.

[0158] Fig. 3 illustrates the lithium level record 32. The lithium level record comprises the patient ID 34, the determined patient's lithium level 38 and the time 36 the patient's lithium level was determined. Optionally, the lithium level record may also comprise the determined patient's creatinine level (39), which is obtained from the same blood sample as the patient's lithium level. The lithium level record may be stored within the database 20 for later use, for instance for statistics.

[0159] Fig. 4 illustrates the output 42. The output comprises as a minimum the patient ID 44 and the estimated 12-hour blood concentration of lithium 46. The output may be stored within the database 20 for later use, for instance for statistics. The output may comprise further information, for instance the patient's name and time the 12-hour blood concentration of lithium was estimated and other relevant information.

[0160] Fig. 5 shows a plot based on data from a Bipolar CHOICE Discovery trial, plotting the association between the time in hours since the last lithium dose and the lithium level among 122 patients yielding 192 measurements after 16 and 24 weeks of lithium treatment. 83894PC01

[0161] 21

[0162] The eLii? equation was developed in secondary analysis using data from the Bipolar CHOICE study, a 6-month multi-site, randomized trial comparing lithium to quetiapine combined with other guideline-informed medications for bipolar disorder. 240 patients took part in the study.

[0163] Table 1 shows data from the Bipolar CHOICE trial, Mean and Standard deviation (SD) of Patient Characteristics by Time Since last Lithium Dose.

[0164] Among the 240 patients, lithium levels (in milliequivalents per liter, mEq\L) were measured via an antecubital vein blood sample after 2, 16 and 24 weeks of treatment at normal follow-up visits. At the same visits, patients reported the number of hours between their last lithium dose and the blood test.

[0165] The target lithium was 0.6 mEq\L, corresponding to an approximate dose of 900 mg lithium. Patients started with lithium treatment at study entry; hence, the 2- week visit was during up-titration. Most patients were on their stable maintenance dose at weeks 16 and 24.

[0166] Gaussian regression was used to predict lithium levels from the time after the last dose. Because lithium levels were not normally distributed, the square root of lithium levels was used as the dependent variable.

[0167] Patients were grouped based on the time since their last lithium dose: <8, 8-12, 12-14, 14-16, 16-20, and >20 hours. Analysis of variance (ANOVA) was performed using the guideline-recommended group of 8-12 hours as the reference group.

[0168] Table 1: Data from the Bipolar CHOICE trial, Mean and Standard Deviation (SD) of Patient Characteristics by Time Since last Lithium Dose. 83894PC01

[0169] 22

[0170] Of the 240 patients, 145 had at least once reported the time since their last lithium dose at a visit when lithium levels were measured, see table 1.

[0171] The primary analysis used patients on a stable maintenance dose, i.e., after 16 or 24 weeks of treatment. A total of 122 patients yielded 192 lithium level measurements. The mean lithium was 0.63 mEq / L (SD=0.33; range=0.05-1.9), the mean lithium dose was 1002.0 mg (SD=259.7; range = 150-1500) and the mean time since the last lithium dose was 12.4 hours (SD=5.6; range = l-36). The 53 (27.6%) who reported that they took their last lithium dose 12 hours before blood sampling had a mean lithium level of 0.74 (SD=0.29) and a mean lithium dose of 1062.2 mg (SD=270.1).

[0172] Fig. 5 shows the plot between lithium levels and the time since the last dose.

[0173] When compared to patients with 8-12 hours since the last lithium dose, patients with 14-16 hours, 16-20 hours and >20 hours had significantly lower lithium levels (Table 1). Patients with 8-12 hours since the last lithium dose had a more than 3-fold higher (0.72 versus 0.21 mEq\L) lithium compared to patients with >20 hours.

[0174] Gaussian regression found a significant association between a longer time since the last lithium dose with lower lithium levels, p-coefficient=-0.013.

[0175] Based on the Gaussian regression model predicting lithium level from time since last dose, the following equation for calculating the estimated 12-hour se-Li level was developed: eLiiz = (Li1''2± p * (12 - h))2where eLii? is the estimated 12-hour blood concentration of lithium. Li is the actual measured patient's lithium level at the patient-reported hour (h) after the last lithium dose and p is the coefficient from the regression model, i.e., -0.013. 83894PC01

[0176] 23

[0177] Alternative el_ii2 equations were also developed and also a second equation eLii2 = Li * exp(6 * (h-12)) was tested.

[0178] FIRST PROOF-OF-CONCEPT TRIAL

[0179] To test the accuracy of the eLii? equations a proof-of-concept trial was performed at the Massachusetts General Hospital bipolar disorder outpatient clinic (Boston, USA). 5 patients were recruited aged 18-65 years with a bipolar disorder diagnosis who were on a stable dose of lithium.

[0180] Two blood tests from each participant were taken via standard antecubital vein blood sample. The first was 12 hours after their self-reported last lithium dose; the second was between 3 and 6 hours later.

[0181] Table 2: Results from the Boston proof-of-concept trial assessing the accuracy of the eLii2 equation eLii2 = (Li1 / 2+ * (12 - h))2.

[0182] Notes: el_ii2 = Estimated 12-hour blood concentration of lithium; Absolute deviation: difference between predicted and actual 12-hour blood concentration of lithium;

[0183] Percent deviation: the difference as a percentage of the actual 12-hour blood concentration of lithium.

[0184] Table 2 shows results from the Boston proof-of-concept trial assessing the accuracy of the eLii? equation for the equation eLii2 = (Li1 / 2+ * (12 - h))2.

[0185] Data are presented as means with deviations. To compare the estimated eLii2 blood concentration of lithium to the actual measured lithium level at other time 83894PC01

[0186] 24 points than the 12-hour level, the absolute and percent deviation between these two values was calculated.

[0187] The 5 participants, mean lithium dose 600 mg, provided one blood test 12 hours after the self-reported last lithium intake and another between 2.9 and 6.17 hours after the 12-hour blood sample. Table 2 shows that for all patients, except #6, eLii2 was closer to the true 12-hour level compared with the actual level. For #6, the deviation was the same for both.

[0188] Table 3: Results from the Boston proof-of-concept trial assessing the accuracy of the el_ii2 equation el_ii2 = Li * exp(P * (h-12)).

[0189] Table 3 shows results from the Boston proof-of-concept trial assessing the accuracy of the eLii? equation for the equation eLii2 = Li * exp(P * (h-12)).

[0190] Table 3 shows that for all patients eLii2 was closer to the true 12-hour level compared with the actual level.

[0191] SECOND PROOF-OF-CONCEPT TRIAL

[0192] A second proof-of-concept trial was performed at Aarhus University Hospital Psychiatry in Aarhus, Denmark.

[0193] We recruited patients hospitalized at the department for affective disorders at Aarhus University Hospital psychiatry. 83894PC01

[0194] 25

[0195] Participation involved up to 9 antecubital vein blood samples during 24 consecutive hours over the course of two days. Participants took their daily lithium dose in the evening, mostly at 8 pm. The first blood sample was taken immediately before this lithium dose. The remaining 8 blood samples were taken at 2, 4, 9, 11, 12, 14, 20, and 24 hours after the first blood sample. Some samples were taken at slightly different time points if preferred by the patient, e.g., at 6 am instead of 5 am for the 9-hour level. Each blood sample was centrifuged within 30-60 minutes after sampling and stored at -80 C. After all participants had been recruited, lithium levels were measured for all samples at the same time.

[0196] Based on the data from the Bipolar CHOICE trial, we assumed a power of 80%, a standard deviation (SD) of 0.3, which was the SD for the mean lithium levels as a function of time (hours) since the most recent lithium dose, and a width of 0.1, assuming a rather narrow confidence interval, which resulted in 154 required blood samples.

[0197] The primary endpoint was whether eLii2, according to the equation eLii? = (Li1 / 2+ P * (12 - h))2, could estimate a 12-hour blood concentration of lithium within an acceptable range of the measured 12-hour blood concentration of lithium without giving falsely high or low lithium values. All lithium data from participants with a measured blood concentration of lithium at 12 hours and at least one other time point was included for analysis.

[0198] We calculated the mean differences including standard deviation (SD) between the eLii2 from the different time points and the determined 12-hour blood concentration of lithium in absolute and relative (i.e., percentage) terms.

[0199] The 23 participants, receiving a mean lithium dose of 19.8 mmol, provided 163 blood tests. Two participants did not have a blood test at 12 hours, leaving 21 participants with 159 blood tests. The mean 12-hour blood concentration of lithium of the 21 participants was 0.57 (SD=0.16, range = 0.27-0.81).

[0200] Fig. 6 shows data from the Aarhus eLii2 proof-of-concept trial plotting the mean values of the actual measured patient's lithium levels and the elii2 12-hour 83894PC01

[0201] 26 estimated lithium levels based on the time in hours since the last lithium dose using the equation eLii? = (Li1 / 2+ p * (12 - h))2. The plot 61 of the blood-based values represents the mean of all participants measured patient's lithium levels. The plot 62 represent the mean elii2 estimated 12-hour lithium level.

[0202] Fig. 6 shows the measured lithium levels and the eLii2 12-hours estimated lithium levels using the equation eLii2 = (Li1 / 2+ p * (12 - h))2during the 24 hours after the last lithium dose for each of the 21 participants with valid 12-hour levels. Ideally, the eLii2 lines, the plot 62 of the eLii2 in Fig. 6, should be horizontal as it aims to estimate the same 12-hour lithium level. The line is not perfectly horizontal but is much closer to horizontal than the actual determined patient's lithium levels 61.

[0203] Fig. 7 shows Table 4: Findings from the second proof-of-concept trial to test the final accuracy of the developed eLii2 equation eLii2 = (Li1 / 2+ p * (12 - h))2. Table 4 shows measured lithium level and estimated 12-hours lithium levels, eLii2, including differences to measured 12-hour lithium levels.

[0204] Table 4 shows in the first column the number of hours after the lithium dose was received that the lithium measurement was taken, the second column "N" shows the number of patients for which a measurement was taken. The column "Measured 12-hour lithium level" shows the means for measurements of lithium level taken 12 hours after the lithium dose was received. The column "Measured mean + / - SD" shows the mean of the measurements made the number of hours stated in column 1 of the lithium level. The column "Actual difference to 12-h lithium level" shows the difference between the measured 12-hour lithium level" and the "Measured, mean+ / -SD". The column "eLii2, mean+ / -SD" shows the mean estimated level at 12 hours according to the equation eLii2 = (Li1 / 2+ p * (12 - h))2. The column "Mean eLii2 difference to 12-h Lithium Level" shows the difference between the estimated 12-hour level (eLii2) and the measured 12-hour lithium level from the third column. The column "Mean eLii2 Difference -Actual" shows the difference between the "actual difference to 12-h Lithium Level" and the mean estimated eLii2. 83894PC01

[0205] 27

[0206] The actual measured 12-hour lithium level and the estimated 12-hours lithium levels (eLii?) including the absolute and percentage deviations between these measures are presented in Table 4 (Fig. 7). When including all blood tests, excluding those at 0 and 12 hours, the mean difference from the actual measured lithium level compared to the measured 12-hour lithium level was 0.14, while the difference was 0.07 for eLii2, the estimated 12-hours lithium levels. The difference between the measured actual lithium level and the estimated 12-hour lithium level eLii2 was larger at the more extreme time points, e.g., at 4, 19 or 20 hours, with eLii2 being notably closer to the measured 12-hour lithium level even at these extreme time points. For example, after 20 hours, the measured mean lithium level was 0.21 (34%) lower than the measured 12-hour lithium level, while the mean eLii2 estimated 12-hours lithium level was only 0.07 (12%) lower.

[0207] For the 117 blood levels not taken at 0 or 12 hours, eLii2 estimated 12-hour lithium level was closer to the measured 12-hour lithium level in 105 blood tests (90%), while for two blood tests, eLii2 estimated 12-hour lithium level was similar to the measured lithium level. Those 10 blood tests, where eLii2 was not closer to the 12-hour level, were all taken 2 hours after the last lithium dose.

[0208] Fig. 8 shows Table 5: Findings from the second proof-of-concept trial to test the final accuracy of the second developed eLii2 equation eLii2 = Li * exp(P * (h-12)). The table 5 is similar to table 4 showing the measured 12-hour lithium level and the estimated 12-hours lithium levels, eLii2, including differences to measured 12- hour lithium levels.

[0209] Table 5 shows in the first column the number of hours after the lithium dose was received that the lithium measurement was taken, the second column "N" shows the number of patients for which a measurement was taken. The column "Measured 12-hour lithium level" shows the means for measurements of lithium level taken 12 hours after the lithium dose was received. The column "Measured mean + / - SD" shows the mean of the measurements made the number of hours stated in column 1 of the lithium level. The column "Actual difference to 12-h lithium level" shows the difference between the "Measured 12-hour lithium level" and the "Measured, mean+ / -SD". The column "eLii2, mean+ / -SD" shows the mean estimated level at 12 hours according to the equation eLii2 = Li * exp(P * 83894PC01

[0210] 28

[0211] (h-12)). The column "Mean eLii? difference to 12-h Lithium Level" shows the difference between the mean estimated 12-hour level (eLii?) and the actual measured 12-hour level from the third column. The column "Mean eLii? Difference -Actual" shows the difference between the "Actual Difference to 12-h Lithium Level" and the mean estimated eLii2.

[0212] The difference between the measured actual lithium level and the estimated 12- hour lithium level eLii2 was larger at the more extreme time points, e.g., at 4, 19 or 20 hours, with eLii2 estimated using the equation eLii2 = Li * exp(P * (h-12)) being notably closer to the measured 12-hour lithium level even at these extreme time points.

[0213] Thereby both table 4 for the equation eLii2 = (Li1 / 2+ p * (12 - h))2and table 5 for the equation eLii2 = Li * exp(P * (h-12)) shows a considerable improvement in estimating the 12-hours lithium level than using the measured lithium level.

[0214] Fig. 9 is a flow-chart of the invention. First step is registering (SI) the dose record comprising the time a patient received a lithium dose and the amount of the dose from the dose registration unit. Second step is assigning (S2) a unique patient ID to the dose record. This is done either automatically by the dose registration unit or entered by the patient or the medical staff member. Then the dose registration unit is transmitting (S3) the dose record to the processing unit, and the processing unit is storing (S4) the dose record within the database. The processing unit is obtaining (S5) the patient's lithium level from the lithium registration using and is then performing (S6) an estimation calculation applying the patient's lithium level and the time the patient received the lithium dose to calculate an estimated 12-hours blood concentration of Lithium (eLii2) and finally outputting (S7) the estimated 12-hours blood concentration of Lithium (eLii2).

[0215] Fig. 10 is a diagram showing the change of the lithium level over a period of 24 hours after the dose was taken. Fig. 10 shows lithium levels, multiplied by 100, after different numbers of hours from 0 hours to 24 hours. In this case two different p values were used. Among blood samples where the creatinine level is lower than 73 pmol / L, we got a coefficient p = 0.0262213, which is for the more specific eLil2 equation. For blood samples where the creatinine level is higher 83894PC01

[0216] 29 than 73 pmol / L, we got a coefficient p = 0.0124438, which is used for the more specific eLil2 equation.

[0217] The x-axis shows the time since the last lithium intake in hours; the time is rounded to the nearest full hour. The y-axis shows the lithium level * 100 in mEq / L.

[0218] As it can be seen the graph 110 for creatinine levels lower than 73 pmol / L, the lithium level in the blood drops faster when the creatinine level is low compared to the graph 111 for creatinine levels higher than or equal to 73 pmol / L. When the creatinine level is high the kidney works less well and therefore is slower to reduce the lithium level in the blood.

[0219] Fig. 10 shows that more precise lithium levels can be obtained by using the creatinine levels to select the coefficient taking into account the patient's kidney function.

[0220] The graphs in fig. 10 are based on samples of blood tests from different patients, where the median value of the measured creatinine level was 73 pmol / L and therefore in fig. 10 the samples was split in two groups having creatinine levels higher and lower than 73 pmol / L respectively.

[0221] Although the present invention has been described in connection with the specified embodiments, it should not be construed as being in any way limited to the presented examples. The scope of the present invention is set out by the accompanying claim set. In the context of the claims, the terms "comprising" or "comprises" do not exclude other possible elements or steps. Also, the mentioning of references such as "a" or "an" etc. should not be construed as excluding a plurality. The use of reference signs in the claims with respect to elements indicated in the figures shall also not be construed as limiting the scope of the invention. Furthermore, individual features mentioned in different claims, may possibly be advantageously combined, and the mentioning of these features in different claims does not exclude that a combination of features is not possible and advantageous.

Claims

83894PC0130CLAIMS1. A system to estimate 12-hours blood concentration of lithium comprising:- a processing unit (10),- a database (20) comprising dose records (22) of the respective times when patients received a lithium dose and the amount of the dose,- one, or more, dose registration unit(s) (12, 14, 15) configured to o register a dose record (22) comprising the time (26) a patient received a lithium dose and the amount (28) of the dose, o assign a unique patient ID (24) to the dose record, and o transmit said dose record to the processing unit (10) to be stored within the database (20),- a lithium level registration unit (16) configured to receive an indication of a patient's lithium level, and the time the patient's lithium level was determined; wherein the processing unit (10) is configured to:- obtain the patient's lithium level (38) from the lithium level registration unit (16), retrieve the corresponding dose record (22) from the database, perform estimation calculation to calculate an estimated 12-hours blood concentration of lithium (eLii?) applying the patient's lithium level (38), the time (36) the patient's lithium level was determined, and the time the patient received the lithium dose (26), and- output the estimated 12-hours blood concentration of lithium (eLiu).

2. The system according to claim 1, wherein the patient's lithium level (38) applied in the estimation calculation of the 12-hours blood concentration of lithium (eLii?) is obtained from a single blood test.

3. The system according to claim 2, wherein the lithium level registration unit further is configured to receive a patient's creatinine level, the estimation calculation to calculate an estimated 12-hours blood concentration of lithium83894PC0131(eLiiz) is further applying the patient's creatinine level, and the patient's creatinine level is obtained from the same single blood test as from which the patient's lithium level is obtained.

4. The system according to any of the preceding claims, wherein the estimation calculation of the 12-hours blood concentration of lithium (eLii?) is performed by applying a declining function.

5. The system according to any of the preceding claims, wherein the estimation calculation of the 12-hours blood concentration of lithium (eLii?) is performed by applying a linear model, an exponential model, a power law model or a logarithmic decay model.

6. The system according to any of the preceding claims, wherein the estimated 12-hour blood concentration of lithium (eLii2) is determined from a function f(Li, h), wherein Li represents the actual determined patient's lithium level and h is the number of hours between the patient received the lithium dose and the time the patient's lithium level was determined.

7. The system according to any of the preceding claims, wherein the estimation calculation of the estimated 12-hours blood concentration of lithium (eLii?) is performed according to the following equation: eLii2 = (Li1 / 2+ p * (12 - h))2, wherein is the coefficient, h is the difference between the time a patient's lithium level was determined and the time where the lithium dose was received by the patient, eLii2 is the estimated 12-hours blood concentration of lithium determined from the patient's lithium level, and Li is the actual determined patient's lithium level after h hours.

8. The system according to claim 7, wherein the coefficient p is between -0.02 and -0.006.

9. The system according to any of the claims 1-6, wherein the estimation calculation of the estimated 12-hours blood concentration of lithium (eLii2) is performed according to the following equation:83894PC0132 eLii? = Li * exp(P * (h-12)), wherein p is the coefficient, h is the difference between the time a patient's lithium level was determined and the time where the lithium dose was received by the patient, eLii? is the estimated 12-hours blood concentration of lithium determined from the patient's lithium level, and Li is the actual determined patient's lithium level after h hours.

10. The system according to claim 9, wherein the coefficient is between -0.065 and -0.019.

11. The system according to any of the claims 7-10, wherein the coefficient p is a fixed value.

12. The system according to any of the claims 7-10, wherein the lithium level registration unit (16) further is configured to receive a patient's creatinine level (39) and wherein the coefficient p is determined based on a patient's creatinine level.

13. The system according to claim 12, wherein the database comprises a table wherein the coefficient p can be obtained based on the patient's creatinine level (39).

14. The system according to any of the preceding claims, wherein the dose registration unit is a patient registration unit (12) configured to receive, from the patient, the dose record (22) comprising the time the patient reports to have received the last lithium dose, and to transmit said dose record to be stored within the database (20).

15. The system according to claim 14, wherein the patient registration unit (12) is a smartphone wherein the patient enters the time that the patient received the last lithium dose via an app.

16. The system according to claim 15, wherein the system is configured to send reminder notifications via the app to the patient in relation to the time when the patient should receive a lithium dose.83894PC013317. The system according to any of the preceding claims, wherein the dose registration unit is a medical staff registration unit (14) configured to receive, from a medical staff member, the dose record (22) comprising a time the patient has received the last lithium dose, and to transmit said dose record to the system to be stored within the database (20).

18. The system according to any of the preceding claims, wherein the dose registration unit is a medicine container (15), configured to generate the dose record (22) comprising the time when the medicine container has dispensed a dose of lithium to the patient, and to transmit said dose record to the system to be stored within the database (20).

19. The system according to any of the preceding claims, wherein the system is further configured to create a prioritization of patients within said database (20) to have a blood sample taken applying i) whether the dose record (22) was received from a specific dose registration unit (12, 14, 15), and ii) the reported time since the last lithium dose was received by the patients.

20. A computer-implemented method for estimating 12-hours blood concentration of lithium, wherein the method comprises- registering by a dose registration unit (12, 14, 15), a dose record (22) comprising a time (26) a patient received a lithium dose and the amount (28) of the dose,- assigning a unique patient ID (24) to the dose record,- transmitting from the dose registration unit (12, 14, 15) the dose record (22) to a processing unit to be stored within a database (20),- storing the dose record (22) within the database,- obtaining a patient's lithium level (38) and the time (36) the patient's lithium level was determined from a lithium level registration unit,- performing an estimation calculation, applying the patient's lithium level, the time the patient's lithium level was determined, and the time the patient83894PC0134 received the lithium dose, to calculate an estimated 12-hours blood concentration of lithium (eLii?), and- outputting the estimated 12-hours blood concentration of lithium (eLii?).

Images