WO2011039676A2

WO2011039676A2 - Assessing patient compliance

Info

Publication number: WO2011039676A2
Application number: PCT/IB2010/054248
Authority: WO
Inventors: Mariana Simons-Nikolova; Johan Muskens; Aleksandra Tesanovic; Rob Theodorus Udink; Lennard Leonardus Petrus Josephus Maria Kuijten
Original assignee: Koninklijke Philips Electronics N.V.
Priority date: 2009-09-29
Filing date: 2010-09-21
Publication date: 2011-04-07
Also published as: WO2011039676A3

Abstract

There is provided a method of assessing a patient's compliance to an intervention specified by a healthcare professional, the method comprising determining a level of compliance for the patient from interactions between the patient and a remote patient management system.

Description

Assessing patient compl

TECHNICAL FIELD OF THE INVENTION

The invention relates to a method and system for assessing the compliance of a patient that uses a remote patient management system to a specified intervention. BACKGROUND TO THE INVENTION

Compliance describes the extent to which a patient follows recommendations or instructions from a physician, nurse, other health practitioner or a remote patient management (RPM) system in relation to a particular intervention, such as taking a medication, following a treatment regimen (which could include taking a regular medication, following a diet, performing particular exercises and/or making other lifestyle changes), taking regular or specified measurements of their physiological characteristics, filling out questionnaires or viewing educational videos or messages relating to the health issues of the patient. Compliance can also relate to whether the patient seeks further advice from a healthcare professional if their condition changes or worsens.

The failure of patients to adhere or comply with a specified intervention is a significant problem, and results in a suboptimal clinical benefit to the patient from the intervention. Failures can include missing doses of a medication, not taking regular measurements of blood pressure or blood sugar, eating too much salt, and so on.

In relation to patients with heart failure, it has been estimated that the compliance to specified interventions while the patient is in a specialty centre (i.e. a medically controlled and supervised environment) is very high (i.e. around 95%). However, once the patient is outside the direct control of the specialty centre (i.e. they are at home), the patient's compliance is poor.

The largest database of acute heart failure patients in the world, the Acute Decompensated Heart Failure National Registry (ADHERE), has data on the compliance of over 100,000 patients to heart failure-based interventions. According to the data in the ADHERE Registry, nearly half of all patients in the database are non-compliant with respect to medication and lifestyle therapy, while another fifth fail to seek advice or assistance in the event that their condition worsens. These figures have been supported by a number of other studies of patients with heart failure.

Non-compliance of a patient with heart failure with medications and diet and fluid restrictions decreases the efficacy of the treatment prescribed and exposes the patient to clinical destabilization, which can lead to increased heart failure related symptoms. Noncompliance has also been found to be a precipitating factor in heart failure exacerbation, leading to poor clinical outcomes such as more frequent physician visits, increased hospitalizations, a longer length stay in hospital, and therefore increased health costs.

At present, patient compliance can be determined by a healthcare professional during a patient visit. The professional asks the patient about their compliance with the medication or treatment regimen, and then estimates the patient compliance, factoring in the professional's estimate on how truthful the patient was in answering the questions. Most patients overestimate their level of compliance to a medication or treatment.

However, such an estimate of compliance is clearly subjective, and can only be done through direct contact between the professional and the patient. If the professional suspects or detects a low compliance level, it is only during the patient visit that the professional can take steps to improve the compliance. There is currently no way for a professional to detect problems with compliance between patient visits.

Alternatively, a patient with a remote patient management (RPM) system can be presented with a questionnaire having questions relating to their compliance with the intervention. However, many studies and statistics show that self-reporting tools like this often result in mis-reporting, particularly with respect to lifestyle, for example by

underestimating the consumption of sodium rich food and fluid intake whilst overestimating physical activity. In addition, many patients are only marginally motivated to invest time and energy in answering self-administrated questions.

Therefore, there is a need for a method and system that provides an alternative way for a patient's compliance with an intervention to be assessed, particularly where the intervention is supported by a remote patient management (RPM) system.

SUMMARY OF THE INVENTION

According to a first aspect of the invention, there is provided a method of assessing a patient's compliance to an intervention specified by a healthcare professional, the method comprising determining a level of compliance for the patient from interactions between the patient and a remote patient management system. According to a second aspect of the invention, there is provided a computer program product comprising computer program code that, when executed by a computer or processor is configured to cause the computer or processor to perform the method described above.

According to a third aspect of the invention, there is provided an apparatus for use in a remote patient management system, the apparatus being for assessing a patient's compliance to an intervention specified by a healthcare professional, the apparatus comprising an input for receiving information on interactions between the patient and a part of the remote patient management system, a computer program product as described above and a processor that is configured to use the received information and to execute the computer program code in the computer program product.

According to a fourth aspect of the invention, there is provided a remote patient management system comprising an apparatus as described above. BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the invention will now be described, by way of example only, with reference to the following drawings, in which:

Fig. 1 is a simplified block diagram of the hardware components in a remote patient management system in accordance with the invention;

Fig. 2 is a flow chart illustrating a method in accordance with the invention;

Fig. 3 is a block diagram illustrating the functional components of the remote patient management system of Fig. 1 in accordance with the invention;

Fig. 4 comprising Figs. 4A, 4B and 4C are a flow charts illustrating the operation of an interaction analyzer in accordance with the invention; and

Figs. 5, 6 and 7 show exemplary outputs of the interaction analyzer in accordance with the invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

In the following, the user of the remote patient management (RPM) system according to the invention whose compliance is being measured is referred to as a patient.

The use of the word "patient" or "patients" in the description or claims is not intended to limit the scope of the invention or to require any kind of doctor/patient relationship.

In addition, the compliance of a patient is assessed in relation to an

"intervention" specified by a healthcare professional and/or the remote patient management system and performed, at least in part, in connection with the remote patient management system, where the intervention can comprise any kind of activity, pharmacological or non- pharmacological treatment (or any combination thereof) that is intended to improve, or lead to improvements in, the health of the patient.

A simplified architecture of a remote patient management (RPM) system 2 is shown in Figure 1. The RPM system 2 generally comprises a patient end 4, a server back end 6 and a healthcare professional end 8.

The patient is provided with a patient terminal 10 (which can be a set-top box or a general purpose computer executing an RPM software program), one or more measuring devices 12 for measuring various physiological characteristics of the patient, and one or more feedback devices 14.

The measuring device(s) 12 present at the patient end 4 can be either selected based on the specific intervention to be overseen by the RPM system 2, or provided to monitor a wide range of different health problems or conditions by the RPM system 2. In this illustrated embodiment, the measuring devices 12 comprise weight scales 12a, a blood pressure sensor 12b, a heart rate sensor 12c and a blood sugar sensor 12d.

However, it will be appreciated that the measuring device(s) 12 can comprise sensors for measuring any type of physiological characteristic(s), including, but not limited to, any of blood pressure, heart rate, blood sugar levels, blood oxygen levels, breathing rate, heart function (including an electrocardiogram), bio -impedance, motion or weight. Those skilled in the art will be aware of other physiological characteristics of a patient that can be measured, and the appropriate measuring devices 12 for doing so. The measuring device(s) 12 can also include an activity monitor and/or a medication dispenser.

The one or more feedback devices 14, such as a TV, a PDA, a smart phone, a notebook or a PC, can comprise a display 16 and one or more user input devices 18, such as a keyboard, keypad, mouse, microphone, camera, touchscreen, etc. The display 16 can be used to present information to the patient, such as reminders or instructions to take a particular measurement or perform an exercise, information on the patient's progress, videos, quizzes, messages and/or questionnaires/surveys to be completed regarding the patient's current symptoms, and the user input devices 18 can be used to receive the patient's responses.

Each of the measuring devices 12 and feedback devices 14 are connected to the patient terminal 10. In this illustrated embodiment, the feedback devices 14 are connected to a bus 20 in the patient terminal 10, which is in turn connected to a processor 22 that manages the operation of the patient terminal 10. In alternative implementations, the feedback devices 14 can be connected to the patient terminal 10 in any other suitable manner, for example using individual wired connections, a wireless connection, or a point to point connection. The processor 22 can execute appropriate RPM software which is stored in a local memory 24.

The patient terminal is connected to an RPM system server 30 in the back end

6. The server 30 comprises a processor 32 and a memory 34 and manages the operation of each patient terminal 10 in the RPM system 2.

The RPM system server 30 is connected to a terminal 40 that is used by a healthcare professional. The terminal 40 comprises a processor 42, memory 44 and user interface 46. The terminal 40 executes RPM software that allows the healthcare professional to view data relating to the patient provided from the patient terminal 10, such as whether the patient is complying with the specified intervention or whether the symptoms or condition of the patient has changed, and to issue further instructions or interventions to the patient via the user interface 46.

Those skilled in the art will appreciate that only the elements of the RPM system 2 required to implement the invention are shown in Figure 1. It will also be appreciated that the RPM system 2 will be used by a large number of patients, each with their own patient terminal 10 measuring devices 12 and feedback devices 14, and a large number of healthcare professionals, each with their own terminal 40. Furthermore, it will be appreciated that the back end 6 may comprise a server farm comprising a plurality of individual servers, rather than just the single RPM system server 30 shown in Figure 1.

In accordance with the invention, the compliance of a patient to an intervention provided or monitored via an RPM system 2 is determined by tracking the interactions between the patient and the RPM system 2. In particular embodiments of the invention, an interaction tracker is provided that generates a log of the interactions between the patient and the RPM system 2, and an interaction analyzer transforms the data in the log into qualitative and/or quantitative data relating to the compliance of the patient with the intervention.

Figure 2 is a flow chart illustrating a method in accordance with the invention. In step 80, information on the interactions between the RPM system 2 and the patient are collected and stored in the RPM system 2. Then, in step 82, a level of compliance of the patient with the specified intervention is determined from the stored interactions. In a preferred embodiment, step 82 comprises comparing the information on the interactions to a schedule of interactions specified for the intervention. Figure 3 shows the functional components of the remote patient management system 2 of Figure 1 in accordance with the invention. At the patient end 4, information on interactions between the patient and the measuring device(s) 12 and feedback device(s) 14 is provided to an interaction tracker 50 in the patient terminal 10. The interaction tracker 50 can be a computer program that is executed by the processor 22 and stored in the memory 24.

The interaction tracker 50 stores the information relating to the interactions in an event logging database 52. The database 52 can be physically stored in the memory 24.

The data in the event logging database 52 is provided to an interaction analyzer 54 in the back end server 30. As with the interaction tracker 50, the interaction analyzer 54 can be a computer program that is executed by the processor 32 and stored in the memory 34. The interaction analyzer 54 makes use of three databases, an RPM database 56 (which stores the results of measurements by the measuring devices 12), a compliance database 58 and an intervention database 60 (which stores details of the specified

intervention or care plan), as will be described in more detail below. The three databases 56, 58, 60 can be physically stored in the memory 34.

Data from the compliance database 58 is provided to the remote healthcare professional's terminal.

Interaction tracker

The interaction tracker 50 logs the interactions between the patient and the RPM system 2 based on events from the measurement device(s) 12 and feedback device(s) 14 of the RPM system 2.

Each time the patient interacts with a measurement device 12 or feedback device 14 in the RPM system 2, an event is logged by the Interaction Tracker 50 in the event logging database 52. Each event in the database 52 is described by at least three attributes - date, time and an event descriptor.

Table 1 below lists some examples of event descriptors. The table is divided into two sub-categories, events logged from the measuring device(s) 12 and events logged from the feedback device(s) 14.

Those skilled in the art will understand the types of signals that can be output from the measurement device(s) 12 and feedback device(s) 14, and how these can be interpreted by the interaction tracker 50 to form the event data to be stored in the event logging database 52. Events logged from Event descriptor

measuring device(s)

Physiological • a physiological characteristic measurement taken characteristic

measurement

• weight

• blood pressure

• pulse

• Sp02

• bio-impedance

• glucose

Biomarkers • a biomarker measurement taken

• BNP

• Creatinine

• Potassium

• Sodium

Physical Activity • activity monitor data recorded

• activity monitor • activity monitor data read out

Medications • a pill (package/cup) dispensed

• medication dispenser • medication dispenser empty

• medication dispenser refilled

Events logged from Event descriptor

Feedback device(s)

Videos • Video v, v = 1, 2, ... available at the patient terminal

• Play video v

• Stop video v

• Pause video v

• Rewind forward/backward video v

• Mute on/off video v

Questionnaire/Survey • Quiz q, q = 1, 2, ... available at the patient terminal

• Start quiz q

• Finish quiz q

• Select answer to question k, k = 1, 2, ... from quiz q

• Change answer to question k, k = 1, 2, ... from quiz q

• Press "next question" from quiz q

• Press "previous question" from quiz q

Messages • Message m, m = 1, 2, ... available at the patient terminal

• Open message m

• Close message m

• Delete message m

Table 1 Interaction Analyzer

As described above, the interaction analyzer 54 uses the log of the interactions between the patient and the measuring device(s) 12 and feedback device(s) 14 to determine qualitative and/or quantitative data on the patient's compliance with the specified interaction.

In the illustrated embodiment, the interaction analyzer 54 combines the output of the interaction tracker 50 (i.e. the data in the event logging database 52) with data from the RPM database 56 and intervention database 60 to determine an estimate of the patient's compliance with the specified intervention.

The intervention or care plan database 60 stores information on the

intervention specified for the patient. For example, this information can comprise a measurement or medication schedule, rules on reminders that can be issued to the patient to perform a measurement or take some medication, or when the patient should be issued with or complete a survey or an educational video.

Table 2 below illustrates some exemplary inputs to the interaction analyzer 54 from each of the event logging database 52, RPM database 56 and intervention (care plan) database 60.

Input to interaction analyzer relating to interactions with measuring and feedback devices

Event logging database

Physiological

characteristic • output of interaction tracker

measurement

• weight RPM database

• blood pressure

• pulse • physiological characteristic measurements, for example 76

• Sp02 kg, 140/85 blood pressure, 76 beats per minute, etc.

• bio-impedance

• glucose • biomarker measurements, e.g. BNP = 135pg/ml_

• day and time of measurement, e.g., 9:05 on 24/01/2009

Biomarkers

• BNP • physical activity measurements on a given day, for example 3

• Creatinine hours of physical activity on 24/01/2009 (2h 35 min of low

• Potassium intensity and 25 min of medium intensity)

• Sodium

• medications dispensed on a given day, for example Nebivolol

1 x 5 mg, Lisinorpil 1 x 2.5 mg dispensed on 24/01/2009

Physical Activity

• activity monitor Intervention (Care Plan) database

• measurement schedule, for example "measurement m should

Medications be done daily by 12:00 pm"

• medication

dispenser • intervention rules for sending reminder "send a reminder if it is 12:00 pm and there has been no measurement in the last

10 hours"

Videos

• physical activity schedule (based on recommendations from the healthcare professionals), for example at least 30 minutes

Quizzes/Surveys of physical activity per day, on at least 5 days per week

• medications schedule (defined by healthcare professionals),

Messages e.g., Nebivolol 2 x 5 mg in the morning and in the evening,

Lisinorpil 1 x 2.5 mg, Furosemide 1 x 40mg, Spironolacton 1 x

12.5 mg

• video, quiz/survey and/or message schedule

• video, quiz/survey and message content

• video, quiz/survey and message content duration

Table 2 The interaction analyzer 54 can determine the patient's compliance with the specified care plan or intervention by comparing the interactions of the patient with the RPM system 2, as represented by the data stored in the event logging database 52 and RPM database 56, with the interactions expected according to the specified care plan or intervention, as represented by the data stored in the intervention database 60.

As described above, the interaction analyzer 54 can determine qualitative and quantitative data on the patient's compliance with the specified intervention. The qualitative data can comprise statements such as "measurement performed on time", "measurement not taken", "measurement performed with time deviation Δ triggered by a reminder", "watched video to completion" or "medication taken on time", for example, whereas the quantitative data can comprise a score for the patient's compliance or a percentage.

Figure 4 is a flow chart illustrating the operation of the interaction analyzer 54 in accordance with an embodiment of the invention.

In step 101 of Figure 4A, the interaction analyzer 54 extracts information on the intervention from the intervention database 60. Thus, the interaction analyzer 54 obtains information on the type of interactions required from the patient and the schedule according to which these interactions should take place.

In step 103, the interaction analyzer 54 extracts information on the patient's interactions with the RPM system 2 from the event logging database 52 and RPM database 56. Thus, the interaction analyzer 54 obtains information on the type of interactions that the patient has had with the RPM system 2, the timing of the interactions, and the result of the interactions.

In step 105, the interaction analyzer 54 determines whether there is an interaction scheduled in the intervention database 60. For example, if the intervention database 60 indicates that a weight measurement should be made by 12:00 pm, the interaction analyzer 54 determines from the information extracted in step 103 whether there has been a weight measurement around or after this time.

If the interaction analyzer 54 determines that no interaction has occurred, the interaction analyzer 54 sets the qualitative feedback for this scheduled interaction to

"interaction missed" (although it will be appreciated that the word "interaction" can be replaced by the name of the specific interaction that has been missed, for example "weight measurement"). In addition, or alternatively, the interaction analyzer 54 sets the quantitative feedback (also known as a compliance score) to 0 (represented by Mi = 0, where Mi is the interaction). This is shown in step 107. If the interaction analyzer 54 determines that an interaction has occurred, the interaction analyzer 54 moves to either or both of step 109 in Figure 4B or step 117 in Figure 4C, depending on the specific interaction being analyzed.

If the interaction was scheduled by a particular time, or period of time, such as a weight, a blood pressure or a pulse measurement, the interaction analyzer 54 performs the steps in Figure 4B. If the interaction comprised an activity that might be partially completed by the patient, such as watching a video, reading a message or text document, answering a survey or performing a set of exercises, the interaction analyzer 54 performs the steps in Figure 4C. It will be appreciated that if the interaction comprised an activity that might be partially completed by the patient and that was scheduled for a particular time, the interaction analyzer 54 can perform the steps in both Figures 4B and 4C.

In step 109 of Figure 4B, the interaction analyzer 54 determines if the interaction in the event logging database 52 was performed in accordance with the schedule in the intervention database 60. Thus, the interaction analyzer 54 compares the time that the interaction was completed with the time indicated in the schedule. If the interaction was completed at an appropriate time (for example within n minutes of the scheduled time), the interaction analyzer 54 can set the qualitative feedback for the scheduled interaction to "interaction on time" (although it will be appreciated that the word "interaction" can be replaced by the name of the specific interaction, for example "weight measurement"). In addition, or alternatively, the interaction analyzer 54 can set the quantitative feedback to 1 (represented by Mi = 1 , where Mi is the interaction). This is shown in step 111.

If the interaction was not performed in accordance with the schedule, the interaction analyzer 54 determines the amount of deviation Δ of the interaction in the event logging database 52 from the schedule (step 113). The interaction analyzer 54 can then set the qualitative feedback for this scheduled interaction to "interaction with time deviation Δ" (although it will be appreciated that the word "interaction" can be replaced by the name of the specific interaction that has been missed, for example "weight measurement"). In addition, or alternatively, the interaction analyzer 54 can set the quantitative feedback to 0.5 (represented by Mi = 0.5, where Mi is the interaction). This is shown in step 115.

Turning now to Figure 4C, the interaction analyzer 54 determines if the interaction in the event logging database 52 was completed fully. If the interaction was completed fully (for example the patient watched a complete video or completed all questions in a survey), the interaction analyzer 54 can set the qualitative feedback for the scheduled interaction to "interaction completed" (although it will be appreciated that the word "interaction" can be replaced by the name of the specific interaction, for example "video"). In addition, or alternatively, the interaction analyzer 54 can set the quantitative feedback to 1 (represented by Mi = 1, where Mi is the interaction). This is shown in step 119.

If the interaction was not completed fully, the interaction analyzer 54 determines the portion, p, of the interaction that was completed (step 121). The interaction analyzer 54 can then set the qualitative feedback for this scheduled interaction to "interaction p% completed" (although it will be appreciated that the word "interaction" can be replaced by the name of the specific interaction, for example "video"). In addition, or alternatively, the interaction analyzer 54 can set the quantitative feedback to O.p (represented by Mi = O.p, where Mi is the interaction). This is shown in step 123.

The feedback set in each of steps 107, 111, 115, 119 or 123 is stored in compliance database 58.

Once the interaction analyzer 54 has determined feedback on a number of events in the intervention database 60, the feedback can be collated and presented in a number of different ways, for example as a list of events with a corresponding compliance score, or in a table, chart or graph. Furthermore, the interaction analyzer 54 can arrange the feedback in a format that is suitable for viewing by a healthcare professional.

The interaction analyzer 54 can provide indications of the patient's

compliance on a day-by-day basis, but it is also possible for the interaction analyzer 54 to collate the feedback to provide an indication of the patient's compliance with particular interactions or the intervention over a longer period of time. This longer period of time is referred to herein as a 'sliding window', and it means that it is possible to view or extract trends or patterns in the patient's behavior (such as particular measurements that the patient is prone to miss) over time. The sliding window allows the compliance data accumulated over a period of time to be used to, for example, track the time at which physiological

measurements were made during a given week, month, etc., track the number of weight measurements that were on time (i.e. in accordance with the schedule) or the number of interactions that were prompted by a reminder from the RPM system 2.

In addition, by collating feedback for a number of different patients that are subject to the same or similar care plans or interventions, it is possible to determine whether patients as a whole are prone to miss (or intentionally skip) any specific events (such as watching a particular video).

Table 3 below sets out some exemplary outputs of the interaction analyzer 54, including specific qualitative and quantitative feedback for the various types of interactions that can be generated day-by-day, and also 'sliding window' feedback that can be used to view trends or patterns in the patient's behavior over a longer period of time.

Output of Interaction Analyzer

Qualitative data Quantitative data

Physiological

characteristic day-by-day day-by-day

measurement • missing M, measurement • Mi = 0

(Mi) • measurement M, on time • Mi = 1

• measurement Mi with deviation Δ • Mi = 0.5

• weight triggered by a reminder

• blood pressure

• pulse sliding window sliding window

• Sp02 • frequency of missing measurement in see Figure 4 and 5

• bio-impedance terms of daily, x-times per week, x-

• glucose times per month, etc.;

• deviation (in terms of time) from a pre¬

Biomarkers defined schedule for measuring a

measurement physiological characteristic;

(Mi) • frequency of deviation in terms of daily,

• BNP x-times per week, x-times per month,

• Creatinine etc.;

• Potassium • extracting patterns, e.g., usual time a

• Sodium patient takes measurement

day-by-day Day-by-day

Physical Activity

• Recommended amount of activity • delta = P PA - (R_PA) R_PA;

• activity monitor

• Performed amount of activity (P_PA) • Mi = 1 if delta ≥ 0 and 0 otherwise;

sliding window sliding window

• Number of monitored days (N) • Mi

• Recommended amount of activity per P_PA*100/(R_PA*N) day (R_PA)

• Performed amount of activity in N days

(P_PA)

day-by-day (per medication) Day-by-day

Medication

• Prescribed type of meds • M, = 1 or 0 (taking

• Dispensed type of meds compliance)

• medication

• Prescribed number of doses (P_N D) • Mi = D_N D/P_ND dispenser

• Dispensed number of doses (D_ND) (dosing compliance)

• Prescribed time of a dose (P_TD) • delta = |T TD -

• Dispensed time of a dose (D_TD) D_TD|

• Interval b/n two consecutive doses • Mi = 1 if delta = 0 and 0 otherwise (schedule compliance)

Table 3

Figures 5 to 7 illustrate some exemplary graphical outputs of the interaction analyzer 54.

Figure 5 is a graph showing the patient's compliance with various types of measurements (weight, blood pressure and pulse) made using the measuring devices 12 over the course of a week. The compliance is illustrated by the frequency with which the patient took the measurement on time, required a reminder to take the measurement or missed the measurement entirely. Thus, it can be seen, for example, that the patient made the weight measurement on time four times, but missed each of the blood pressure and pulse measurement three times.

Figure 6 is an alternative graph, using the same data set as Figure 5, showing the times at which various measurements are made on particular days of the week. The graph also shows the time at which a reminder is set to be issued, so it is possible to see how many measurements were made as a result of a reminder being issued. Thus, it can be seen, for example, that all three measurements were made on time on Monday and Tuesday, while all three measurements were late and required a reminder on Wednesday.

The quantitative data used to populate these graphs can also be used to generate an overall compliance score for the patient for the week. The compliance score data for the three measurements is set out in Table 4 below. Mo Tu We Th Fr Sa Su Compliance

Weight 1 1 0.5 1 1 0 0 4.5/7 = 0.65

Blood 1 1 0.5 0.5 0 0 0 3/7 = 0.43

pressure

Pulse 1 1 0.5 0.5 0 0 0 3/7 = 0.43

Table 4 Thus, the compliance score for the weight measurement is 4.5 out of a possible total of 7, giving a 65% compliance rate, while the compliance score for each of the blood pressure and pulse measurements is 3 out of 7, giving a 43% compliance rate.

Figure 7 is a graph showing how often particular videos (vl, ..., v8) are watched by the user. It can be seen that video 8 (v8) has been watched the most, while videos 5 (v5) and 7 (v7) have not been watched at all. Furthermore, the graph shows that video 3 (v3) has not been watched completely as the frequency is less than 1.

Table 6 below illustrates how an overall compliance score for the video part of the intervention can be determined from the quantitative data.

Table 5

In embodiments of the invention, the interaction analyzer 54 can account for periods of time in which the patient's compliance cannot be calculated (for example if the intervention is suspended while the patient is on holiday or if there is a problem with some element of the RPM system 2), so these periods do not adversely affect a patient's

compliance score or graph.

In embodiments of the invention, the interaction analyzer 54 can determine a score for the patient's compliance with a number of different interactions in the intervention using the following equation:

m

compliance , CS = ^ cci · Μ _| (1)

i=1

where Mi is the compliance score for the i-th interaction (for example representing whether a blood pressure measurement was taken, whether video vl was watched, etc.), m is the total number of interactions being considered (which can include multiple instances of the same type of interaction) and ¾ is a weighting factor.

The weighting factor ¾ is set for each particular type of interaction based on its importance in the overall intervention. For example, in an intervention designed to monitor a patient with hypertension, the weighting factor may be set high for measurements of the blood pressure and pulse rate, and relatively low for watching a video or for completing a survey.

The values of the weighting factor can be determined in one of two ways. The first way is to use a simple mathematical approach, such as averaging. The second way is based on statistical methods like factor and cluster analysis that generate patterns from collected intervention data during a clinical study.

For example, factor analysis, based on correlations between variables, tries to group and/or reduce a set of variables into a new and better interpretable set. The factor analysis calculates factor loading coefficients that can be used as weighting factors. Another technique is cluster analysis which is based on distances and (dis)similarities between objects and is mostly used to group the objects in a dataset. For a set of patients interacting with a RPM system the cluster analysis could give an answer to the question "Is there a sizeable number of patients (a cluster) who constantly score high on one compliance score Mi and low on another?" Identifying such clusters will imply which Mi are the most important by assigning them a higher weighting factor.

In some embodiments of the invention, the RPM system 2 can include a module for learning the patient's preferences for interacting with the RPM system 2 from the compliance scores or the interactions logged in the event logging database 52. The patient's preferences can relate to, for example, the timing of the daily measurements and the interaction with the coaching (i.e. video, audio and messaging) content. These preferences can either be learnt by the RPM system 2 through the use of the RPM system 2, or the preferences can be manually input by the patient or a healthcare professional.

It will be appreciated that a healthcare professional may use the output of the interaction analyzer 54 and compliance database 58 in a number of ways, for example, the healthcare professional can see the level to which the patient is complying with the intervention, the particular interactions that the patient is prone to miss or not complete fully, etc.

Where the healthcare professional is responsible for the care of a number of patients that each use the RPM system 2, the feedback from the interaction analyzer 54 can be used to highlight those patients that are least compliant with the specified intervention, and that therefore require further consultations with the healthcare professional or more closer monitoring. Thus, the output of the interaction analyzer 54 can be used to prioritize the patients that require the most attention of the healthcare professionals.

The output of the interaction analyzer 54 for a particular patient can be used by the healthcare professional to help assess the patient's behavioral profile (for example whether they are inclined to comply with advice or instructions from a healthcare

professional). This in itself can be used by the healthcare professional to select an

intervention that is more appropriate for that patient.

In some embodiments of the invention, the or an output of the interaction analyzer 54 can be provided to the patient terminal 10 so that the patient can directly see their level of compliance with the specified intervention. Providing the feedback (whether it is positive or negative) to the patient in this way can result in the RPM system 2 motivating the patient to improve their compliance.

Finally, the output of the interaction analyzer 54 can be used to assess the quality of specific content used in the intervention (such as videos, audio clips or messages) based on the compliance scores for a number of patients. If the compliance scores show that the patients often fail to finish a particular video, listen to all of a particular audio clip or read all of a particular message (for example if the message is opened and then closed in a very short space of time, e.g. < 2 seconds) then it might be that the content of that interaction needs to be revised to make it more engaging for the patient.

There is therefore described a method and system that allows a patient's compliance with an intervention to be assessed.

While the invention has been illustrated and described in detail in the drawings and foregoing description, such illustration and description are to be considered illustrative or exemplary and not restrictive; the invention is not limited to the disclosed embodiments.

Variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality. A single processor or other unit may fulfill the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. A computer program may be stored/distributed on a suitable medium, such as an optical storage medium or a solid-state medium supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the Internet or other wired or wireless

telecommunication systems. Any reference signs in the claims should not be construed as limiting the scope.

Claims

CLAIMS:

1. A method of assessing a patient's compliance to an intervention specified by a healthcare professional, the method comprising:

determining a level of compliance (82) for the patient from interactions between the patient and a remote patient management system (2).

2. A method as claimed in claim 1, wherein the intervention comprises one or more specified interactions between the patient and the remote patient management system (2)·

3. A method as claimed in claim 2, wherein the interactions comprise interactions between the patient and measurement and/or feedback devices (12, 14) at a patient end of the remote patient management system (2).

4. A method as claimed in claim 1, 2 or 3, further comprising the step of:

storing information (80) on interactions between the patient and the remote patient management system (2) in a database (52, 56) in the remote patient management system (2).

5. A method as claimed in claim 4, wherein the step of storing information (80) comprises storing information on what interactions between the patient and the remote patient management system (2) occurred and when the interactions occurred.

6. A method as claimed in claim 5, wherein the intervention comprises a schedule of one or more specified interactions, and the step of determining a level of compliance (82) comprises comparing when the interactions between the patient and the remote patient management system (2) occurred to the schedule of specified interactions.

7. A method as claimed in claim 6, wherein the step of determining a level of compliance (82) comprises determining a compliance score for each specified interaction from the difference between the schedule and when a corresponding interaction occurred.

8. A method as claimed in claim 6 or 7, wherein the step of storing (80) further comprises storing information on the extent to which the patient completed the specified interaction, and the step of determining a compliance score (82) comprises determining a compliance score based on the extent to which the patient completed the specified interaction.

9. A method as claimed in any preceding claim, further comprising the step of:

providing an indication of the level of compliance of the patient to a healthcare professional end of the remote patient management system (2).

10. A method as claimed in any preceding claim, wherein the interactions comprise any one or more of taking a measurement of a physiological characteristic or biomarkers of the patient, monitoring a specified physical activity, playing a video or audio clip, displaying messages or completing a questionnaire or survey.

11. A method as claimed in any preceding claim, wherein the intervention comprises any kind of pharmacological or non-pharmacological treatment, or combination thereof, that is intended to improve or lead to improvements in the health of the patient.

12. A computer program product, comprising computer program code that, when executed by a computer or processor is configured to cause the computer or processor to perform the method as claimed in any of claims 1 to 11.

13. An apparatus (30) for use in a remote patient management system (2), the apparatus (30) being for assessing a patient's compliance to an intervention specified by a healthcare professional, the apparatus (30) comprising:

an input for receiving information on interactions between the patient and a part of the remote patient management system (2);

a computer program product (34) as claimed in claim 12; and a processor (32) that is configured to use the received information and to execute the computer program code in the computer program product (34).

14. An apparatus (30) as claimed in claim 13, wherein the apparatus (30) is a server for the remote patient management system (2).

15. A remote patient management system (2) comprising an apparatus (30) as claimed in claim 13 or 14.