Abstract
Introduction
Intermediate telehealth devices can be defined as connectors between entities of telehealth systems. 1
They fall into two classes. The first class is adapted devices,which are those originally designed for other uses and have been adapted for telehealth systems, usually via integrating software into the device. These include mobile phones and personal computers. The second class is proprietary devices, which in contrast are those specifically designed for use in a telehealth system. 1
From a systems perspective, there is evidence that intermediate devices are the key invariant component in a telehealth system. A particular telehealth offering can be used for multiple diseases and contain varied types of actors, but its intermediate device or devices are usually fixed: possessing particular sets of capabilities. 1 For this reason, their capabilities form an important area of study.
Previous Case Studies
Some descriptions of the use of individual intermediate devices already exist in the literature. For example, Luptak et al. 2 describe a study in which patient feedback was obtained on intermediate device used in daily chronic care.
Case-study evidence is also emerging that such devices can form part of systems that have tangible medical impact. Several primary care trusts within the United Kingdom have implemented telehealth systems incorporating intermediate devices. Examples include systems for patients with heart failure and devices in systems to monitor chronic obstructive pulmonary disease (COPD). 3 Device use in the latter has helped community nurses to prioritize their work, resulting in a decrease by 80% of home visits. 3
The Perspective of Healthcare Providers
However, there has been no systematic study of the capabilities of current intermediate devices or of how those capabilities are clustered on current devices.
From the perspective of those considering implementing telehealth systems for chronic care, this means that it is difficult to know whether their expectations could be met with current systems. For example, the answer to a question such as “can current telehealth systems monitor multiple diseases and if so what range of diseases can be covered” is dependent on whether the intermediate device of a system will connect with and process data from multiple point-of-care devices. Questions on how patients are monitored, prompted, and advised are also dependant on the capabilities of intermediate devices. The same is true for questions of data security and confidentiality, key information capture, and system usability.
Research Questions
This article assesses currently available intermediate devices in telehealth systems.
It initially seeks to do two things: first, to understand the capabilities of such devices, and second, to show how such capabilities are clustered on the current generation of devices.
Using this information, it then seeks to summarize the capabilities of the devices as seen from an actor-centric viewpoint, that is, the viewpoint of a potential patient, caregiver, health professional, or payer.
Methods
Criteria for Selecting Devices
As little information is currently available on adapted devices, this study was limited to proprietary devices. It was also limited to the devices that met the following criteria: 1. Devices that have a degree of certification, that is, Food and Drug Administration (FDA) 510(k) or CE mark. 2. Devices that have been designed primarily for a patient rather than a health professional. 3. Devices that can be used outside the hospital environment.
Using these criteria, a total of 14 intermediate devices were analyzed. Table 1 lists the devices evaluated, the organization using or distributing them, the type of certification, and appropriate references.
Proprietary Intermediate Devices with Data Display Capabilities
Data Capture
Proprietary devices available between November 2009 and April 2010 were analyzed. Company Web sites and brochures were assessed to obtain information about the proprietary devices. Where there were insufficient details, the organizations were contacted requesting additional information, and where necessary, a visit to the organization was made.
Evaluation Questions
A set of questions were developed to evaluate proprietary intermediate devices.
The questions correspond to four categories.
Setup/configuration: what is the required process to setup the intermediate device? (Table 2).
Criteria Used to Assess Proprietary Intermediate Devices (Setup/Configuration)
Available features: what are the features available on the intermediate device? (Table 3).
Criteria Used to Assess Proprietary Intermediate Devices (Available Features)
Inputs: how does the intermediate device receive its inputs? (Table 4).
Criteria Used to Assess Proprietary Intermediate Devices (Inputs)
Outputs: what are the outputs produced by the intermediate device? (Table 5).
Criteria Used to Assess Proprietary Intermediate Devices (Outputs)
Clustering
To be able to see how capabilities are clustered in currently available intermediate devices, the clusters have been presented in the article in tabulated form (Tables 6 –19).
Availability of Proprietary Intermediate Devices
V, variable, depending on request.
Environments in Which Proprietary Intermediate Devices Can Be Used
Chronic Conditions Managed by Proprietary Intermediate Devices
In this table, where the chronic condition is shown as unspecified, the organization did not describe the condition.
COPD, chronic obstructive pulmonary disease.
Types of Point-of-Care Devices Providing Inputs for Proprietary Intermediate Devices
Proprietary Intermediate Devices Designed for Multiple Users
Prompts Available on Proprietary Intermediate Devices
Alerts to Patients on Proprietary Intermediate Devices
Advice Provided to Users on Proprietary Intermediate Devices
Automatic or Manual Results Input for Proprietary Intermediate Devices
Proprietary Intermediate Device Inputs
Wired Information Transfer Technology Between Point-of-Care Devices and Proprietary Intermediate Devices
USB, Universal Serial Bus.
Wireless Information Transfer Technology Between Point-of-Care Devices and Proprietary Intermediate Devices
Information Transfer Technology Between Proprietary Intermediate Devices and Other Components of Telehealth Systems
GPRS, General Packet Radio Service; GSM, Global System for Mobile Communications; M2M, mobile to mobile.
Data Transfer Destination for Intermediate Devices
For each cluster table, the horizontal axis shows the device number in the study. The vertical axis shows the capabilities. Where a capability is present, a tick is shown in the table; if absent, it is shaded in dark gray so that the clustering can be easily seen visually. Where the data were unobtainable, the corresponding position in the table is left blank. Where additional coding of the data is required, this coding is shown in a key below the table.
Results
This section describes the findings of the analysis of proprietary intermediate devices based on the questions detailed in Tables 2 –5. For each dataset, where the data percentages do not add up to 100%, the remaining data were unobtainable in this study.
Within each section of the results, there are two parts. The first part concerns individual capabilities of the devices: how often they are present and what is their nature. The second concerns how these individual capabilities are clustered on current devices.
Availability
Ninety-three percent of intermediate devices are purchasable from a single provider, and 7% (device 5) are available from multiple providers (three suppliers in total).
As shown in Table 6, much less information was available from manufacturers on devices provided within a system with 79% of data being unavailable. Twenty-one percent of devices are sold for use within a system package. Within this group, one device (device 8) can also be bought off the shelf. Device 5, available from multiple suppliers, is only sold for use within a system package.
Cost information was generally unavailable from the manufacturers. Manufacturers indicated that costs varied with quantity and with user requirements.
Setup
Devices require neither explicit configuration by the patient nor establishment of any online account (though information was unavailable in 21% and 29% of cases, respectively).
Environments
Fourteen percent of intermediate devices can be used in both home and hospital— 50% are exclusively for home use and 36% can be used in any location.
Table 7 shows the clustering of intended device environments. All devices that were usable outside the home and hospital environment were also designed for use in that environment.
Multiple Conditions and Point-of-Care Connection Options
All of the intermediate devices evaluated can be customized to manage several health conditions including asthma, COPD, diabetes, heart disease, and hypertension. Twenty-one percent of devices were said to be used for managing unspecified “chronic conditions.”
Table 8 shows the clustering of conditions by device. The majority of devices are used for a cluster of four diseases: COPD, diabetes, heart disease, and hypertension. In contrast, asthma is much less well served.
Ninety-three percent of intermediate devices evaluated allow multiple point-of-care devices to be connected to them. The remaining 7% of intermediate devices obtain their data from another intermediate device. Point-of-care devices include glucometers, weighing scales, blood pressure (BP) monitors, pulse oximeters, spirometers, and peak flow meters.
Table 9 shows the different types of point-of-care devices providing inputs to intermediate devices. The most popular were BP monitor, weighing scale, glucometer, and pulse oximeter. The less popular ones were peak flow meter and spirometer. The coagulometer and urine analyzer were least represented.
Multiple Users
Thirty-six percent of intermediate devices are designed for the use of multiple users and 64% by only a single user (Table 10).
Prompts, Reminders, and Confirmations
Voice prompts are available on 64% of devices and 29% do not have this feature.
Eighty-six percent of intermediate devices provide reminders to users and 7% do not.
In cases where users enter data or select an option on the intermediate device, 36% display a confirmation to the user of the selected option and 21% do not.
Sixty-four percent of intermediate devices ask the user symptomatic questions, whereas 29% do not.
Fifty percent of intermediate devices display a confirmation to the user that their results have been transferred to the health professional, whereas 21% do not. This confirmation took the form of a progress bar or a message displayed to the user.
Table 11 shows the clustering of the prompts capabilities of intermediate devices.
Alerts
Alerts are provided directly to patients in 29% of intermediate devices (Table 12), whereas 64% of devices do not provide alerts to patients. In each case, this is in the form of messages displayed on the intermediate device informing the user that their result is not within the expected target range.
Advice and Communication with the Health Professional
Fifty-seven percent of intermediate devices provide instructions to the user as to how to carry out a particular test, whereas 21% do not. In one device (device 5), training videos are provided for the user.
No device was found to have the capability for patients to send additional messages to their doctor (though information was unavailable for 29% of cases).
One device (device 12) allows an alarm call to be made to the doctor.
Health professionals are alerted when the patients' results are outside specified thresholds. (However, none of these alerts are through the intermediate device.)
Twenty-nine percent of intermediate devices enable patients to receive messages from the health professionals compared with 42% that do not have this feature. In one of the devices (device 5), a two-way video call can be organized by the health professional to discuss issues with the patients. Table 13 shows the device capabilities for advice and communication.
Historical Data
Patients can access historical data on the device in 36% of intermediate devices (devices 1, 5, 6, 8, and 13) compared with 43% that do not provide this feature.
Authentication
Twenty-nine percent of intermediate devices (devices 1, 5, 8, and 13) require authentication by the users, although this feature requires activation; 50% do not require this.
Automatic or Manual Results Input
Results of point-of-care tests are automatically transferred from the point-of-care device to the intermediate devices in 50% of cases. Results are manually entered in 7% of devices. In 29% of devices, results transfer is both automatic and manual.
Table 14 shows the clustering of results transfer based on mode of input.
Device Inputs
The inputs for the intermediate devices are either data from point-of-care devices and/or responses to questionnaires to help assess the symptoms of the patient. Seventy-nine percent of intermediate devices allow patients to interact with them through buttons, whereas 14% had a touch screen interface.
Table 15 shows the availability of these features.
Information Transfer Technology
To transfer results to the intermediate device from a point-of-care device, 64% of devices provide both wired and wireless connections, whereas 21% provide only a wireless connection. The wireless connections include Bluetooth, General Packet Radio Service (GPRS), infrared, radio, Global System for Mobile Communications, mobile-to-mobile cellular GPS, and Wi-Fi. The wired connections are through Universal Serial Bus (USB) or serial cables.
Table 16 shows the details of the wired connections from the point-of-care device to intermediate device, with the serial port being the most popular.
Table 17 shows the details of the wireless connections from the point-of-care device to the intermediate device, with Bluetooth being the most widely used.
Table 18 shows the information transfer technology in operation between the devices and other components of telehealth systems. As can be seen, few providers made this information available. There is no pattern in the types of information transfer technology used, although GPRS was used by three devices (2, 8, and 9).
Data Transfer Destination
Results and responses to the questions are transferred to a software application (29%), a secure server (57%), or another intermediate device (7%).
Table 19 shows the breakdown by device.
Discussion
In this section, the key capabilities and their variance on current proprietary intermediate devices are summarized. Capabilities of proprietary devices are then considered based on actor-centric perspectives, that is, those of the patient, health professional, and payer and system provider. It should be noted that a caregiver perspective is not given, as no interactions of the devices with caregivers were found.
Key Capabilities of Current Proprietary Intermediate Devices
Three main types of capability were identified from the current analysis. (1) To display information to patients: these are reminders, instructions of how to conduct tests, symptomatic questions and confirmation of their entry, or the successful transmission of results to another entity. Most proprietary devices provided reminders and asked the patient questions. Just over half of proprietary devices showed users how to do their test. Half of proprietary devices showed a confirmation of results transmission, whereas few confirmed the user entry. (2) To act as a channel that receives results automatically from point-of-care devices or manually via data entry by the patient. Half of devices allowed automatic input of results, whereas few allowed manual results entry. (3) To act as a channel for the forwarding of results and questionnaire responses to another entity. This may be another intermediate device, external server, or software application. Most proprietary devices transferred results to an external server.
Provider Perspective
By definition, all the devices evaluated in this study were for use in the home. In a study by Hopp et al. 28 showing the management of diabetes, the devices were only used within the home environment; however, providers consider multiple environments to be a market, as more than one third of devices were also suitable for hospital use.
Devices are clearly provider specific, with only a single device being available from multiple suppliers. From the limited information available, providers appear keen on selling a whole system package; only one device is available off the shelf. The state-of-the-art is therefore that intermediate devices are usually part of closed proprietary systems.
All of the intermediate devices evaluated can be customized to manage several health conditions, with the majority of devices usable for a cluster of four diseases: COPD, diabetes, heart disease, and hypertension. Hence, nearly all intermediate devices are designed for connection to more than one point-of-care device. Point-of-care devices that can be connected include glucometers, weighing scales, BP monitors, pulse oximeters, spirometers, and peak flow meters.
This requirement for point-of-care device connectivity leads to information being widely available from suppliers on information technology for data transfer from point-of-care device to intermediate device. Nearly all intermediate devices allow transfer using both wired and wireless media. In contrast, little data are available as to how the intermediate device connects to rest of the proprietary system.
Interestingly, the multiple connectivity features of intermediate devices, for use to connect with point-of-care devices, contrast with current point-of-care devices. Results from a recent study focusing on glucometers showed that only 51% allowed the data to be transferred. Of these, the majority (90%) used a USB cable, whereas 10% used infrared connection. 29 None used any other wireless technology.
Providers seem divided as to whether single- or multiple-user intermediate devices will gain them market share, with only a third of devices being multi-user capable. Aside from the opportunity to sell more devices, if each device is only for a single user, providers may consider that this device limitation reduces requirements for setup and authentication and helps with data security for the patient.
Providers also appear to be currently divided as to the necessity of either voice or video technology. Voice prompts were present on over half the intermediate devices, but video technology was available on only one. The case for the utility of video technology in intermediate devices seems yet to be made.
In contrast, providers are essentially unanimous in device designs, which excludes the initiation of direct communication with a health professional by the patient.
Patient Perspective
Providers appear keen to minimize the information technology knowledge required of the patient. For the patient, this translates to simple setup and authentication procedures (few devices had any authentication) and automatic data transfer from the point-of-care device to the intermediate device. Perhaps this is also a reason why only a third of devices allow patients access to historical data.
Less than a third of devices enable patients to receive messages from the health professionals, with one device allowing a two-way video call initiated by the health professional.
With regard to device interaction with patients, nearly all devices provide reminders to patients; over two thirds are capable of quizzing patients about their symptoms. The majority of devices also provided test instructions. In contrast, the figure for point-of-care devices (glucometers) for reminders is 19%. 29 This difference indicates where intermediate devices can bring benefits in patient independence and compliance above those available from the current generation of point-of-care devices.
In contrast, few intermediate devices are designed to provide alerts, for example, following an unexpected result. An anecdotal reason, given by some suppliers, is to avoid raising false alarms. However, this compares with again taking glucometers as examples of point-of-care devices to 77% that do provide alerts. 29
Certain chronic diseases are risk factors for others, for example, diabetes for cardiovascular disease, cerebrovascular disease, and renal disease. 30 Therefore, it is a clear benefit from the patient's perspective that one intermediate device can be potentially used with several chronic conditions.
Health Professional Perspective
Unlike for the case of the patient, intermediate devices have little direct contact with the health professional: the health professional sees aspects of the system rather than the intermediate device.
The main role of health professionals in relation to the intermediate device is that of selecting the types of questionnaires that will be used to regularly assess the patient's health. This role of health professionals is echoed by Hopp et al., 28 where questions are specifically configured for patients. Hopp et al. 28 added that health professionals trained patients in how to use the device to collate their responses. The aspect of training patients was not investigated in this article.
Payer Perspective
The detail of cost of devices was generally unavailable and system providers indicated that the variability of the cost of systems is based on user requirements. As devices are provided as part of a telehealth solution, it is usually the cost of the system that the payer sees.
Providers see multiple chronic disease capabilities as a selling point, whereas anecdotal evidence from one provider indicates that payers are initially interested in buying provision for a single chronic disease.
Limitations
To predict the directions of future development of proprietary devices, an extensive survey would need to be carried out to ascertain the expectations of the actors as to the additional features they need.
There is very limited data and analysis available in the literature on proprietary intermediate devices. Hence, it is difficult to compare and contrast the data and analysis of this article with the work of other researchers.
Conclusions
Proprietary intermediate telehealth devices are those that are specifically designed as connectors between the entities of telehealth systems.
In this article, their capabilities have been analyzed in current telehealth systems for the management of chronic diseases.
From the perspective of the different actors concerned with telehealth systems, the following can be concluded:
Provider perspective: Intermediate devices are usually part of closed proprietary systems, that is, they are provider specific and normally sold as part of a whole system package. Providers produce disease-customizable devices, capable of connection to more than one point-of-care device via wired and multiple wireless media; their connectivity is considerably ahead of the current generation of point-of-care devices. In contrast, little data are available from providers on how the intermediate device connects to rest of the proprietary system. Providers currently design the majority of devices for a single user.
Patient perspective: Simple setup and authentication procedures (few devices had any authentication) and automatic data transfer from the point-of-care device to the intermediate device seem key aspects of device design for patients. The majority of devices provide reminders to patients, quiz patients about their symptoms, and provide test instructions; this contrasts with current point-of-care devices. However, few intermediate devices are designed to provide alerts, again in contrast to point-of-care devices. For the patient, there is a clear benefit that one intermediate device can be potentially used with several chronic conditions.
Health professional perspective: Proprietary intermediate devices have little direct interaction with the health professional; their current main role is restricted to selecting the types of questionnaires that will be used to assess the patient's health.
Payer perspective: The details of costs of devices were generally unavailable and system providers indicated that the variability of the cost of systems is based on user requirements. As devices are provided as part of a telehealth solution, it is usually the cost of the system that the payer sees.
Footnotes
Acknowledgments
The authors acknowledge the EPSRC and Cranfield IMRC for funding this research.
Disclosure Statement
No competing financial interests exist.
