The design of a patient consulting system in managed clinical networks

(1)

RIJKSUNIVERSITEIT GRONINGEN

The design of a patient consulting system in

managed clinical networks

Master’s thesis

Supply Chain Management

E. Tiggelaar, S1904280 Supervisor: H. Broekhuis 2nd_{supervisor: D.P. van Donk} Contact person field of study: H.W. Nijman

(2)

1

Abstract

Motivation and research question: In managed clinical networks (MCN’s), care is organized in a way in which patients have to visit multiple care providers with different disciplines in often different care facilities. From the diagnosis until the aftercare, patients may want to ask questions, give feedback or communicate other messages. This research focuses on designing a system that can facilitate two-way communication between patient and care provider in the whole network, i.e. a patient consulting system. To design such a system, goals and requirements have to be known. Identifying these will provide more knowledge on how patients and care providers would like to experience the care provision.

Method: Literature on the process of communicating messages in a way that motivates the receiver to undertake action was used to compose a list with goals and requirements. By conducting interviews with patients and care providers, based on what was found in literature, goals and design preferences for a patient consulting system were identified. Besides, comparable existing systems were identified and described on their advantages, disadvantages and the extent to which they comply with the goals and requirements stated by patients and care providers. Experts were interviewed to determine the feasibility of certain systems in practice. Two possible designs were then developed. In a second loop of interviews was investigated to what extent the goals and requirements found, were met with these designs.

Results and design: The result of the first round of interviews was a list with goals and requirements for the system of both the patients and care providers. Surprisingly, only two of the requirements set by patients could, depending on how they are implemented in the design, be conflicting with those of care providers. Taking all the information together, two basic designs were developed. One design is an email system, where a patient can send an email whenever they want, containing any message, to a care provider. This design can easily be extended to meet more requirements. The other system is a webcam system where a patient can have a consult from home via the webcam instead of travelling to the hospital. This system is only available to patients when the care providers are available. Both systems make it easier for a patient to reach any care provider in the MCN.

(3)

2

Preface

(4)

3

1. Introduction

Getting diagnosed with cancer has a huge impact on the life of patients. A time of insecurities and a lot of worries commences. The holistic needs of patients, so not only physical, but also social and psychological needs, may vary in different stages of the provided care. To ensure that care is provided according to the variety and severity of the needs of patients, it is important that those needs are known (Waller et al., 2010). However, care providers often do not know exactly what the needs and questions of patients are. If care is provided in a network in which different care providers on often different locations are responsible for the different phases of the care process, it becomes even more difficult to have knowledge about patient-specific needs. In such a network, contact periods between a care provider and a patient are shorter. This means that there is less time to build a relationship between patient and care provider, which may complicate the sharing of information. Patients may not feel comfortable to express feedback to a care provider they just met. However, expressing needs or feedback will give the care provider knowledge on what to improve on the provided care. Also, because of all the insecurities during their treatment, patients may want to ask questions to their care provider. However, patients may be hesitant to ask questions to a care provider they do not know well or may not be aware to whom they can ask questions. Patients may therefore worry a long time about issues. Asking questions earlier may lead to some peace of mind for the patients. This research focuses on finding goals and requirements to design an patient consulting system in managed clinical networks, in which patients can have contact with their care providers, when they want to, to express all the things they want. This system could facilitate better care to the patients, by providing care that is adjusted to the needs of patients and by taking some questions and worries away.

In managed clinical networks (MCN’s) care is organized in a way in which different care providers work together on a common goal, while keeping their own tasks and responsibilities (Rosendal, Ahaus & Huijsman, 2009). The different care providers can be located in one care facility, but it is also possible that they are located in different care facilities. One of the important goals in MCN’s is providing patient centered care, i.e. adjusting the provided care to what patients value as important (De Booys, 2009; Singer, Burgers, Friedberg, Rosenthal, Leape & Schneider, 2011). The needs and the type and number of questions can for example vary during the treatment process. Schildmann, Ritter, Salloch, Uhl & Vollmann (2013) found for example that patients that were diagnosed with pancreatic cancer had different information needs in different stages of the treatment. Right after cancer was diagnosed, patients were not interested in detailed information and left decision making to the physician. Later in the treatment, they were more interested in their illness and started gathering information about it. Because of these changing needs, it is important that such a patient consulting system is ongoing and accessible in multiple moments in time.

(7)

6

factor makes that patients share their needs, feedback or questions with a care provider? And which factors make that care providers respond to the information that patients share? The aim of this research is to investigate what these design requirements are and translate the requirements into a design.

This research will contribute to the literature by providing more insights in the requirements for a patient consulting system of both patients and care providers and how this best can be translated in a design. It will be investigated under which conditions patients are willing to communicate a message and under which conditions care providers are motivated to use the message to improve the care they provide. To the best of the authors’ knowledge, an overview of these requirements is not available in literature. This research can therefore contribute to and provide insights on patients’ needs to share information during a difficult, sometimes life threatening, treatment. This research also provides more insights in the role of care providers by investigating what they find important in such a system and in care in general. This information is useful when designing care paths or facilities in which the patient should be involved to provide their opinion on the delivered care. This research will take place in the gynecological oncological care center in the university hospital in Groningen (hereafter: UMCG), but obtained insights in this care field can be used in other care fields, with the same type of population and type of treatment, as well. Patients in the gynecological, oncological care are mostly women of 60 years and older. Women in the same age class, who also face a difficult treatment, will probably have a similar point of reference and will therefore, most likely, have the same requirements. Since the care in the gynecological, oncological care is organized in MCN’s, also similarities in the requirements of other care fields that are organized in a MCN are most likely to be found.

(8)

7

2. Theory

This research aims to get an understanding of the requirements of a patient consulting system, which will be translated into a design. With this system, a patient will be able to communicate all information a patient wants to share with a care provider, for example expressing a need or a question. Both the design, so the way how patients can provide feedback, and the content, so the topics discussed, should comply with the wishes and needs of both patients and care providers in order for them to use the system (Brown, Sitzia, Richardson, Hughes, Hannon & Oakley, 2001).The goal of this system is to give patients the opportunity to communicate with their care provider outside the regular contact moments. Possible requirements for this system of both patients and care providers will be identified in this section. These requirements will be based on the goals stakeholders are expected to have regarding this patient consulting system (figure 2.1).

Figure 2.1: Relationship between goals, requirements and the final system

Mulder & Ellinger (2013) developed a model that depicts the feedback process. Although it is designed for feedback, it can also be used for other forms of communication in which a response from the receiver is required. The model of Mulder & Ellinger (2013) is focused on the feedback provision of a manager towards an employee. This model is therefore adjusted to fit the situation in which a patient communicates a message to a care provider, by removing the characteristics of the organization influencing the quality of feedback and by adjusting the outcomes (see figure 2.2 for the adjusted model). According to this model, the perceived quality of the feedback by receivers is determined by the content of the feedback, the timing of the delivery, the way it is delivered and the credibility of the sender. The perception of quality is influenced by the characteristics of the feedback receiver. Feedback can create cognitive, motivational, affective and behavioral responses. Since the goal of the feedback provision is to change the care provision, the focus will be on creating behavioral responses. However, since a behavioral response is influenced by cognitive, motivational and affective responses, these types of responses should also be taken into account. The outcome will be a changed care provision for the patient. This section will start with the intended outcome and then reflect back on how this can be accomplished. Possible goals and requirements of the system will be based on literature. The characteristics of the feedback receiver will be accounted for when collecting data, by interviewing different care providers with different characteristics. This will therefore not be treated in this chapter.

Figure 2.2: The feedback process (adjusted from Mulder & Ellinger, 2013)

Perceived quality of feedback Content

Timing Delivery mode

Credibility of the sender

Feedback receivers’ characteristics

Norms, attitude, knowledge

Receivers’ reactions Cognitive Motivational Affective Behavioral Outcomes

(9)

8

2.1 Receivers’ reactions

The desired effect of communicating a message would be that the care provider answers a question or improves its way of working and adjusts the care provision to the needs of patients. This can be classified as a behavioral reaction. How can a message lead to a behavioral reaction? Hattie & Timperley (2007) found that praise and positive feedback did not lead to a big change in the behavior of students. This can be explained by the fact that praising confirms that the feedback receiver is behaving in a good way and does therefore not have to change his behavior. However, care providers need confirmation, to know that they provide care according to the patients’ wishes. So, although positive feedback does not lead to a change in performance, the system should be able to record both positive and negative messages. There are four types of reactions to (negative) messages: acceptance, so willingness to receive the feedback, partial acceptance (‘yes, but…’ responses), rejection and disregarding the message (Laforest, 2002). Feedback and other messages can only be used for improvement if a care provider accepts it (Gabelica, van den Bossche, Segers & Gijselaers, 2012). According to Mulder & Ellinger (2013), acceptance of a message depends on the quality of the message. The content of the message, the timing and the mode of delivery influence the level of quality (Mulder & Ellinger, 2013). Other authors support this finding, but also add another dimension, namely the perceived credibility of the source (Bing-You, Paterson & Levine, 1997; Ilgen, Fisher & Taylor, 1979; Sargeant, Mann & Ferrier, 2005; Steelman & Rutkowski, 2004). It can be argued that the level of credibility determines the perceived level of quality. If the sender is perceived not to be credible, the perceived quality tends to be lower. This dimension will therefore also be discussed as a factor influencing quality.

The goal of the patient consulting system is therefore to deliver a message to the care providers that is perceived to be of high quality.

2.2 Quality of feedback

The perceived quality level of feedback will determine if feedback will be used for improvement. As argued before, the quality is determined by the content, timing, mode of delivery of feedback and the credibility of feedback provider. These factors will be discussed below.

2.2.1 Content

(10)

9

In the case of managed clinical networks, a message sometimes has to motivate multiple care providers. The question or feedback could namely be intended for an individual care provider, a couple of care providers or all care providers in the network. A patient consulting system should enable care providers to understand the specific needs and the message of patients (Richardson et al., 2007). The content of the message should therefore be clear. A misunderstanding of a message may lead to poor health care provision (Brown et al., 2001). In a MCN, since a message can be intended for multiple care providers, it is even more important that the message is clear.

In order to provide better patient-centered care – one of the goals in MCN’s -, patients should be able to express all of their needs and questions. However, it should be ensured that the needs or questions are care oriented, to avoid that patients communicate irrelevant messages to the care providers. Patients’ needs might also change over time, for instance after the surgery or they might get more questions in a later stage in the treatment. Therefore, providing feedback and asking questions should be an ongoing process, whereby patients can communicate a message at any point in time, whenever they want. Extra attention to these messages should be paid at key moments in the treatment process, such as the time of diagnosis, surgery or completion of the treatment (Richardson et al., 2007).

The goal regarding the content is to deliver a message, which can be anything a patient wants to share, that motivates care providers to undertake action, according to the needs of patients. The system should therefore comply with the following requirements:

- The patient consulting system should be able to record all kinds of care related messages.

- The patient consulting system should encourage that the message is elaborated on, with a limitation to prevent that it becomes too long.

- The patient consulting system should ensure that all care providers that have to deal with this message understand the message and are motivated by it.

- If care providers do not understand what is meant by the message, they should be able to ask the patient for clearance.

2.2.2 Timing

(11)

10

The goal of the system regarding timing is that patients should be able to express their needs or feedback whenever they want to and that care providers use the feedback to adjust the care to the needs when they meet the patient. This lead to the following requirements:

- The patient consulting system should be accessible and available whenever patients want to. - The patient consulting system should be available for care providers when necessary.

2.2.3 Mode of delivery

The mode of delivery includes both the way in which patients provide a message and the way care providers receive the message. In managed clinical networks, some messages are intended for a single care provider, other messages can be intended for all care providers during the treatment process. In order to make sure that the message is received by the right care provider, a patient consulting system should facilitate that the messages can be accessed by multiple care providers, but that only the actors that are concerned with a certain message, have access to this message.

Another aspect of the mode of delivery is if a message should be communicated verbal (face-to-face or via telephone) or written (via paper, mail or social media). Verbal communication is more synchronous than written communication: people interact in real time and there is little time between the expressions of one person and the response of the other person (Berger & Iyengar, 2013). With written communication, which is more asynchronous, it takes longer for a person to respond. This time can be used to construct and refine the response, until the message conveys perfectly what the respondent wants to express (Berger & Iyengar, 2013). However, written communication also takes more effort and time and involves less social presence, which for example can complicate the sharing of emotions and therefore the understanding of messages (Baralou & McInnes, 2013). Written communication is also more permanent compared to verbal communication. For the care providers this can be an advantage. The message is written in the words of the patient and can be shared and read by different care providers in the MCN. If the message is written down by a care provider to which the patient expressed verbal feedback, it is possible that the care provider missed some important aspects of the message. When patients write the message down themselves, it contains what the patient wants to say. The message also contains how a patient feels at the moment of writing. This is valuable information, since it can be difficult to relive the exact same emotion at a later moment in time. On the other hand might the permanent character be a disadvantage for the patients. If they wrote something down in a rush of emotions, they might regret this later on, but they cannot take their words back. Patients will therefore be more careful with expressing certain things and may withhold certain information. The mode in which the patient feels freer to express everything they want, should be chosen as the right mode.

Verbal communication Written communication

Synchronous Time between remark and response

Direct response Time to construct and refine response

Easy to construct response More effort to construct response

Sharing of emotions Less social presence (no emotions)

Responses are volatile Responses are permanent

(12)

11

The goal of the consulting system is that the message is delivered to the concerned care provider(s) in a way that captures the message as it was meant by the patient. It should for the patient also be clear to who they can go with their questions, feedback or needs. This leads to the following requirements: - The patient consulting system should be accessible by the care provider(s) concerning the message. - The patient consulting system should capture the message in the way as it was meant by the patient. - The system should allow a form of communication that motivates patients to express everything they want to share.

- It should be clear where a patient can go with questions, needs or concerns.

2.2.4 Sender credibility

Characteristics of the provider of the message make that the message they provide is perceived as credible or not and will influence performance of the receiver of the message (Strijbos et al., 2010). The provider in this paper is the patient that communicates a message towards one or multiple care providers that are part of the MCN. Credibility is often explained from the perspective of peers or managers providing feedback, with expertise and high performance as one of the most important sources of credibility (Ilgen et al., 1979; Steelman & Rutkowkski, 2004). However, patients do not have much expertise about the care that is provided, but they know what they like and dislike about the provided care and what they would like to see change. So patients can be seen as ‘experience experts’ on care provision.

The goal of the system regarding credibility is to assure that the feedback provider is perceived as credible by the different care providers. The knowledge on sender credibility leads to the following requirements:

- The patient consulting system should provide guidance to the patients when composing a message, to ensure that the ‘experience expert’ knowledge is emphasized.

- The patient consulting system should ensure that different care providers, with different visions of credibility, see the source as credible.

2.3 Willingness to communicate a message

(13)

12

by providing negative feedback about a care provider, the care they will receive in the future from this care provider will be negatively changed.

Earlier experiences with communicating messages will also influence the willingness to do this again. This is another aspect of ostracism. If the patient has expressed feedback or asked a question, but had a negative experience with this, i.e. the relation with the care provider changed or nothing was done with this feedback, the patient can feel rejected and experience social pain. Social pain can more easily be relived and is more painful than physical pain (Chen, Williams, Fitness & Newton, 2008). Because people can recall the feeling of rejection easily, they will avoid stimuli associated with it (Chen et al., 2008). Patients will therefore be less willing to provide feedback if they had a bad experience with it in the past. Especially in a MCN, it should be assured that something is done with the message. If the patient communicates a message in the beginning of the network, but has the feeling that the feedback is not used for improvement, this patient will be less willing to provide feedback later on in the process. Closely related to this is trust. A patient has to trust the care providers and the consulting system that they will treat the feedback that is provided confidentially. This means that the care providers have to ensure that they will not treat the patient differently, no matter what the message is that was communicated. This also means that it has to be ensured that the data about patients will be stored in a secure way and that only the concerned care providers have access to this data.

The goal regarding the willingness to communicate a message is therefore to design a system in which patients feel free and are encouraged to express everything that is on their mind. This leads to the following requirement of the patient consulting system:

- The patient consulting system should be seen by patients as a system that, when used, will increase the satisfaction with the provided care for that patient.

- The patient consulting system should provide a confirmation to patients that the message is received and that something will be done with it.

- The patient consulting system should assure that the data is stored in a secure way and that only the care providers concerned can have access to it.

- The patient consulting system should assure that the message is only used to improve the provided care and not for other purposes.

Based on literature, reasoning and expectations, a list with goals and requirements is composed. However, it is not ensured that patients and care providers also have these requirements for a consulting system. It should therefore be investigated, by making use of the list above, which goals and requirements patients and care providers have. This will be done in the following sections. The global approach can be found in figure 2.3, but will be elaborated on in chapter 3.

Figure 2.3: Process of validating goals and requirements found in literature

Initial list of goals and requirements

Check with stakeholders

Final list of goals and requirements

(14)

13

3. Methodology

This research tries to find an answer on the question how a patient consulting system can best be designed, by identifying goals and requirements. Design science develops knowledge that will be adopted by users in the field (van Aken, 2004). A design study was therefore conducted. Peffers, Tuunanen, Rothenberger & Chatterjee (2007) identified six stages in a design study process: identify the problem, define objectives of a solution, design and development, demonstration, evaluation and communication (figure 3.1). This paper focuses on stage one through the design in stage three. The other three stages and the development of the system will not be covered, due to time limitations. The first stage, identifying the problem, is already discussed in the introduction. Stage two, identifying requirements for a solution, is partly finished. However, stakeholders still have to be interviewed to validate, refine and/or elaborate on the objectives and requirements found. This section will first elaborate on the context and stakeholders. Hereafter will be explained how the data gathering and design of the patient consulting system was conducted.

Figure 3.1: Design study process model (Peffers et al., 2007)

3.1 Context and stakeholders

3.1.1 Context

(15)

14

For patients means this form of organizing that they will have to deal with different care providers in different care facilities. This will complicate providing care according to the wishes of patients.

The most important groups of stakeholders are the care providers and patients. More stakeholders can be identified, like insurance companies or hospitals itself. However, only patients and care providers will be discussed in this thesis, since they will have to work with the system. Information on these groups will be provided next.

3.1.2 Stakeholders

Different care providers are concerned with the treatment of the patient. These are gynecologists and a general practitioner, but for instance also the oncologist and internist. Also other specialists can become concerned with the treatment, depending on possible complications or drawbacks in the treatment. The care for gynecological oncological patients will mainly be provided in the peripheral hospital, which is mostly close to patients. However, the surgery will take place in the UMCG, i.e. the central hospital. Figure 3.2 shows the treatment process of the patient. A patient will go to the general practitioner (GP) with complaints. The GP will send the patient to a gynecologist with a specialization in oncology in a peripheral hospital, if the complaints are suspicious. This gynecologist will diagnose the patient. Patients that need to undergo a surgery will be sent to the central hospital (in most cases the UMCG). Patients sometimes have to undergo a neo-adjuvant chemotherapy. This can take place in the peripheral or the central hospital. The patient will be operated by a multi-disciplinary team in the central hospital, but the patient will have most contact with the gynecological oncologist (GO). The chemotherapy and aftercare will most often take place in the peripheral hospital again.

Peripheral Central Central Peripheral

Pre-diagnosis Diagnosis (Neo-adjuvante chemo) Surgery Chemo Aftercare

GP = General practioner

GOA = gynecologist with a specialization in oncology GO = Gynecological oncologist

MDT = Multi-disciplinary team

Figure 3.2: The treatment process of a patient

3.2 Data collection and design

The design of a patient consulting system took place in two consequent loops of data collection. The result of the first loop was the design of two possible systems, which were evaluated in the second loop. This format was chosen, based on how Van Aken, Berends & van der Bij (2012) depict the design phase (the third step in the model of Peffers et al. (2007)). They split this phase into 5 steps (figure 3.3). After the problem is known and specifications are developed, a design on paper is developed based on the synthesized specifications. Then, the design on paper will be evaluated and if the design meets all requirements, the system will be designed in practice. If the design does not meet all requirements, it could be that more specifications (option Z) or another design is needed (option X). This process is repeated until a good system is found (Hevner, March, Park & Ram, 2004). For this paper, the first loop

(16)

15

of data collection ended at the synthesis phase and the second loop ended after the evaluation. The result of this thesis is a number of recommendations on how to refine the design.

Recommendations

Loop 1 Loop 2

Figure 3.3: Process of system design (van Aken et al., 2012)

To come to a good design, data from different sources is needed (Holmström, Ketokivi & Hameri, 2009). In this research, data from two sources was collected:

A. from stakeholders, i.e. the users, to obtain more insight in the goals and requirements of the system. B. from experts and existing systems, to obtain more insight in the design and feasibility of comparable designs of patient consulting systems.

Knowledge from both sources was synthesized at the end of the first loop to develop two possible designs. In the second loop, both sources evaluated the design. Their evaluations were taken together to base the recommendations for the final design on. This process is depicted in figure 3.4.

First loop Second loop

Obtaining insights

from stakeholders Evaluate design

Synthesis Synthesis

Obtaining insights Evaluate design

from experts in design

Figure 3.4: Process of data collection and system design

The data collection process of both the first and second loop will now be described.

3.2.1 First loop of data collection

In the first loop, interviews with patients and care providers, i.e. the stakeholders, were conducted to refine the goals and requirements found in literature. Experts were interviewed to obtain more knowledge on the feasibility of patient consulting systems. Comparable existing systems were evaluated for the same purpose.

Problem analysis

Developing specifications

Synthesis Evaluation Design

(17)

16

Stakeholders

Semi structured interviews with patients and care providers were chosen as an appropriate mode to gather data from stakeholders, because interviews provide space for explanations and clearance (Yin, 2009). The goal of the interviews with stakeholders was to find out which goals and requirements they have for such a system. Some of the goals and requirements found in literature concerned only patients, some only care providers and some concerned both stakeholders. Patients, for instance, are not directly concerned with the question when care providers should have access to the system. It, however, does concern them within how much time they get a response. For all goals and requirements that influenced a stakeholder group in some way, an open question was developed to ask the opinion of the patients about that topic and therefore, to find out if they also saw it as a requirement. Two different questionnaires were therefore developed, one for patients and one for care providers. Open questions were used to avoid steering the answers in a certain way. Also questions were asked in which patients were asked to talk about a certain situation they experienced, to find new goals or requirements. See appendix A for the patient interview scheme and appendix B for the interview scheme for care providers.

Patients in different stages of the treatment process were interviewed. These stages were: 1. some days after surgery and 2. when a patient came back for a check (after 3 months or later). It was not possible to interview patients from other stages, due to time limitations and difficulties in organizing those interviews. The number of patients that was interviewed was n=16 (see table 3.1). All interviewees were women and the average age was 60.25 years. It was not possible to interview more patients after surgery, since they did not want to be interviewed or were not in a clear and healthy enough state after the surgery.

Patients Number

After surgery 4

At a check after treatment 12

Total 16

Table 3.1: number of patients interviewed during the first loop

Care providers with different specialisms were interviewed (table 3.2). As was found in figure 2.2, the personal characteristics of the feedback receiver influence the perceived level of quality. By asking care providers with different medical backgrounds - and hereby assuming that characteristics have influence on the specialism one chooses - the differences in characteristics are taken into account. It also accounted for the differences in requirements of different specialisms.

(18)

17

providers would be enough to cover a wide range of opinions and that no new insights would be found when interviewing more care providers.

Specialism Number of care providers with this specialism

Gynecological oncologist 2

Gynecologist with a specialization in oncology 3

Case manager 2

Radiotherapist 1

Medical oncologist 1

Table 3.2: The number of different specialist that were interviewed

After all interviews were conducted, the interviews were transcribed and coded. The codes were based on requirements found in literature. The statements of patients and care providers belonging to certain codes were analyzed and goals and requirements were identified from this.

Experts

Three experts were interviewed, who can be divided in two groups: an expert on the relations between patients and care providers and experts on patient consulting systems.

The expert on the relations between patients and care providers was interviewed early in the process, to get some initial ideas on what patients and care providers want. For all the goals and requirements found in literature, an open question was made to ask the expert the expected opinion of stakeholders on that topic. Based on this information it became clearer where the emphasis in the interviews with stakeholders should be.

Two experts that already implemented an online information system were interviewed. The goal of these interviews was to get more insights in the implementation of the system, how the system works and what the reaction of the patients was. A list with topics of features of this system that are interesting for the patient consulting system was made. For each of these topics a question was developed, which was asked to the experts.

Other, already existing communication systems in the health care environment were observed, to get an idea what kind of systems already exist and what their goals are. These systems were found by searching on the internet and by asking the care providers and experts if they knew about existing systems. The systems were studied by reading about them or by making use of them.

Design

(19)

18

3.2.2 Second loop of data collection

In the second loop, the designs were evaluated by both stakeholders and an expert. According to Peffers et al. (2007) and Hevner et al. (2004), evaluation involves comparing the objectives of a solution to observed results. Depending on the nature of the problem, those results can take any form. For instance, as is the case in this research, the evaluation of the design on paper by stakeholders.

Stakeholders

Stakeholders were asked to rate to what extent they expect that the system ensures that a goal or requirement will be met (for instance: to what extent can be ensured that care providers show they see a patient as credible) or to what extent a goal or requirement is covered with a system (for instance: to what extent supports the system the communication of different kinds of messages). The requirements found in the second loop are the refined version of the requirements found in literature, therefore were only these requirements taken into account. The goal of the interviews was to find out what the opinion of stakeholders on the designs was and to what extent the designs comply with the goals and requirements found.

Patients (n=5), all women, average age 50.4 years, who came for a check after their treatment, were asked to evaluate the two designs. Also care providers (n=4), with different specialisms for the same reason as described before (see table 3.3), were asked to evaluate the design. One requirement (it should be ensured that patients are informed that they can use the system), was deleted since patients found it difficult to rate this requirement. The interview schemes for patients and stakeholders can be found in appendix C and D.

Specialism Number of care providers with this specialism

Gynecological oncologist 1

Case manager 1

Medical oncologist 1

Nurse practitioner 1

Table 3.3: The number of different specialist that were interviewed

Experts

In the second loop, an expert (n=1) is asked to evaluate the design in the same way as the stakeholders. However, more in depth questions about the feasibility were also asked. Since this expert is a nurse practitioner, the data she provided will also be used as data for the care providers, to enlarge the number of respondents.

Design

(20)

19

4. Results

Based on the interviews with stakeholders, goals and requirements of these groups were identified. In this chapter first the goals and requirements of patients and care providers for a patient consulting system will be discussed. Then the knowledge provided by experts on the feasibility of such systems will be presented. Lastly, a summary and the relations between all this information will be provided.

4.1 Stakeholders

4.1.1 Patients

Patients were generally positive about the idea of developing an patient consulting system (n=10). They state that it is good when a patient feels safe and have a tool that helps them clarify the provided information. Patients saw that they could get some peace of mind as an advantage of this system (n=6). Other advantages that were mentioned are the shorter line to a care provider (n=1), better communication between patient and care provider, since patients feel more free to say what is on their mind (n=2) and that care providers can work more effectively, since small questions can easily be asked by email and not during a consult (n=1). A last advantage is that information provision becomes more uniform and less dependent on one specialist (n=1).

Despite these advantages, some patients were less positive. They state that they already can get in contact with the care providers when they want and do not see an advantage of using another system for this (n=5). However, patients want to have some kind of possibility to contact their care provider, next to the normal consults (n=14). Disadvantages of such a system that were mentioned are that the workload of care providers becomes too high (n=6), the security of data (n= 7) and that contact becomes less personal (n=2).

Content

Patients indicated that they want to have a possibility to communicate both positive and negative feedback (n=9) to ask questions (n=10) and to express their needs (n=6). The goal regarding the content should therefore be to communicate any type of message patients would like to share with a care provider.

The following requirements were mentioned:

- The patient consulting system should support the communication of different kinds of messages. - Patients want to explain and motivate their messages (n=8). Patients also indicated that they would like to have guidelines when composing the message, to come up with certain questions or comments they would not have come up with, without the guidelines (n=5). The system should therefore contain some guidelines in composing the message.

Timing

(21)

20

timeframe in which patients want a response varied from a direct response to a response within 2 weeks (and a number of possibilities within this timeframe). However, it is important to communicate when the response will be provided (n=1).

It should be noted that, even though it was expected that patients in different stages of the treatment would have different needs and require different things, no significant differences were found in the answers of patients from different stages of the treatment. It was also found, from statements from care providers, that patients have most questions in the diagnosis phase and after surgery. However, patients also had questions in other phases and for feedback or other types of messages an increase in certain phases was not found. The system should therefore be available during all phases, from diagnosis until after care. The goal regarding timing should therefore be for the system to be available when patients want.

- The patient consulting system should at least be available in all phases of the treatment, during the day and evening (n=11).

- It should be communicated to patients when a response will be provided (n=1).

Mode of delivery

Patients prefer to have verbal contact with their care provider, to explain things and to get a direct answer (n=13). Three patients did not have a preference for verbal or written communication. It was found that patients prefer to communicate with their own care provider, since (s)he knows most about the disease and the patient (n=7). Patients therefore also prefer to communicate with the care facility responsible for the part of the treatment where the patient wants to say something about (n=4). So if a patient has a question about chemotherapy that took place in Zwolle, the patient would prefer to contact Zwolle. However, for comments on the surgery in Groningen, the patient would prefer to contact Groningen. This implies that, because of the MCN, the system should be connected to all hospitals in the MCN. Some patients prefer to go to the general practitioner with questions (n=2). Other patients just want a good and fulfilling response (n=5). The goal regarding the mode of delivery is therefore that patients are able to communicate a message the way they want (verbal or written) to a care provider with whom they want to share the message.

- It should be possible to have some kind of verbal communication (n=13).

- The patient should be able to communicate with the care facility or care provider responsible for that part of the treatment where the message is about (n=4) and the patient should be able to indicate which care provider should respond (n=7).

- Even though most patients did not mind that other care providers see their message (n=12), since other patients did mind (n=3) and since patients should be protected, a patient should be able to indicate which care provider(s) can have access to a message.

Sender credibility

(22)

21

explain and communicate (n=7), when something is done with the feedback or question (n=3) and when the care provider listened, was interested in the patient and placed themselves in the situation of the patient (n=7). Patients had the feeling they were seen as less credible when care providers think too easily about things that are difficult to understand by the patient (n=2) and when things went wrong in the treatment, for instance when the wrong diagnose was made (n=3) or when information was not provided (n=2). All patients who had the feeling they were seen as less credible, linked this with a negative emotion about the hospital where it took place. So, even though sender credibility is about how care providers see the patients, it is also important that care providers let patients feel they see them as credible, by explaining what is happening and providing proper information, by showing which actions are undertaken based on their messages and by showing interest in the patients. This will make that patients think more positively about the provided care. The goal regarding sender credibility is therefore that care providers provide a good and clear response and show interest in the patient. The following requirement was mentioned:

- It should be ensured that care providers show interest in the patient, clearly explain and communicate information and indicate when and how they will undertake action based on the provided message.

Willingness to communicate a message

Reasons indicated by patients why they do not provide feedback or ask questions are that they do not want to bother care providers, since they are busy (n=2) or because they do not like to say negative things in general (n=1). Patients are not afraid that the data they provide will be used for the wrong purposes (n=9). However, when they have the feeling that nothing is done with their comments, especially when it concerns negative feedback, five patients would not provide comments again. Other patients would go to another hospital (n=6). This would have consequences if a patient consulting system for the whole MCN would be in place. Patients would not use the system or even leave the MCN and go to a hospital outside the MCN. The goal regarding the willingness to communicate a message is therefore that patients are intrinsically willing to use the system.

- Patients should be informed by care providers that they can use the patient consulting system, also for negative messages and that the contact will not bother the care providers.

- It should be ensured that patients have the feeling that the comments they communicate are taken into account and that something is done with it.

4.1.2 Care providers

(23)

22

The advantages that were identified by care providers were that patients easily can get in contact with the care providers and ask questions, which they otherwise would have asked too late or not at all (n=9). Also, less travelling (n=1), more and better information (n=2), a better care provider-patient relationship (n=1) and a safe feeling for patients (n=1) are seen as advantages. An advantage for the care providers is that they can work more efficiently (n=3), because they know more about the patient and where the patient worries about. Consults can therefore become more focused on those questions and worries. It will also be more efficient since care providers can answer when they want to and when they have time (n=5).

Content

Care providers mentioned different things that should be able to be communicated, like questions (n=4) or feedback (n=3). One of the care providers summarized this when saying: ‘whatever it is that patients want to say, it should be possible’. The patients want, according to the care providers, to reach different goals with the message they communicate, for instance patient specific information (n=3), an answer (n=4), reassurance (n=4) or an improvement in the care provision (n=1). The goal regarding content is therefore that whatever it is that patients want to say or reach with their message, it is possible to communicate that.

- The patient should communicate a clear message, with some motivation (n=9). However, the message should not be too long (n=4).

- It should be ensured that the responses of care providers are communicated in language that is understandable by patients (n=2).

Timing

Care providers would like to have some freedom in answering to patients (n=5). They would like to respond when they have time for it and not when a patient wants to have an answer. However, they also state that they should have some kind of deadline, in order to make sure that patients will get a timely response. Especially if the message has to be forwarded to one or more care providers in the MCN, deadlines are important. The goal regarding timing is therefore that care providers are able to decide when they want to respond, within a certain time limit.

- Care providers should not have to respond directly, so they can look up some background information and respond when they have time. This also means that emergency questions should not be communicated with this system, but that another form of communication should exist for emergency, which should always be available (n=7).

- It should be communicated to the patients when they can expect a response (n=5).

- Care providers should have access to the data that is interesting for them when they want to (n=6).

Mode of delivery

(24)

23

mainly has to do with the choice in verbal or written communication. In a lot of conversations between care providers and patients is the non-verbal communication very important. That is why care providers choose verbal communication as the preferable mode of communication (n=4). However, two care providers prefer written communication when the message concerns multiple care providers. Written communication allows a care provider to respond in its own time and to look up more information or ask other care providers in the MCN before answering. Written communication can also more easily be communicated to other care providers (n=2). This could be important in the MCN, since multiple care providers are involved with the care of a patient. Other care providers state that it depends on the type of message which mode they prefer (n=3). As can be seen, the preferred modes are very mixed. The goal regarding the mode of delivery should therefore be that care providers are able to respond to patients with a mode of delivery they prefer and find suitable for that type of message.

- The patient consulting system should be easily accessible and easy to use, for both patients and care providers (n=7).

- The system should be reliable, it should always work properly (n=1).

- The security of data should be ensured (n=4). However, the security should not become a barrier for the access of the system, for instance by having too much access codes (n=1).

- There should be a form of triage in the system (n=6). It is not possible to have one person responding on everything that happens in the whole MCN (n=1).

- The communication line with patients should be different from the communication line with colleagues (for example separate email addresses) (n=2).

- There should always be an option to have verbal contact with patients, to ask for clearance or to have conversations where non-verbal communication is important (n=8).

- When a message is important for multiple care providers, they all should have access to this message (n=5).

Sender credibility

Care providers stated that it sometimes happens that some patients contact the care providers very often with sometimes, in the eyes of the care providers, irrelevant questions (n=8) which make that care providers sometimes get a bit irritated. If the system covers all patients in the MCN, even more patients can (and will) send a message. However, the care providers stated that those ‘irrelevant questions’ do not influence the credibility of the patient or the quality of the answers they provide (n=8). They state that patients contact them because they want something and it is their job to respond and help the patient, so all of the patients and questions should be taken serious. The goal regarding credibility is therefore that the care providers see responding to patients with this system as their duty and part of their job.

The following requirement was mentioned:

(25)

24

Willingness to communicate a message

Care providers want to facilitate a service to the patients by using the patient consulting system. However, the care providers state that the patient should want to use it and should not be forced to use it (n=6). The goal regarding the willingness to communicate a message is that patients are intrinsically motivated to use the patient consulting system.

The following requirement was mentioned:

- Patients should be informed about the existence of the patient consulting system and be motivated by care providers to use the system (n=6).

4.2 Experts and systems

4.2.1 Experts

Experts were interviewed about a kind of consulting system, called mijn digipoli. The goal of this system is to provide better and personalized information to patients. Patients can login on a website and find information and interactive images about their specific disease. This information can be seen as modules which are put together in different ways for different patients. A feature of this website is that patients can ask questions via an email system (see figure 4.1). The questions come in at the nurse practitioner, who answers most questions. The questions that cannot be answered by this nurse are forwarded to the specialist. Both experts stated that this form of triage is working well and makes that the specialist does not get too many emails.

Figure 4.1: Graphical representation of ‘mijn digipoli’.

The system is developed in an organic way, by starting small with only a couple of patients. They were asked about what they liked and disliked and the system was adjusted according to this. Hereafter was the group of users enlarged and the same process was repeated. This form of implementation is advised by both experts. The experts also stated that patients are satisfied with this system and are making use of it. Also older people make use of this digital system. When they do not have access to the internet,

(26)

25

they are asking help of family or friends. The experts also stated that they noticed that older people are making more and more use of the internet. Patients are also happy with the option to ask questions – in a questionnaire was found that 25% of the respondents think that the option to ask questions is the most important feature of the site. The experts noticed that patients find it easier to send an email with a question than to call to the policlinic. A disadvantage of the system is that it takes a lot of time to write the modules. However, once the modules are written, they can easily be put together and the experts say that this system saves time. Patients ask fewer questions, because of the improved information provision and consults are organized more efficiently, since patients are better prepared. However, this is an estimation made by the experts, it is not measured.

The expert on the interaction between patients and care providers was enthusiastic about the idea of developing a patient consulting system. He stated that it is important that patients know what is going on with them. Patients should be supported with this whenever possible. The expert noticed that, when doing research and talking to patients, two things are important to patients: the way a doctor treats the patient, which has to do with listening and taking the patient serious, and the information provision, not only in the consult, but also afterwards. This was also found in the interviews with patients when discussing credibility. The expert also says that not all elderly patients can or are willing to use the internet, but that the number of elderly patients using internet is growing and that internet use is becoming more integrated. However, the expert experienced that patients enjoy personal contact with their care provider, so this system should be an addition to the normal, personal contact, not a substitution of it.

4.2.2 Existing communication systems

(27)

26

Name Description Advantages Disadvantages Overall

Mijn digipoli Online personal information provision and option to send emails.

Patients are more informed about their disease.

It takes a lot of time to write

the modules. ₊

Online intake Online anamnese before the first consult.

The anamnese does not have to be done anymore and consults become more structured.

Patients have not seen their doctor yet and do not know the person who sees their answers.

+-

Vertel het ons A hospital wide system where patients can post feedback or remarks.

It is an official place to post comment and it is therefore likely that the comments will be used for improvement.

It is impersonal: the person who reads the comment first does not know the patient or

the story behind it. +-

MijnRadboud A website where patients can have a look in their own dossier and see results, see letters between care providers and ask questions.

Patients are up to date on their own medical status.

Care providers may feel that they are being monitored. Patients may not understand all information and may therefore worry about nothing.

+

Mijnzorgnet.nl Patients can create their own community, gather

information about their disease and can invite people to talk about and share this information.

Patients can get in contact with care providers and specialists other than their own and can get more or different explanations.

Care providers have to voluntary make an account. It could therefore be that your care provider does not have an account so (s)he cannot be reached.

+

Digitalepoli.nl Website where care providers can place information and where they can communicate with patients.

Patients will have access to up to date information and can easily get in contact with care providers.

Care providers have to voluntary make an account. It could therefore be that your care provider does not have an account so (s)he cannot be reached.

+

Facetalk Patients can have an online consult via the webcam, using a secured internet connection, with one or multiple care providers.

Patients do not have to travel a long time for a short consult.

It is not always enough in a consult to show something and talk about it, sometimes a care provider has to feel it or touch it.

+

Digitale Parkinson poli

Website where patients can read, write and share information and experiences.

Patients can read information and can get in contact with other patients and care providers.

Mostly experiences and information from patients is shared and it is questionable how objective that information is.

+

Ipads Patients can ask questions from home, using an iPad, and attach a code to it. The time for a response depends on the code.

Patients can easily get in touch with their care provider.

Care providers could experience a high workload, since they have to respond within 15 minutes if the attached code is red.

+

(28)

27

4.3 Summary

As can be seen from the previous sections, there is some overlap between the goals and requirements of patients and of care providers and the information provided by experts. The following table shows which information is overlapping (or contrary) and which information stands alone.

Category Patients Care providers Experts

Kind of message

The system should support the communication of all kinds of messages.

In mijn digipoli all types of questions are asked.

Composition of message

The system should contain some guidelines in composing the message.

The patient should

communicate a clear message, with some motivation. However, the message should not be too long.

Availability The system should at least be available during the day and evening.

Response time It should be communicated to patients when a response will be provided. Care providers should not have to respond directly. It should be communicated to patients when they can expect a response.

Users of mijn digipoli know that they will get a response within 24 hours and appreciate this time limit.

Communicate verbally

It should be possible to have some kind of verbal communication.

There should always be an option to have verbal contact with patients.

All experts state that such a system should be an addition of the normal contact, not a replacement.

Respondent The patient should be able to communicate with the care facility or care provider responsible for that part of the treatment where the message is about and should be able to indicate which care provider should respond.

There should be a form of triage in the system.

In mijn digipoli, a triage system is in place in which the nurse answers most questions. The experts stated that this is working very well.

Accessibility A patient should be able to indicate which care provider(s) can have access to a message.

When a message is important for multiple care providers, they all should have access to this message. Care providers should have access to the data when they want to. Behavior of care providers

It should be ensured that care providers should show interest in the patient, clearly explain and communicate things and indicate that they will undertake action based on the provided message.

It should be ensured that the responses of care providers are communicated in language that is understandable by patients.

An expert noted that the way a care provider treats a patient and the information provision are important to patients.

Motivation to use system

It should be ensured that patients are informed by care providers that they can use the system and that the contact will not bother the care providers.

It should be ensured that patients are informed about the existence of the system and be motivated by care providers to use the system.

(29)

28 Actions

undertaken

It should be ensured that patients have the feeling that the comments they

communicate are taken into account and that something is done with it.

It should be ensured that the starting point when responding is that every patient is credible and every question and remark is important.

Ease of use The system should be easily

accessible and easy to use.

System reliability

The system should be reliable, it should always work properly.

Data security The security of data should be ensured. However, the security should not become a barrier to access the system, for instance by having too much access codes.

All systems found are secured with a password and sometimes even with an authentication with their DigiD.

Medium The communication line with

patients should be different from the communication line with colleagues.

Table 4.2: Overview and relations between the provided information of patients, care providers and experts

(30)

29

5. System design

In this chapter, two possible designs will be introduced and the feedback provided by stakeholders on the designs will be discussed.

5.1 Designs

From the requirements appears that patients like to have verbal contact with their care providers. However, patients can already get in touch verbally with their care providers in two ways, during the consults and they can call the care facility. In most cases, the patient calls the nurses at the policlinic, who can answer some of the questions. The questions that they cannot answer, are passed on to the specialist, who calls the patient back with an answer. Patients that undergo chemotherapy can reach the specialist directly.

Since patients indicated that they like this kind of contact and since some patients indicated they do not need another form of contact, it is advised to maintain the current way of communication next to the new system. It is expected, based on experiences of experts, that over time, more and more people will use the new system instead of the current way of communicating. Even though substitution was not a requirement, the designed systems could, in the future, both substitute a part of current way of communicating. The experts (n=3) and some care providers (n=3) also stated that the use of internet in this patient population probably will increase over time. By anticipating on the future and by considering the amount of possibilities the digital world has to offer, which was found while exploring existing systems, two digital systems were designed.

A general description of these systems will be provided first. Then a list will be provided that shows in which way all requirements found in the interviews are met in these systems. It was difficult to determine the value of the requirements based on the interviews, because some requirements were mentioned by one stakeholder, while other stakeholders could also find it important but forgot to mention it. All requirements were, in this stage, therefore seen as equally important. The systems were therefore designed in such a way that all requirements were met. In the second loop of interviews will be investigated which requirements are more important than others.

1. E-mail system

(31)

30

Figure 5.1: Schematic overview of the email system

2. Online communication with the use of webcams

With the second system patients can communicate with their care providers via a webcam. The technique for this is already existing, namely facetalk. Care providers should have certain hours reserved in which patients can make an appointment via the care administration for a consult (see figure 5.2). An advantage of this system is that patients do not necessarily have to go to a hospital anymore, for face to face contact, but can stay at home in a safe environment. It could be an option for patients that do not have the facilities for a webcam at home, to facilitate this in the peripheral hospitals. The patients do not have to go the central hospital, but only have to go to a hospital that is closer. Especially for the MCN this would be a good solution, since patients sometimes have to travel 150 km for a consult of only 15 minutes. However, not all consults can go via this system, since the care provider sometimes has to do physical research.

Figure 5.2: Schematic overview of the webcam system

Table 5.1 shows an overview of the requirements and how these are met in each system. Where possible are the requirements of patients and care providers in the same category taken together (based on table 4.2). Patient General practitioner Medical oncologist Nurse Gynecological oncological specialist Radiotherapist ……. Sends mail to Sends mail

The design of a patient consulting system in managed clinical networks