INNOVATION ADOPTION CRITERIA OF MEDICAL DEVICES: IDENTIFYING THE MOST CRITICAL INNOVATION ATTRIBUTES FROM SUPPORT STAFF PERSPECTIVE

(1)

1

INNOVATION ADOPTION CRITERIA OF MEDICAL DEVICES:

IDENTIFYING

THE

MOST

CRITICAL

INNOVATION

ATTRIBUTES FROM SUPPORT STAFF PERSPECTIVE

Master thesis, Msc, Supply Chain Management

University of Groningen, Faculty of Economics and Business June 2014

Student name: Bowei Ning Student I.D.: s2444992 E-mail: b.ning@student.rug.nl Supervisor/university Prof. dr. D.J.F. Kamann Co-assessor/university

Prof. dr. J.de. Vries

Supervisor/field of study

J.A. Bax

Manager Procurement, UMCG, Groningen

Acknowledgment: helpful comments on earlier drafts of this thesis were given by my supervisors D.J.F. Kamann, J.A. Bax, and my friend Sebastiaan Minkes. I thank J. T. Zeilstra, for her help and kindness when I did my case study; all the participants of my project, for their time and attention; all my friends, especially for their help and

(2)

2

Abstract

Purpose – This study aims to shed light on the innovation adoption decision-making process, as well as the key attributes influencing the adoption decision of medical innovations from the perspective of support staff, so as to contribute to the formulation of innovation adoption criteria. Design/methodology/approach– The literature review, case study and survey were applied to get an in-depth insight into the innovation adoption decision-making process of generally-used medical devices and the influences of support staff when making a decision on the adoption of innovative medical devices. Ten most important innovation attributes have been identified in the survey round. Analytic Hierarchy Process (AHP) was employed to create a hierarchy of priorities of the ten most important innovation attributes that were identified by support staff in the University Medical Center Groningen (UMCG).

Findings – This study illustrated the status and influence of support staffs in the innovation adoption decision-making process. The priorities of ten most important innovation attributes identified by support staffs represent their emphasis and focus when making a decision on the adoption of new medical devices. In addition, the innovation adoption decision-making process was visualized in this research.

Practical implications – The insight into the innovation adoption decision-making process and the most critical innovation attributes can help suppliers to improve the effectiveness and efficiency in the development of innovative medical products. As for the healthcare provider organization, innovation adoption criteria can be used as the guidelines to enhance the system-level adoption of innovations.

Originality/value – This research is to explore the most critical innovation attributes from the perspective of support staff in the healthcare industry, which is a new territory for research on healthcare. It helps to fill the gap in both theory and practice and make a contribution to the development of innovation adoption criteria of medical devices. Furthermore, the innovation adoption decision–making flow chart was established through interviews with the participants of two procurement cases.

Keywords: adoption of innovation; generally-used medical device; support staff; innovation attributes; prioritizing;

(3)

3

1. Introduction

“During the past 30 years, a wide array of innovations, both administrative and clinical, has

flooded health care systems worldwide, offering potentially beneficial advances in the diagnosis and treatment of diseases and the delivery of medical services in a growing number of clinical domains” (Rye and Kimberly, 2007). On one hand, the adoption of healthcare innovation not

only improves the productivity and economic growth but also benefit the patients, citizen and society (Roback et al., 2007). On the other hand, the adoption of innovation exerts financial pressures to the healthcare industry. In the past, the adoption of medical innovation is mainly determined by doctors. However, in recent years, besides doctors and nurses, more groups of stakeholders like buyers in purchasing departments, support staffs and other staffs are involved in the decision-making process for innovation adoption. There are essentially two categories of stakeholder: internal stakeholders, who are those actively involved in the innovation adoption process; and external stakeholders, who are those affected by the innovation adoption. Each single internal stakeholder plays a different role in the buying process and therefore exerts a definite influence on innovation adoption decision. The major challenge nowadays lies in balancing the interests of suppliers with those of the internal stakeholders like doctors, nurses and other medical staff while satisfying the demands of the business unit managers in hospitals in financial terms simultaneously. In order to tackle such challenge and manage the relations with suppliers and internal users, it would be an advantage to know on what basis users and other stakeholders value certain new medical devices: what do they consider relevant, important and on what basis do they decide to adopt new medical devices.

(6)

6

set of attributes is crucial in the decision–making process helps to create an innovation adoption criteria, so as to satisfy the needs to optimize the adoption decision in healthcare industry.

Figure 1.1 Determinants influencing the innovation-decision process (based on Rogers, 2003 and Fleuren et al, 2004)

According to Drazin and Schoonhoven(1996), “innovation research has been dominated by two

types of empirical work: (a) cross-sectional studies aimed at identifying contextual, organizational, and individual predictors of innovativeness; (b) longitudinal event history studies aimed at predicting diffusion rates for specific innovations across organizational populations” (Denis et al., 2002). Besides these general researches, Meijer (2014)1 studied the ideal types of healthcare Group Purchasing Organizations (GPOs) based on the categorization of medical products in order to achieve greater effectiveness. As he said, medical products can be categorized according to the asset specificity and uncertainty. It is possible to make a prediction on which ideal type offers an effective solution on purchasing issues in view of the characteristics of product categories. On the basis of the findings of Meijer (2014), we can deduce that the introduction of innovative medical devices, which is regarded as vital procurement project, may differ among different types of medical products. However, there is no

1

(7)

7

literature categorizing medical devices based on the product characteristics “asset specificity” and “uncertainty”, let alone exploring the innovation adoption process on each type of medical devices. On one hand, it to some extent hinders the application of GPOs theory on the healthcare industry. On the other hand, it may inhibit the improvement on the effectiveness and efficiency of medical device procurement. Moreover, though our review of the literature, we found that even though amount of theories have identified various types of determinants that may facilitate or impede actual innovation adoption, they fail to specify the most influential attributes from different stakeholders’ perspective in healthcare industry.

To fill this gap, we divided the medical devices into two categories: generally-used medical device and user-specific medical device. The generally-used medical devices are defined as the medical products used by multiple stakeholders, like for example gloves, syringes, beds. Various groups of stakeholders are involved in the adoption decision-making process of this type of innovative products. In terms of the user-specific medical devices, they are used by certain stakeholders such as plastic surgeons and cardiologists. For this type of medical device, in most cases only users are involved in the decision-making process. In addition, in order to further explore the adoption criteria of new medical devices, it is significant to solve two issues. One is how the innovation adoption decision is made under the participation of various stakeholders; the other one is what set of attributes is most influential on different types of stakeholders. Here, the stakeholders who participate in the decision-making process are distinguished into physicians and non-physicians. In general, non-physician stakeholders include nurses, support staffs and other service providers. According to Cambridge Dictionary2, support staff is defined as the employee who works for an organization to ensure its normal operations and supports other staffs who are involved in the organization's main business. Various stakeholders, such as buyers, experts of medical devices, maintenance staffs, are classified as the support staff. In the healthcare industry, the support staff plays an indispensable role in both the normal operation of the medical system and the innovation adoption decision-making process of general-used medical devices.

(8)

8

This paper aims at giving more insights on the complete innovation adoption process of healthcare provider organization. The emphasis is laid on the adoption of innovative generally-used medical device from the perspective of support staff. Based on this, the status and the role of support staffs who are included in the decision–making process were analyzed, and the most crucial set of attributes impacting adoption decisions made by support staff was explored. On account healthcare provider organizations are investing significantly on medical innovations, and most of them are hindered by the problem of appropriate system-level diffusion of innovations, the insight into the factors that facilitate or inhibit introduction and secession in these organizations is conductive to enhancing the appropriate system-level diffusion of innovations (Rye & Kimberly, 2007). Furthermore, with a deeper investigation, innovation adoption could be facilitated by reducing existing barriers. The in-depth understanding of innovation attributes helps to improve the effectiveness and efficiency in the development of innovative medical products. In this way, the rejection rate of innovative products and possibility of discontinuance after implementation can be decreased. (Bax & Kamann, 2013).

The research question of this paper is:

What are the most critical innovation attributes taken into consideration by support staff, when deciding on the innovation adoption of generally-used medical devices?

Based on this research question, three related sub questions are proposed:

1. What innovation attributes influencing the individual innovation adoption decision have been identified in the literature?;

2.What’s the position of support staff and how does the support staff influence the innovation adoption decision-making process?;

3. According to support staff, what are the most important attributes influencing the individual innovation adoption decision of generally-used medical devices?

(9)

9

assessment. The answers on the sub questions gave a complete overview of the critical attributes influencing the individual innovation adoption-decision from support staff perspective.

(10)

10

2. Theoretical background

This article aimed to figure out the impact of different determinants on the innovation adoption decision of support staff in healthcare industry. So here we did a literature review on innovation, innovation adoption process, the determinants and attributes on innovation adoption as well as the stakeholders mapping theory, in order to form a theoretical foundation to support our study.

Innovation, as a core conception, is discussed from various perspectives. It can be grouped into diversified types on the basis of different standards. Innovations in health care are usually classified into product, process, or structure (Varkey et al., 2008; Omachonu & Einspruch, 2010). The characteristics of innovation influence the adoption of an innovation in organizations (Rogers 1995; Tornatzky and Klein, 1982). As shown in the theoretical model, the core subjects of innovation adoption process are innovations and stakeholders. Besides, the innovation adoption process correlated with the buying process to a large extent, forming a basic architecture of innovation adoption process in the healthcare industry. Multiple stakeholders are involved in the decision-making flow to decide on the procurement of new medical devices. Moreover, the determinants and attributes of innovation adoption have different impacts on different stakeholders when making a decision on whether or not to introduce an innovative medical product. The body of knowledge on innovation adoption attributes provides a basis of this study, contributing to the design of questionnaires and interviews. As shown in the figure 2.1, elements in this chapter are combined, forming the basic structure of innovation adoption decision-making process.

(11)

11

Figure 2.1 Theoretical Model

2.1 Innovation

Many definitions of innovation have been raised in the past decades. Also, different definitions exist in literatures. “Some researchers conceptualize innovation as a discrete product or

program, while others conceptualize innovation as a process” (Rye and Kimberly, 2007).

Rogers (2003) argued that “innovation is an idea, practice, or object that is perceived as new by

an individual or other unit of adoption”3. This definition provided an example for the former view. Schon (1967) exemplifies latter view, defining the innovation as the process of bringing inventions into use.

2.1.1 Types of innovation

In recent years, diversified types of innovation are explored by various researchers. Bax (2011) made an overview on various classification methods and illustrated them in detail. According to Bax (2011), Gopalakrishnan and Damanpour (1997) divided innovation into product innovation

3

(12)

12

and process innovations. Product innovations are the tangible output of the process. It can be divided into three categories: cheaper but equal products, improved products serving same needs and new products serving different needs. Process innovation means the implementation of production or delivery method which is innovative or improved significantly. Compared to the product innovation, the process innovation changes the (production) process, which transforming input to output. These two types of change are interdependent.

In addition to Gopalakrishnan and Damanpour (1997), innovations can be divided in stand-alone and systemic: Stand-alone innovations are the products developed by a single organization and do not require additional products or systems to operate, whereas the systemic innovations require complementary products or system to operate. They are generally developed by several organizations (Huizingh, 2008; Bax, 2011). The adoption and implementation of innovative products will be impacted negatively if the supporting products or systems are not adjusted.

Besides, Cooper (1998) came up with another classification of innovation. Two types of innovations have been identified, which are radical- and incremental innovations. “Radical

(disruptive) innovations result in truly new products or routines often having no comparable predecessor and requiring a high degree of change to implement. While incremental innovations are defined as the gradual renewal of existing products or routines having only a moderate impact” (Bax, 2011).

(13)

13

Figure 2.2 Technology push versus market pull (source: Martin, M.J.C., 1994; Bax, 2011)

2.1.2 Characteristics of innovations

“Researchers have revealed a number of characteristics of an innovation, as perceived by a potential adopter, to be of influence on its rate and speed of adoption” (Frambach, 1993).

Though there isn’t a standard classification of innovation characteristics, the influence of characteristics has found empirical support in a large scale.

(14)

14

potential adopter. This characteristic includes both divisibility and reversibility. If the technology is divisible, the adoption decision concerns whether the new device should complement the traditional healthcare alternative and, if so, to what extent. Reversibility refers to the ease with which the previous practice can be reinstated, and the economic loss associated with a reversed decision. Both divisibility and reversibility are important factors influencing the riskiness of a purchase. With respect to observability, it can be understood from different angles. For example, how visible are the results of using the innovation? Can others recognize the benefits? (Roback et al., 2007). New technology is more easily to be adopted when the advantages are highly visible. Besides these, Roback et al. (2007) also mentioned the different names for the same characteristic. For instance, trialability and observability are recognized as learning or uncertainty factor in some literatures.

Compared to the work of Roback et al. (2007), Frambach (1993) focus more on the general definition of characteristics, which also could be applied to the business-to-business settings. In addition, Frambach (1993) proposed that the innovation characteristics came up by Rogers should be supplemented with the consideration of uncertainty. Besides, he also explored the relationship between innovation adoption and different characteristics. Relative advantage, compatibility, trialability and observability of innovation were defined as positively related to the adoption of innovation, whereas the complexity of innovation, the uncertainty surrounding, and expectation of fast technological development impact the innovation adoption in a negative way. 2.1.3 Innovation in healthcare industry

According to Varkey et al. (2008), innovations in healthcare industry are relevant to product, process, or structure. The product innovation consists of goods or services which are paid by the healthcare provider organization. A process innovation refers to a novel change in the production or delivery method. As stated by Varkey et al. (2008), process is essential for delivering a product or service even though it is not directly paid by the patients. Usually, structural innovation impacts the internal or external infrastructure, and creates new business models.

(15)

15

(1998) also mentioned the difficulty to make a change on the current medical practices in healthcare provider organizations. “Innovations in patient care, treatment practices and hospital

procedures may include significant health risks related to financial, social, and ethical issues”

(Collier, 1994; Faulkner & Kent, 2001). Moreover, it is laborious to make changes in healthcare organizations due to the regulations and laws on the adoption of healthcare innovations (Faulkner & Kent, 2001). According to Huntington, Gilliam, and Rosen (2000), the mental pressure caused by healthcare performance gap and physicians’ willingness to guard their own autonomy and reputation contribute to a culture of blame and secrecy. This may hinder the generation of innovation and organizational learning. Furthermore, organizational practice or structure innovations require research methods derived from social studies. However, clinicians are more familiar with experimental research methods that are feasible for clinical research. This may lead to the result that some organizational practice or structure innovations are evaluated as lack credibility in the eyes of many medical practitioners (Pope, 1995; Pope & Mays, 2000). This represents another aspect that inhibits the adoption of other types of innovations. For all these reasons, how to facilitate the innovation adoption in healthcare industry constitutes the most urgent problem to the medical suppliers.

2.2 Innovation adoption process

According to Rogers (1995), innovation adoption is “the decision to make full use of an

innovation as the best course of action available”. Besides, the innovation adoption process is

defined as “the process through which an individual or other decision-making unit passes from

first knowledge of an innovation, to forming an attitude toward the innovation, to a decision to adopt or reject, to implementation of the new idea, and to confirmation of this decision”.

Frambach (2002) also discussed the definition of the adoption process, that is, “the adoption

process is a sequence of stages a potential adopter of an innovation passes through before acceptance of a new product, service or idea (hereafter product)”. Based on the description of

(16)

16

implementation. According to Frambach and Schillewaert (2002), the initiation stage encompasses awareness, consideration, and intention sub-stages. In the implementation stage, the decision is made by organization to procure and implement innovation.

With regard to this two-step decision-making process, Rogers (2003) stated that usually there are different sets of individuals involved in these two stages, one set of employees makes decision for the adoption of innovation while the other set of employees takes part in the implementation of innovation. In addition, Rogers (2003) divided innovation-decisions into four categories: optional, collective, authority and contingent. The optional innovation-decision represents the decision made by the individual. For the collective innovation-decision, it was based on the consensus of all individuals of the decision making unit. In terms of the authority innovation-decision, it is made for the entire organization by few individuals with relative power and influence. A contingent innovation decision or “forced adoption” happens when the usage of an innovation by individuals is uncertain and contingent upon a prior organizational adoption decision (Ram and Jung, 1991; Rogers, 1995).

(17)

17

Figure 2.3 Model of five stages in the innovation-decision (Source: Rogers, 2003; Bax, 2011)

2.3 Organizational buying

As stated by Webster & Wind (1972), organizational buying behavior is a complicated process which encompasses multiple goals, various stakeholders, and potentially conflicting decision criteria. It often requires information from many sources and involves many inter-organizational relationships. Robinson, Faris, and Wind (1967) introduced "buy phases" to indicate a series of activities usually conducted in the organizational buying situation. In general, these activities include the identification of needs, establishment of specification and requirements, search for alternatives, evaluation of proposals, selection of suppliers, selection of order routine, and performance evaluation (Johnston & Lewin, 1996). Various roles are involved in the organizational buying process, such as influencer, buyer, decider, and gatekeeper (Webster & Wind, 1972). Stakeholders in the buying center are motivated and influenced by the complicated interaction of individual and organizational objectives (Webster & Wind, 1972). Besides, the organization impacts the buying center through the subsystem of people, tasks and structure (Webster & Wind, 1972). As illustrated by Webster and Wind (1972), “the entire organization is

embedded in a set of environmental influences including economic, technological, physical, political, legal, and cultural forces” (Webster & Wind, 1972).

The General Model for Understanding Organizational Buying Behavior by Webster and Wind (1972) is widely used. In this model, the four classes of variables determining organizational buying behavior are provided. They are individual, social, organizational, and environmental. Within each class, two categories of variables are identified. One is task variables, which are directly related to the buying problem. The other one is non-task variables, which extend beyond the buying problem. As stated by Webster and Wind (1972), any given set of variables has both task and non-task dimensions. Besides, the non-task variables can be divided into achievement motives and risk-reduction motives: the achievement motives refer to personal advancement and recognition, whereas the risk-reduction motives relate to the uncertainty about the alternatives, the outcome of the alternatives and the way relevant others will react to the outcome.

2.4 Innovation adoption determinants and attributes

(18)

18

and external characteristics. Seven attributes were included in the “individual (leader) or internal characteristics”. They are attitude towards change, centralization, complexity, formalization, interconnectedness, organizational slack and size. Moreover, six attributes (system openness, relative advantage, compatibility, complexity, trial ability and observability) were placed into the same category “external characteristics”. Meyer and Goes (1988) attributed 15 attributes into four determinants: environmental set, organizational set, leadership set, innovation decision set. Compared to these researchers, Frambach (1993) divided 29 attributes into more detailed categories, which are adopter characteristics, information processing characteristics, innovation characteristics, competitive environment, network participation, innovation development and marketing strategy. Fleuren et al. (2004) categorized 50 attributes into four determining factors, which were focused on the healthcare sector: socio-political context, organization, adopting person (user) and innovation. Greenhaigh et al. (2004), came up with 53 attributes for nine factors: innovation, adopter, assimilations, communication and influence, system antecedents, system readiness for innovation, outer context, implementation and routinization, and linkage among components. A meta-analysis of the literature reveals that there exists a significant overlap between the lists of attributes in numerous articles. Most of them emphasized the impact of individual, social, innovation, organizational and environmental factors on the innovation adoption decisions. Furthermore, the characteristics of these five determinants resemble the features of four factors influencing organizational buying decision to a large extent.

2.5 Stakeholders in the innovation adoption process

2.5.1 Definition of stakeholder

In the stakeholder literature, both the broad definitions and narrow definitions exist. Freeman and Reed (1983) pointed out that there would be serious differences of opinion between broad and narrow definitions of “Who or What Really Counts”. According to Mitchell et al. (1997), the broad definitions lay emphasis on “specifying the empirical reality that anyone can affect or be

affected by an organization's actions”. Whereas narrow definitions “attempt to specify the pragmatic reality that managers simply cannot attend to all actual or potential claims” (Mitchell

et al., 1997), proposing a variety of priorities for managerial attention. As illustrated by Mitchell et al. (1997), Freeman's definition, which is “A stakeholder in an organization is (by definition)

(19)

19

objectives” (1984), is considered to be one of the broadest definitions. The reason lies in that it

merits all to be a stakeholder. There exist some comments on this definition. As stated by Phillips (2003), Sternberg (1997), and Mitchell et al. (1997), if everyone is a stakeholder of everyone else, there is little value-added in the use of the stakeholder concept (Olander, 2007). The Stanford definition—those without whose support the organization would cease to exist— has confirmed this view. It was regarded as one of the representatives of the narrow definition. Mitchell et al. (1997) has figured out the major difference between broad and narrow views. As stated by Mitchell et al. (1997), narrow views of stakeholders are on the basis of “the practical

reality of limited resources, limited time and attention, and limited patience of managers to deal with external constraints”. Narrow views of stakeholders are more focus on defining relevant

groups based on whether they are directly related to the firm's core economic benefits. Various examples were proposed by Mitchell et al. (1997). Such as, Clarkson (1995) defines stakeholders as voluntary or involuntary risk-bearers, which emphasize aspects of risk in relationship with the firm, and Post et al. (2002) define the stakeholders as potential beneficiaries and/or risk bearers, contributing voluntarily or involuntarily to the organization’s wealth-creating capacity and activities. Besides, a few scholars define stakeholders from a moral perspective, arguing that the essence of stakeholder management lies in the creation and maintenance of moral relationships (Wicks, Gilbert, & Freeman, 1994; Freeman, 1994), or the firm's fulfillment of its obligations to stakeholders via fair distribution of the benefits and harms of the firm's activities (Evan & Freeman, 1988; Langtry, 1994; Donaldson & Preston, 1995).

(20)

20

Mitchell et al. (1997) address this problem. As indicated by Mitchell et al. (1997), although influencers do not have legitimate claims or perhaps any claims at all, they do have an impact over an organization or project. Besides, power and legitimacy are different but sometimes overlap, so that theory of stakeholder identification must accommodate these differences (Mitchell et al., 1997). For these reasons, Mitchell et al. (1997) defined power, legitimacy and urgency as core attributes in a comprehensive stakeholder identification model. Then stakeholders can be clarified into different categories according to their possession of one, two or all three of the following attributes: the stakeholder’s power to influence; the legitimacy of stakeholder relationships; and the urgency. (Olander, 2007).

2.5.2 Stakeholder mapping and power/interest matrix

In general, groups of stakeholders rather than one individual stakeholder are most influential in the formulation of strategy of projects. Stakeholders in different groups often have conflicting expectations on a certain project. “Frequent conflicts between stakeholders revolve around long-

term versus short-term objectives, cost efficiency versus jobs, quality versus quantity, and control versus independence” (Newcombe, 2003). Recently, stakeholder analysis or stakeholder

mapping has evolved to assess the importance of stakeholder expectations (Johnson and Scholes, 1993). There are various stakeholder mapping techniques. Mendelow (1981) proposed a model, which categorized stakeholders according to the attributes “power” and “dynamism”. “Power

ranges from low to high, and dynamism ranges from static to dynamic” (Olander & Landin,

2005). In a static environment, stakeholders are less likely to alter their power base, whereas in a dynamic environment stakeholders may derive their power from the alterations in the bases. Based on the Mendelow’s model, Johnson and Scholes (2008) established the power/interest matrix with replacing the axes of dynamism by interest. This matrix can be used to analyze the following questions (Olander & Landin , 2005):

 “How interested is each stakeholder group to impress its expectations on the project decisions?”

 “Do they mean to do so? Do they have the power to do so?”

(21)

21

Figure 2.4: Stakeholder mapping, the power/interest matrix (source: Newcombe,2003)

In the zone A, the stakeholders are those who have little power to influence the projects and low level of interests in them. Minimal efforts are required from managers. Stakeholders with high level of interests in the projects but little power to impact them can be categorized into zone B. They should be kept informed of the core strategies and decisions which have been made. Discontent of these stakeholders over specific decision or strategy may have a negative impact on the projects. Therefore, good negotiation and communication with this type of stakeholders are vital to the execution of projects. As key player, stakeholders in Zone D have a high priority in making a decision. Their acceptability of decisions is a major consideration when formulating project strategy. Regarding the stakeholder in Zone C, they have great power to influence the project but with low level of interest in the project. They are often the stakeholders who are the most difficult to manage. According to Newcombe (2003), for this type of stakeholders, their level of interest in the organization’s strategies will remain low provided that they feel satisfied with the policies adopted; in case that they become dissatisfied, they can increase their interests and turn into key players easily. Besides, they will step into Zone D simultaneously.

(22)

22

2.5.3 Stakeholders in the innovation adoption process of healthcare industry

Adoption of medical innovations is related to the decision of healthcare organization to acquire a medical technology (Ghodeswar & Vaidyanathan, 2007). As said before, the innovation adoption decision is made by intra-organizational use at the individual level. Multiple stakeholders are involved in the innovation adoption decision-making process. According to Freeman (1984), stakeholders in an organization are “any group or individual who can affect, or is affected by, the

achievement of a corporation’s objectives”. In healthcare provider organization, stakeholders

refer to consultant clinician, senior nurses, allied health professionals, patient, patient’s family/carers, procurement staff, manager and accountants, and other service providers (public or private sector). In general, stakeholders working in the hospital could be categorized into two types: physicians, non-physicians. As stated by Thomas et al. (2003), it is of great importance to include non-physician counselors into the analysis of innovation adoption process due to their importance in initiating and implementing treatment. Thomas et al. (2003) did a survey in 2003, aiming at exploring who adopts a certain drug and why. In this case, both physicians and non-physicians were taken into account. Five potential factors were defined to study the adoption patterns of these two different types of stakeholders. According to the results, we could conclude that non-physician adoption patterns differ in several respects from the physician-based model. That is why we need to explore the impacts of different determinants on each type of stakeholders separately.

(23)

23

decisive stakeholders to decide whether or not to adopt and implement a medical product innovation. Besides, as one type of non-physician stakeholders, nurses play an important role in the innovation adoption decision-making process as well, especially for the medical supplies like syringe and bandage. However, with regard to the support staff in the healthcare provider organizations, few studies demonstrated the status or importance of support staff in the decision-making process.

(24)

24

3. Methodology

For the methodology part, we decided to employ literature review, case study and survey to figure out our questions. By reviewing literatures, we came up with the list of attributes that are vital in the decision-making process. For the case study, two large procurement projects were studied to explore the innovation adoption decision-making process as well as position of support staff in the decision-making process. In this way, a basic framework of innovation adoption decision-making process was formed, laying a foundation for the further study. With regard to survey, the ten most important attributes influencing the decision-making in the innovation adoption process were identified and prioritized by the support staff.

In this chapter, the research design and methodology used in the thesis was illustrated. The design was closely related to the research sub questions. Firstly, the list of innovation attributes was confirmed through literature review to answer the first sub question. Secondly, case study aiming at answering the second sub question was performed to gain an in-depth insight into the making (buying) process. Through the case study, the innovation adoption decision-making process was identified, and influences of support staff on the decision-decision-making (buying) process were illustrated. Thirdly, the survey was to figure out the core attributes when support staff makes decisions on the innovation-adoption of user-specific medical devices. The main goal is first to consider a broad set of innovation attributes, and then find the most critical attributes by generating priorities. By combining the results of case study and the outputs of survey, the research question was answered.

Figure 3.1 Overview of methodology

(25)

25

3.1 Literature review

In order to answer the first sub question “What innovation attributes influencing the individual

innovation adoption decision have been identified in the literature”, various articles and

literatures were reviewed. The literature review provided a fundamental framework for the case study. A preliminary overview of innovation attributes influencing the individual innovation adoption-decision was identified.

The literature generally indicated five categories of determinants influencing the overall innovation adoption-decision in healthcare provider organizations: individual, social, innovation, organizational and environmental determinants. In this survey, the focus was on one specific group of determinants: the innovation attributes. A few reasons accounted for this focus: Firstly, innovation attributes played a leading role. As stated by Rogers (2003), the variance in the adoption rate is mainly explained by the five perceived characteristics of the innovation itself. The individual, internal and external characteristics seemed to have less influence. Secondly, as the unit of analysis in this survey was the individual support staff in UMCG, all other groups of determinants, such as the individual and organization determinants, remained the same. In every discrete innovation adoption-decision, only the innovation attributes differed. Finally, only a very limited number of studies were found investigating the innovation attributes and their hierarchy. Despite the importance of the innovation attributes, in the review of existing research on the adoption of innovation by healthcare provider organizations, Rye and Kimberly (2007) found the innovation attributes were the least examined category of all groups of determinants.

(26)

26

create a long-list of 27 innovation attributes derived from literature review. (See appendix A and C)

3.2 Research Case Study

To answer the second sub question, we decided to use a comparative multiple case study to shed light on the whole decision-making process of medical innovation adoption, discussing the impacts of support staff and visualizing the decision-making process of innovative generally-used medical device. The reasons behind the choice were threefold. First of all, this is an open research with a “how” question, for which case study is the most suitable methodology. Second, it puts emphasis on the natural setting in which the phenomena under investigation occur (Eisenhardt and Graebner, 2007). Finally, it offers a good way to study phenomena in highly complex settings (Stuart et al., 2002). These three reasons necessarily indicate that an explorative case study is the most suitable methodology for our research.

The application of multiple cases has quantities of advantages. First, the findings of multiple cases are more grounded, more accurate, and more generalizable compared to single case studies (Eisenhardt and Graebner, 2007). Each case performs as a distinct analytic unit/experiment, serving as a replication, contrast, and extension to emerging theory (Yin, 1994). Second, multiple cases can both argument external validity and help guard against observer bias (Karlsson, 2010). Moreover, comparisons across cases enable investigator to ascertain whether a finding is constantly replicated by several cases or if it is idiosyncratic to a specific case (Eisenhardt, 1991). Finally, research questions and theoretical elaboration can be explored more broadly.

In this case study, two procurement projects were studied, based on which the support staffs who are involved in the innovation adoption decision-making process were identified. Besides, a list of potential participants was provided by the interviewees for the later questionnaires. 3.2.1 Case selection and case descriptions

(27)

27

introduction of patient-safety needles. Detailed information (like quantity, type, etc.) was offered by the principal of these two projects, as shown in the Table 1:

Project Safety needles Beds

The quantity 95000 850

The type Introcan Safety 3W, BBraun HR900, Hill-Rom

Reason to adopt it During the verification period, users have shown that the needle meets the predetermined requirements

During the verification period, users have shown that the bed meets the predetermined requirements

Time period (start-end) January 2012 – September 2013 January 2012 – October 2014

Table 1: Description of two purchasing projects

The reasons why we decide to perform case study based on these two procurement projects were twofold. Firstly, these two large-scale projects involved various stakeholders in the decision– making process, which make these two projects more representative compared to other small-scale procurement projects. Secondly, as one of the largest hospitals and an economic engine for the Northern region, Medical Center of Groningen (UMCG) is a typical and representative hospital in the Netherland. Compared with other small healthcare organizations, its medical system and decision-making process are more sophisticated. So the results attained from the case study of UMCG projects are relatively more valuable and convincing.

3.2.2 Data collection

Data was collected primarily through structured interviews. An interview protocol with structured questions was developed. It covered questions about the general aspects of the innovation adoption process, and specified questions involving jobs and responsibilities of support staffs, the role they played in purchasing projects, perception of their role and their experiences with the innovation adoption decision-making process. Most interviews were conducted within one week of April 2014.

(28)

28

In order to get an impression of the position of support staff in the decision-making process of innovative medical device, respondents were asked about their perceptions of their participation degree as well as their influences on the decision-making. By taking snap-shots of the role of support staff played in the innovation adoption process, we could construct a view of how support staff with different responsibilities involved in the decision-making process. Moreover, experiences of being involved in the innovation adoption decision-making process have been discussed with the informants. Examples such as in which project the innovative medical device was adopted even though they were not in favor, and the opposite, in which project the innovative medical device was refused even though they were in favor of it, were provided in retrospect by the participants. The use of multiple informants made it possible to have an overview of the decision-making process of innovation adoption in healthcare industry and served the purpose of figuring out how important of support staff as stakeholder in making a decision on whether or not to introduce an innovative medical device. Furthermore, it also helps to decrease the possibility that the different informants with different viewpoints would “engage in convergent retrospective sense making” (Eisenhardt and Graebner, 2007).

3.2.3 Rigor

The criteria for rigor used in this study are credibility, transferability, dependability, and confirmability (Guba & Lincoln, 1989). In order to ensure credibility of findings, data were collected from many participants, and debriefing was used among members of the research team to discuss and challenge findings. Descriptions were obtained and studied, encompassing numerous direct quotes from informants to assist readers in assessing the transferability of the results. Finally, an audit trail of the process of data analysis and emerging themes was kept to promote dependability and confirmability.

Trustworthiness criteria Original criteria

Phase of research

Measure taken in this study

Credibility (Ensures that findings are congruent from the perspective of participants)

Internal validity

Data collection

 Earlier familiarization of culture of UMCG

(29)

29 Composition

external archival data resource

 Informants reviewed draft

 Peer debriefing through two supervisors

Transferability (Degree to which findings can apply or transfer in a different project)

External validity

Data analysis

Composition

 Establish a chain of evidence of coding

 Provided detailed description of two projects wherever possible without comprising anonymity of participants

Dependability (Assesses the consistency of data collection, data analysis, and theory generalization over time)

Reliability Data collection

Composition

 Applied case study protocol

 Developed case study database

 Gave thick description of rationale and methodology employed

Confirmability Objectivity Composition  External audit through the discussion of findings with the researchers who are not relative to this study at conference

 Peer debriefing trough two supervisors

 Listed shortages of this research presented via limitations of this paper

Table 2: Trustworthiness assessment of this study (adapted from: Creswell, 2003; Lincoln & Guba, 1985; Yin, 2009)

3.2.4 Quality criteria

Data were collected from multiple sources. Interviewer was responsible for communicating with participants and taking notes which would serve as a chain of evidence. At the end of each interview, the participant was asked to take a look at the notes so as to ensure that there was no misunderstanding and mistakes. After all the interviews were finished, all the data was inserted into a word file to develop a database for the following data analysis, which was to ensure the reliability of the collected data. In addition, notes were kept of possible themes in the data. 3.2.5 Data analysis

(30)

30

of knowledge that is required for cross-case analysis (Voss et al., 2002).The within cases analyses were followed by cross-case analysis. The core advantage of cross case analysis rest with that evidence can be reviewed from single cases through multiple lenses, forcing investigators to look beyond initial impressions through the systematic search for patterns (Eisenhardt, 1989). This helps to mitigate the risks that the meaning of findings from the single case analysis be exaggerated. Moreover, if findings are corroborated among cases, generalisability of results obtained from single cases can be enhanced (Voss et al., 2002; Eisenhardt, 1989).

In our case study, no priori hypotheses were made. Data were analyzed by first studying two procurement projects to come up with the innovation adoption decision-making process. Tables and quotations were created to facilitate further comparisons between and among cases to identify similarities and differences between them. These were then used to develop emerging constructs and theoretical logic. This shed light on how support staff affects innovation adoption decision-making and why they have different impacts.

3.2.6 Findings

In this section, emerging insight was obtained through analyzing the data iteratively. Analysis of data embraced three steps. The first step was concerned with the innovation adoption making process. The second step sought to explore the influence of support staff in decision-making process. The third step aimed to come up with new innovation attributes by consulting support staff informants. Results gained from the overall analysis were presented as follows. First, the innovation adoption decision-making process was elaborated in detail, and visualized in the form of a graph (see Figure 3.2). This was followed by coding the data to cross case analysis. This section ended with the identification of additional attributes by deleting the repetitive attributes, adding new attributes and improving existing attributes through appending instances or elaborations.

3.2.6.1 Step1: explore the innovation adoption decision-making process

(31)

31 Group 1 Group 2 Specify the package of demands & requirements

Testing and evaluating the products of potential suppliers First choice of supplier Verification of first choice Group 3 GO NO GO Evaluation of products offered by chosen supplier Stage 1: Stage 2: Stage 3:

Figure 3.2: Innovation adoption decision-making process

(32)

32

provided by the principal, based on which informants for later individual interviews were confirmed.

3.2.6.2 Step 2: analyzing the influence of support staff on decision-making

In the second step, nine separate interviews were conducted with support staffs who were involved in either the project of safety needle or the introduction of beds or both.

Although all the informants are support staff in UMCG, the job and responsibility of individual informant varies. These nine support staff interviewees come from different departments ranging from the medical technology department to the Prikpoli department. The type of their responsibility is broad, which range from providing advices on the new projects:

“when new medical device is introduced, I need to check if the new medical ingredients are suitable to the regulations…also give advices on risks that caused by the new techniques or products” (Interviewee A, staff member of medical technology department; beds), to organizing and supervising blood collection: “responsible for the organization of blood collection”

(Interviewee I, manager of Prikpoli; safety needles). The detailed information can be seen in the Appendix B.

Regarding the role of support staff in innovation adoption decision-making process, it could be information provider, decision maker or gatekeeper. As the information provider, they may help to specify the package of demands. “…list our requirements, and arrange the people who give

training to my colleagues. Then we evaluate these products (offered by potential supplier) and give our preferences of products” (Interviewee H, coordinator of technologists for nuclear medicine; safety needles).

As the decision maker, they act as the representative of their own department and give the preference of potential supplier of this innovative medical device:“we give our preference of

supplier.” (Interviewee D, coordinating nurse of the team for medical devices of the Intensive Care organization; safety needles & beds). Besides, reason for their preference is offered simultaneously.

As the gatekeeper, they may evaluate and test the products of chosen supplier and make a

go/no-go decision: “we test the supplier’s products in practice, and decide if they can be adopted”

(33)

33

Most of support staff (seven of nine interviewees) tend to play at least two different roles in the decision-making process. It depends on the individual job and responsibility. In terms of the satisfaction of their roles, most of the interviewees prefer to express their satisfaction. Whereas two of the interviewees think their roles are not so satisfying. Interviewee A said that the scope of his role is limited. He was not involved in every purchasing project but certain number of projects: “in some cases I was not involved in the decision-making process.

Sometimes I was informed too late.” (Interviewee A, staff member of medical technology department; beds) Interviewee C stated that she had both good and bad experiences in the innovation adoption decision-making process.

One of the main findings concerning the difference in support staff’s impact on the innovation adoption decision is that their jobs and responsibilities are divergent. Experts and professionals of medical device can influence the innovation adoption decision to a large extent. This may due to the fact that they have a good knowledge of European Guideline as well as Regulations, and they are in charge of screening the new medical device. Support staffs of Emergency department and ICU department are also of great importance in the decision-making process, largely owing to the specialty of their departments. Despite the person who has a special responsibility or in specific department, other support staffs who act as coordinators or advisors have certain impacts on the innovation adoption decision. In most cases their impact “depends on the project and what type of products to be introduced”

(Interviewee H, coordinator of technologists for nuclear medicine; safety needles).

(34)

34

CE Mark Regulation, European Guideline and other regulations; checking if training is needed and if supplier will offer such training. Besides, the records of products will be checked”

(Interviewee H, Qualified professional of sterile medical devices; safety needles).

Furthermore, interviewees were being asked if they had such experiences that the innovative medical device was adopted even though they were not in favour of it, and the opposite, the innovative medical device was refused even though they were in favour of it. It seems like support staffs who have limited impact on the decision-making are more likely to have such experience, or vice versa. Various reasons account for this phenomenon. For instance, sometimes people with small influence are not involved in the decision-making process: “sometimes I was

not involved in the decision-making process. When something went wrong, and they came to me for suggestions, it was too late” (Interviewee A, staff member of medical technology department;

beds), or they can’t decide which device to be adopted. And sometimes the preferred supplier is

not able to develop specific innovative device. Just as interviewee C stated: “The proposal is

refused because the current supplier is not able to develop this specific care” (Interviewee C, quality assurance and coordinator in the department of Surgery; beds). On the other hand, costs may also lead to the failure in introduction of innovative medical device. Interviewee I (manager of Prikpoli department; safety needles) confirmed this point.As he said,“hospital may consider many

other factors, like costs, suppliers, etc. They are very important in the introduction of new medical device”. In addition, there also exists a possibility that the features of certain type of medical devices are against specific rules of hospital, contributing to the failure of adoption:

“Nexiva needle was not adopted at last, since the colour of its switching button was against the rule used in hospital” (Interviewee F, Support staff in Emergency Department; safety needles)

3.2.6.3 Step 3: identifying new innovation attributes

(35)

35

repetitive. For example, the proposed attribute just differ in the description with attributes found in the literature, or is one narrow aspect of certain existing attribute. On the basis of this rule, the additional attributes proposed by informants were evaluated and disposed. The overview of proposed attributes and relevant solution for each additional attribute were shown as below:

Table 3: Overview of all additional attributes

Table 4: Treatment scheme for all the additional attributes

The attribute “maintenance free”, which also came from interviewee A, is covered by the existing attribute “appealing to use”. The same goes for the attribute “reusable and reprocessing”, the one proposed by the interviewee E. Each of these two proposed attributes refers to features which contribute to the user-friendliness of innovative medical device. Both of them are added as

Inte rvie we e Ad d itio na l a ttrib ute s

Interviewee A 1. Traceability: the need for tracking and tracing equipment in the clinic 2. Maintenance free:no or low need for maintenance configuration management 3. Patient safety issue:no differences between the same type of equipment all over the clinic Interviewee B 1. Necessity or importance: how important or necessary of this innovation

2. Scale of improvement: to what extent the same kind of medical device improved

Interviewee C 1. Stress reduction : reduce the stress of patients from psychological, physical and financial perspectives 2. Unambiguity: the innovation is clear to be used and is not contradictory with the use of other devices 3. Research creditability: the outcomes of this new medical device have already been published in journals Interviewee D 1. End-user safety

2. Patient safety

Interviewee E 1. Reusable and reprocessing: Innovation can be easily cleaned and sterilized Interviewee F 1.Safety to end-user : the new device ensure the end-user work safe

2.Good aftersales: supplier are supposed to care about the users’ experience with their medical device, maybe make improvement accordingly Interviewee G 1. Adaptable: the innovation is supposed to be flexible to be used on different conditions

Interviewee H 1. Safety to final user

2. Quality of product: the innovation should be stable and can be used in continuous way 3. Continuous supply: the innovation should be available from supplier whenever it needed Interviewee I No

Ad d itio na l a ttrib ute s So lutio n R e a s o n

A1 delete does not belong to the category of innovation attributes A2 added as example covered by the existing attribute “appealing to use” A3 added as example covered by existing attribute “compatibility” and “complexity” B1 delete covered by existing attribute “radicalness” and “frequency of use” B2 delete covred by the existing attribute “radicalness”

C1 added as example covered by the existing attribute “relevance (benefit) for the patients” C2 added as example covered by existing attribute “compatibility” and “complexity” C3 added as example covered by the existing attribute “uncertainty (on the advantages)” D1 added as new attribute new and unique

D2 revised based on the existing attribute existing attribute can be improved tobe more comprehensive E1 added as example covered by the existing attribute “appealing to use”

F1 added as new attribute new and unique

F2 added as example covered by the existing attribute “argumentation support” G1 delete covered by the existing attribute “reinvention”

H1 added as new attribute new and unique

(36)

36

(37)

37

available from supplier whenever it needed”. These two attributes illustrate the existing attribute “argumentation support” (the innovation is supplied as an augmented product) from different perspectives. As a result, both of them were used as the further explanation of attribute “argumentation support”. Interviewee G came up with one additional attribute “adaptable”, which is covered by the existing attribute “reinvention” (the potential adopter can adapt or modify the innovation to own needs). The only difference is that they describe the same issue in different ways.

(38)

38

3.3 Survey

In order to answer the third sub question, an explorative survey research was employed to gain an insight into the set of attributes that were vital at the time when support staff made a decision on the adoption of medical device. This survey provided a basis for more in-depth analysis of these attributes and their influences. There were three reasons for our choice. Firstly, “an exploratory survey can help to determine the concepts to measure in relation to the phenomenon of interest, the best way to measure them and how to discover new facets of the phenomenon under investigation” (Karlsson, 2010). It fits for our case due to the lack of researches on the innovation adoption attributes in healthcare industry. Secondly, an exploratory survey can be used to uncover or provide preliminary evidence of the association among concepts (Karlsson, 2010). Furthermore, the results of survey can be generalized. Last but not least, it helps to explore the validity boundary of a theory (Karlsson, 2010).

This survey encompassed two rounds. The first round was to create a shortlist of the ten most important innovation attributes from the long list of attributes that were identified via literature review and case study. This was followed by employing pair-wise comparisons to rank these ten most important attributes with the method of Analytic Hierarchy Process (AHP).

3.3.1 Unit of analysis

We defined the unit of analysis as support staff in the hospital. Considering that collecting data from support staffs all around the Netherlands is infeasible, we confined the unit of analysis to support staffs in University Medical Centers Groningen (UMCG). The reason why we choose the support staff in UMCG lies in that UMCG is a reputable and typical hospital in Netherlands, combining high-level university patient care with medical research. Besides, due to its location, it is more convenient for interviewers to be there and make face-to-face interviews.

(39)

39

of our survey were chosen. There is no clear criteria have ever been referred to indicate the optimal quantity of participants required in expert session. “Even though increased number of

participants assures the reliability and validity of the findings to some extent, after a certain number of participants the saturation-level is reached, adding even more participants make no contribution to additional findings” (Bax, 2011). Participants of our survey composed of the

support staff informants of case study and their suggested colleagues, amounting to 33 persons. This was supposed to be sufficient. In addition, the possible participators for round one and round two were the same population. The reason to send the second questionnaire to all the potential participants of round one (n=33) instead of only ask the real participators (n=17) lies in that in most cases, more replies can be attained from a larger resource. Furthermore, the results required from larger resource are more likely to be generalized.

3.3.2 Data collection

Data were collected through online questionnaire by using the tool “Survey Monkey”. In terms of questionnaire, it is a research instrument consisting of a series of questions for the purpose of obtaining useful information from respondents. The advantages of questionnaire lie in lower cost and less effort from both the questioner and the answerers. Besides, problems due to group interaction could be avoided. This “prevents the authority, personality or

reputation of some participants from dominating others in the process” (Karlsson, 2010).

Also, respondents are free to decide to remain anonymous. This to some extent prevents respondents from being disturbed by other factors. Because of the advantages as described, the questionnaire is chosen as the instrument to collect data. In this questionnaire, both closed-ended questions and open-ended questions are included. Most of the questions are closed-ended questions. Open-ended questions are framed for the clarification purpose.

(40)

40

both the deadline and contact information of the researcher are included. The second part encompasses some general questions about the respondent and his (her) background, as well as some questions about perceptions of support staff on their participation and the importance of their opinions. The third part involves questions to identify ten most important innovation attributes. The final part involves questions about the readiness to personally clarify the answers given.

In the first round, support staffs were asked to rate ten most important innovation attributes taken into consideration when they made decisions on the adoption of a new and innovative medical device. According to the output of the first round, the second questionnaire was formed. The method of Analytic Hierarchy Process (AHP) was employed to generate priority scales through pair-wise comparisons of factors based on the ten most important innovation attributes sorted by support staffs in the former round.

The questionnaire was sent directly to the business-email addresses of potential respondents from “Survey Monkey” via email invitation. All potential participants did not know who else were involved in this survey. This aims to get rid of the possibility that possible respondents discuss their ideas and opinions with each other. A complete overview of the questionnaire I and the questionnaire II are shown in appendix D and E.

3.3.3 Data analysis methods

In the questionnaire-round, the ten most important innovation attributes were selected by the 33 potential respondents. These ten attributes were expected to have different impacts on the individual innovation-adoption decision of support staffs. Therefore, to determine the priority of each innovation attribute identified as important by support staff in the first round, the method of Analytic Hierarchy Process (AHP) was employed. AHP is a mathematical decision-making method developed by Saaty (2008), aiming at deriving priority scales when multiple criteria must be considered. It is a structured technique of measurement through pair-wise comparisons of factors based on the judgments of support staff. The reason why AHP is sorted is because “the

most effective way to concentrate judgment is to take a pair of elements and compare them with a single property without concern for other properties or other elements” (Saaty, 1990). AHP

INNOVATION ADOPTION CRITERIA OF MEDICAL DEVICES: IDENTIFYING THE MOST CRITICAL INNOVATION ATTRIBUTES FROM SUPPORT STAFF PERSPECTIVE