0
The wonderful world of unsought goods:
research towards the factors that affect the
adoption of phase III clinical trials and the
determinants of retention
1
The wonderful world of unsought goods:
research towards the factors that affect the
adoption of phase III clinical trials and the
determinants of retention
By
Lisanne Buijsse
University of Groningen
Faculty of Economics and Business
Master Thesis Marketing Management
January 2010
Dokstraat 65
3082 RK Rotterdam
+316 136 03 866
lisanne_buijsse@hotmail.com
s1486233
Supervisor RUG:
Dr. Jaap E. Wieringa
2
Management summary
Consumer goods, shopping goods, specialty goods, and unsought goods; these are the four types of consumer goods. Although most people know how to define the first three types of goods, unsought goods are less well-known. They can be defined as ‘goods that the customer does not know about or knows about but does not normally think of buying’. Examples of unsought goods are life insurances and encyclopedias. The goal of this research is to gain more insight into what exactly motivates customers to adopt an unsought good and define the determinants of retention.
In order to do this, both qualitative and quantitative research were performed. Sanofi-aventis, an international pharmaceutical company, was used as a case study with the medicines that patients test in phase III clinical trials as the unsought good. Literature was used to define the factors that affect the adoption of an unsought good and the determinants of retention. This resulted in a conceptual model. Subsequently, qualitative research was performed to determine the correctness and completeness of the conceptual model before testing the variables on a greater scale. For this qualitative research part, ten participants of clinical trials and four health care professionals were interviewed. The interviewees confirmed that the information quality, type of vendor, quality and length of the relationship between the customer and vendor, reputation of the vendor, salesmanship of the vendor, advertising, perceived value, the number of additional services, perceived risk, the opinion of personal sources such as family members and/or friends, the negative reputation of the pharmaceutical industry, and whether a person is doing volunteer work or not all affect the adoption of a phase III clinical trial. Customer satisfaction, the quality and length of the relationship between the customer and vendor, the quality of the additional services, and the opinion of personal sources such as family members and/or friends were defined as determinants of retention.
3
vendor’ had to be rephrased into ‘commitment towards the vendor’ and ‘vendor satisfaction’. After that, the variables were tested with the use of a binary logistic regression analysis. The results of this analysis showed that perceived risk negatively affects the adoption of a clinical trial. Commitment towards the vendor, vendor satisfaction, perceived value, the number of services that accompanies the good and whether a person is doing volunteer work or not all positively affect the adoption of a phase III clinical trial. The opinion of personal sources tended to be a mediator of perceived value; the perceived value of patients increases once they have had a positive conversation with their family members and/or friends about the clinical trial. Although the type of vendor did not significantly affected the adoption of a clinical trial, more than half of the participants indicated that they would not have participated in the clinical trial if a research nurse had asked them. The same goes for non-participants; more than half of them would have participated in the trial if a physician instead of a research nurse had asked them. Unfortunately, it was not possible to identify the determinants of retention, since the sample included participants who had the intention to finish the entire trial only. Therefore it was impossible to conduct a binary logistic regression analysis.
4
Preface
This thesis is written to complete the Master of Science in Business Administration at the University of Groningen. It was a long road, but a very rewarding journey! During the past months, I have learned a lot and was able to develop both my academic and personal skills. Of course I could not have written this thesis without the help of a number of people.
First of all I would like to thank the direction of the Clinical Research Unit (CRU) of sanofi-aventis Netherlands B.V. for giving me the opportunity to conduct this research. I have experienced my internship at the CRU as a very pleasant and interesting period. Special thanks go out to Mirjam Brijker – de Jong, my supervisor at sanofi-aventis, for her guidance during the process. Without her advice and constructive feedback, it would not have been possible to finish this thesis. In addition, I would like to express my gratitude towards Terttu Haring, the director of the Clinical Research Unit, for providing the recourses necessary to complete this research. I also want to thank all my colleagues from the CRU for their interest, support, and advice.
Besides my colleagues at sanofi-aventis, many thanks go out to Jaap Wieringa from the University of Groningen for his feedback, support, and patience. I have considered my meetings with him as very inspiring. Moreover, I would like to thank Marjolein Achterkamp for reviewing this thesis.
Last but not least I would like to thank my parents and friends for their encouragement and interest. Without their love and (financial) support, it would not have been possible to complete my study period in Groningen.
Lisanne Buijsse
5
Table of contents
Management summary ... 2
Preface ... 4
Chapter 1 Introduction ... 8
1.1 Problem statement
... 9
1.1.1 Research objective ... 9 1.1.2 Research question ... 9 1.1.3 Sub questions ... 91.2 Theoretical relevance
... 9
1.3 Social relevance
... 10
1.4 Structure of the thesis
... 10
Chapter 2 Literature review ... 11
2.1 Definition and characteristics of an unsought good ... 11
2.2 Factors affecting the adoption of an unsought good
... 13
2.2.1 Information transmission ... 14
2.2.2 Vendor ... 14
2.2.2.1 Type of vendor ... 15
2.2.2.2 The interpersonal relationship between customer and vendor ... 15
2.2.2.3 Reputation of vendor ... 16 2.2.2.4 Salesmanship of vendor ... 17 2.2.3 Advertising ... 17 2.2.4 Perceived value ... 18 2.2.5 Service provided ... 19 2.2.6 Perceived risk ... 19
2.2.7 Influence of personal sources (i.e., family or friends) ... 20
2.2.8 Environmental factors: the influence of the media on an industry’s reputation ... 20
2.3 Determinants of customer retention
... 21
2.3.1 Customer satisfaction ... 21
2.3.2 Interpersonal relationship between customer and vendor ... 22
2.3.3 Service provided ... 23
6
Chapter 3 Research design: qualitative research ... 26
3.1 Case study description ... 26
3.1.1 Sanofi-Aventis ... 26
3.1.2 Sanofi-Aventis Netherlands B.V. ... 27
3.2 Data collection and sample description ... 27
3.2.1 Patients ... 28
3.2.2 Health care professionals ... 29
3.3 Data analysis ... 29
Chapter 4 Qualitative research results ... 30
4.1 Factors that affect the adoption of an unsought good ... 30
4.2 Determinants of retention... 34
4.3 Additional variables ... 35
4.4 Conceptual model ... 36
Chapter 5 Research design: quantitative research ... 38
5.1 Data collection and sample description ... 38
5.2 Questionnaire ... 39
5.3 Data analysis ... 40
Chapter 6 Quantitative research results... 42
6.1 Characteristics of the respondents ... 42
6.2 Factors affecting the adoption of an unsought good ... 43
6.2.1 Factor analysis and scale reliability ... 43
6.2.2 Binary logistic regression analysis ... 45
6.2.2.1 Correlation matrix ... 45
6.2.2.2 ‘Best’ regression model ... 45
6.2.2.3 Hypothesis testing ... 50
6.3 Determinants of retention... 54
6.4 Overview of the quantitative results ... 56
Chapter 7 Conclusions ... 57
7.1 Definition and characteristics of an unsought good ... 57
7.2 Factors affecting the adoption of an unsought good ... 57
7
Chapter 8 Recommendations, limitations & directions for future
research ... 59
8.1 Recommendations ... 59
8.2 Limitations ... 61
8.3 Directions for future research ... 61
8
Chapter 1 Introduction
Pills, ointments, powders, and injections; medicines are available in all shapes and sizes. Although most people know how to acquire them, relatively few are aware of the fact that developing new medicines is a difficult and time-consuming process (Nefarma Farmafeiten 2008). For example, it takes an average of 12 years before a drug is developed and introduced in the market (Nefarma Farmafeiten 2008). The development process starts when researchers discover an active chemical. In order to know whether this active chemical is safe and efficient, clinical trials are conducted. During these trials, the new medicine is tested in three different groups: laboratory animals, healthy volunteers, and patients. Especially the third phase - in which the medicine is tested in patients - is challenging, as it is rather difficult to recruit patients for clinical trials. There may be several reasons for this issue.
First, phase III clinical trials, or, in other words, the medicines that are tested in such trials, can be classified as unsought goods, meaning that most patients are not aware of the existence of these trials. As a result, physicians or research nurses1 have to make patients aware of the possibility to participate in such a trial in order to try these new medicines. Second, some patients are scared to participate in a clinical trial, as such trials are often associated with high risks and participants are frequently compared with guinea pigs by outsiders. Third, in Western Europe patients only get reimbursed for costs incurred, so there is no financial incentive that could act as a reason to participate in a clinical trial. Due to these difficulties, recruitment deadlines are often postponed which causes delays. Delays in the recruitment may result in higher project costs (Lovato et al. 1997) and could decrease the enthusiasm of staff and participants (Richardson et al. 1998).
In addition to the recruitment problems, it is also challenging to keep patients motivated to finish the entire clinical trial. Clinical trials can be quite time-consuming and a number of patients may deal with unpleasant side-effects. As a result, some patients decide to quit. Of course, the more patients decide to quit, the higher the drop-out rate will be. This is unfortunate, as high drop-out rates greatly weaken the
1
9
statistical power of the study. Therefore, gaining more insight in the patient recruitment and retention process is extremely important for solving these issues. Understanding why patients decide to participate in and stay loyal towards a clinical trial can help in the formulation of future marketing strategies for recruiting and retaining patients.
1.1 Problem statement
In order to put the results in a broader perspective, this research focuses on unsought goods. The new medicines or improved versions of existing medicines that are tested in phase III clinical trials are an example of an unsought good. Although participating in a phase III clinical trial might be seen as patients providing a service, it should be considered as a good, since both physicians and researchers from the pharmaceutical industry often classify the medicines that patients test as a full-fledged replacement of their existing medication (Nefarma Farmafeiten 2008).
1.1.1 Research objective
The objective of this research is twofold: i) gaining more insight into what exactly motivates customers to adopt an unsought good and ii) investigating which factors determine whether customers stay loyal (or not) towards this type of good.
1.1.2 Research question
“How can customers be stimulated to adopt an unsought good and which actions should be undertaken to ensure that customers stay loyal towards this type of good?”
1.1.3 Sub questions
1) How can unsought goods be defined and what are its characteristics?
2) Which factors affect customers’ response with regard to the adoption of an unsought good?
3) Which factors determine whether customers stay loyal (or not) towards the good?
1.2 Theoretical relevance
10
Lennon and Stoel 2005; Page and Luding 2003; Monat 2009), perceived value (Zeithaml 1988; Chang and Wildt 1994; Tarn 1999), and the reputation of the vendor (Yoon and Kijewski 1993). Besides the existing literature on buying behavior, a number of articles with regard to customer retention have been published. For example, customer satisfaction (Gustafsson, Johnson and Roos 2005), trust (Ranaweera and Prabhu 2003), and service quality (Venetis and Ghauri 2004) all positively affect customer retention rates. It is, however, not clear if these factors also apply to unsought goods, as researchers who were involved in these studies did not make use of unsought goods when testing the variables. This research will therefore contribute to existing literature by defining which factors affect customers’ response regarding the adoption of an unsought good. In addition, the determinants of retention are defined.
1.3 Social relevance
One of the core activities of the pharmaceutical industry is the development of new medicines. In order to develop new medicines, it is essential to perform clinical trials. It is impossible to perform these trials without having a large group of patients who are willing to participate in it. Having knowledge on what motivates patients to join a clinical trial as well as knowing what can be done to make sure that these patients finish the entire trial, makes the execution of a phase III clinical trial much easier and future delays in the development of new medicines may become less likely.
1.4 Structure of the thesis
11
Chapter 2 Literature review
This chapter contains an overview of the literature regarding unsought goods, (new) product adoption, and customer retention. As there is relatively little marketing literature regarding unsought goods, the first section comprises of the definition and characteristics of this type of good. The second paragraph describes the factors that can affect the adoption of an unsought good. In addition, the determinants of retention are discussed in the third section. The chapter ends with a conceptual model in which the findings of the previous paragraphs are summarized.
2.1 Definition and characteristics of an unsought good
Consumer goods can be classified into convenience goods, shopping goods, specialty goods, and unsought goods. In marketing terms, the description of an unsought good is as follows: ‘A good that the customer does not know about or knows about but does not normally think of buying’ (Kotler 1994: 436). Examples of unsought goods are life insurances, encyclopedias, and new products involving a large innovative element. Table 1 shows how unsought goods differ from other types of consumer goods in terms of purchase behavior, brand loyalty, and the marketing mix.
Table 1: Classification of consumer goods (Kerin, Hartley and Rudelius 2003:228)
BASIS OF COMPA-RISON CONVENIENCE GOOD SHOPPING GOOD SPECIALTY GOOD UNSOUGHT GOOD Purchase behavior of consumers Frequent purchases; little time and effort spent shopping Infrequent purchases; needs much comparison shopping time Infrequent purchases; needs extensive search and decision time
Very infrequent purchases, some comparison shopping Brand loyalty of consumers
Aware of brand, but will accept
substitutes
Prefer specific brands, but will accept substitutes
Very brand loyal, will not accept substitutes
Will accept substitutes
Product Toothpaste, cake mix, hand soap
Cameras, TVs, clothing
Rolls Royce cars, Rolex watches Life insurance, encyclopedias, products involving a large innovative element Price Relatively inexpensive
Fairly expensive Usually very expensive
Varies
Place Widespread; many outlets
Large number of selective outlets
Very limited Often limited
Promotion Price, availability, and awareness stressed Differentiation from competitors stressed Uniqueness of brand and status stressed
12
In contrast to other types of consumer goods, unsought goods are available at a limited amount of outlets and purchases are often very infrequent. In addition, customers are not very brand loyal, as they are willing to accept substitutes as well.
Besides the differences in purchase behavior, brand loyalty, and the marketing mix, unsought goods also differ from other types of consumer goods in terms of the customer decision process. Normally, customers progress through five stages during the customer decision process: problem recognition, search for solutions, evaluation of alternatives, choice, and outcomes (Engel, Blackwell, and Kollat 1978: 224). The customer decision process regarding unsought goods is different in the sense that the process is entirely managed by marketers. Table 2 shows the role of the marketer in every stage of the process.
STAGE TECHNIQUES FOR MANAGEMENT
Need recognition
Sales representatives activate the need by asking questions about the individual’s current circumstances.
Information search
The sales representative performs a lengthy presentation in which the product features and benefits are explained. Questions and objections are dealt with as they arise.
Evaluation of alternatives
The salesperson ‘outsells’ competitors by pointing out ways in which their products are inferior.
Choice decision
The salesperson closes the sale by using a phrase or technique that elicits a yes or no decision.
Post-purchase evaluation
The company may follow up on the sale by sending comment forms, the salesperson may call back to elicit feedback, or a sales manager may call back. In the case of buyer’s remorse, there will be a return visit by the salesperson or sales manager.
Table 2: Decision-making model for unsought goods (Blythe 2007:311)
13
not to adopt the good. If the customer has doubts, the marketer should try to overcome these doubts, for example by offering after-sale services. In the last stage, the marketer needs to assess if the customer is satisfied with the good and undertake action if this is not the case.
Although the decision-making model of unsought goods looks relatively straightforward, it is quite challenging to market unsought goods. The purchase of such a good is often associated with relatively high risks (Tait and Walker 2000) and customers do not have the intention to buy this type of good. This is why unsought goods are usually marketed with use of promotion methods depending on both personal recommendation and personal selling (Tait and Walker 2000) as well as aggressive advertising (Kotler et al. 1998).
2.2 Factors affecting the adoption of an unsought good
Understanding why and how customers make purchase decisions is vital to maximizing profitable sales, developing new products, growing companies, and designing predictive sales models (Monat 2009). This paragraph discusses the factors that influence the adoption of an unsought good.2 As it is rather difficult to market unsought goods, knowledge of the factors that can affect the adoption of such a good can help marketers in developing a more effective future marketing strategy for this type of good. There are numerous factors that can affect the decision of whether or not to adopt a certain good or service. This paragraph starts by explaining how the transmission of information influences the adoption of an unsought good. In the second section, the role of the vendor is described in more depth. The third section deals with the influence of advertising. Subsequently, the effect of perceived value, additional services, and perceived risk on the adoption process is discussed. This is followed by a section in which the influence of personal sources is explained. The paragraph ends with describing the impact of the environment (in this case the media) on the adoption of an unsought good.
2
14
2.2.1 Information transmission
The search for information is one of the first stages in the purchase decision process. It is therefore important for marketers to have knowledge about information acquisition strategies. ‘Information’ can be defined as data that is organized in a certain way or that is given structure – that is, placed in context – and thus endowed with meaning (Glazer 1991). The means by which information about a product or service can be transmitted towards customers consist of both the mass media and interpersonal communications. The search for information is especially important for customers who are planning on adopting a new, innovative product or a product that is associated with high risks. These customers often search for information to reduce the degree of perceived risk (Mahajan, Muller and Bass 1995). Unsought goods meet these two criteria. This indicates that the product information customers receive could affect the adoption of such a good. For example, when customers receive irrelevant information of low quality and do not understand the content, chances are that the good will not be adopted. This leads to the first hypothesis of this research:
2.2.2 Vendor
The vendor plays an important role in the adoption process of an unsought good. This could be explained by the fact that the promotion of unsought goods usually takes place through personal selling. Personal selling is a rather effective marketing mix component. It has a number of advantages. First, it is a two-way form of communication. The vendor can accommodate the message when receiving feedback from the message receiver (i.e., the customer) and vice versa. In addition, personal selling creates a basis of trust through which the client’s perception of risk will decline. In some cases, personal selling is done by the manufacturer. In other cases, a third party (i.e., an intermediary) is involved. This is also the situation in our case study. As pharmaceutical companies are not allowed to contact patients themselves, the physician or research nurse fulfills the role of a third party vendor.
15 2.2.2.1 Type of vendor
A typical sales force could be segmented in several ways. Both age and current performance are examples of possible segmentation bases. Another way of segmenting a sales force is by status. Some customers are very sensitive to the status of a vendor. Such customers attach importance to power, which includes a high status, respect, knowledge, wealth, and more privileges (Patterson, Cowley and Prasongsukarn 2006). This could influence the adoption process, as customers who attach greater importance to status are probably more likely to adopt a good or service when it is offered by a vendor who has a high position in the organizational hierarchy. I therefore specify the second hypothesis as follows:
2.2.2.2 The interpersonal relationship between customer and vendor
The field of relationship marketing has been investigated by quite some scholars, especially in service and channel settings. More and more companies attempt to build profound, meaningful, long-lasting relationships with their customers. The interactions customers have with companies have evolved from impersonal economic exchanges to participation in long-term relationships with key internal stakeholders, such as management or employees (Bhattacharya and Sen 2003). The relationship with the vendor plays an important role in the adoption of an unsought good. A ‘close’ relationship with the vendor can decrease the amount of perceived risk and can thus positively influence the adoption of an unsought good. Both the quality and length of a relationship determine whether the relationship is close or not.
Quality of the interpersonal relationship between customer and vendor
Relationship quality is defined as the ‘degree of appropriateness of a relationship to fulfill the needs of the customer associated with the relationship’ (Hennig-Thurau and Klee 1997). High relationship quality means that the customer is able to rely on the vendor’s integrity and has faith in the vendor’s future performance because the level of past performance has been consistently satisfactory (Crosby, Evans and Cowles 1990). Customer satisfaction (Crosby, Evans and Cowles 1990), commitment to the vendor (Hennig-Thurau, Gwinner and Gremler 2002), and trust (Crosby, Evans and Cowles 1990; Wong and Sohal 2002) are often used to measure the quality of the
16
relationship between the customer and vendor. As both trust and commitment to the vendor are directly linked with purchase intention (Macintosh and Lockshin 1997), it is assumed that the quality of the relationship between the customer and vendor positively influences the adoption of an unsought good. This leads to the following hypothesis:
Length of the interpersonal relationship between customer and vendor
Literature suggests that the duration of the relationship between two parties can influence future purchase intentions (Kalwani and Narayandas 1995; Noordewier, John and Nevin 1990). In some cases, the vendor has known the customer for quite a while. When the vendor subsequently offers the customer a good or service, it is possible that this customer feels more or less obliged to adopt the offer, because he or she has known the vendor for such a long time. This leads to the fourth hypothesis:
2.2.2.3 Reputation of vendor
A good reputation is priceless. Friedrich Nietzsche, a German philosopher, once said that ‘it is easier to cope with a bad conscience than with a bad reputation’. Reputation can be described as a global perception of the extent to which an organization or person is held in high esteem or regard (Weiss, Anderson and MacInnis 1999). It should be emphasized that reputation is not the same as image. Although both terms are conceptually the same, image reflects what a firm (or person) stands for whereas reputation indicates how well this firm or person has done in the eyes of the customer (Weiss, Anderson and MacInnis 1999). Customers use reputation as a means of inferring quality of the product or service (Herbig and Milewicz 1995). As a positive reputation could affect customers’ buying intentions (Yoon and Kijewski 1993), reputation could be a decisive factor in the adoption process of a good or service. I therefore specify the fifth hypothesis as follows:
H3: The quality of the interpersonal relationship between the customer and vendor positively affects the adoption of an unsought good.
17 2.2.2.4 Salesmanship of vendor
The characteristics, competences, and skills of a vendor play an important role when selling a good or service, as an effective sales performance can enhance a customer’s purchase intention (Szymanski 1988). But what is the definition of a “good” salesperson?3 According to Kennedy, Ferrell and LeClair (2001), a competent salesperson should have enough product or service knowledge in order to effectively perform the personal selling function. A salesperson should be able to tell the customer what the advantages and disadvantages of the product or service are and why he or she has to acquire the product. Besides having enough product or service knowledge, effective salespersons also have richer and more interrelated knowledge structures about their customers than do less effective salespersons (Sujan, Sujan and Bettman 1988). Listening skills are another important aspect. Listening is a higher-order construct composed of three dimensions: 1) sensing, 2) evaluating, and 3) responding (Ramsey and Sohi 1997). Salespersons who understand the art of effective listening achieve higher performance goals (Sujan, Weitz and Kumar 1994). Time spent on the job also influences a salesperson’s performance. More time selling and serving clients, along with total time spent on the job, leads to a higher level of performance (Brashear et al. 1997). Based on these outcomes, I specify the sixth hypothesis as follows:
2.2.3 Advertising
Besides personal selling, unsought goods could also be promoted by means of advertising. Advertising is defined as information paid by an advertiser about brands, organizations or ideas in order to influence the knowledge, attitude, and/or behavior
3
Personality traits are not included in this research, as scholars often question the validity of personality trait measures.
H6: The vendor’s degree of salesmanship positively influences the adoption of an unsought good.
18
of the target group in the suitable direction for the advertiser (Floor and Van Raaij 2003). Research has shown that advertising is especially useful in case of products that are associated with relatively high risks. According to Byzalov and Shachar (2004), exposure to advertising increases customers’ tendency to purchase the promoted product because the informative content of advertising resolves some of the uncertainty and thus reduces the risk associated with the product. Based on this, the assumption that advertising initiates the interest in an unsought good is easily made. I therefore specify the seventh hypothesis as follows:
2.2.4 Perceived value
Perceived value is seen as ‘benefits received relative to costs’ (Zeithaml 1988). In more simple terms, perceived value is the difference between perceived benefits and costs (McDougall and Levesque 2000). Perceived value can differ widely from one customer to another, as value is highly personal and idiosyncratic (Zeithaml 1988). It is worth noting that the term perceived value should not be confused with customer satisfaction (which is defined as meeting the needs of a customer). Even though the constructs are distinct, perceived value occurs at various stages at the purchase process, including the prepurchase stage, whereas satisfaction is a postpurchase evaluation (Sweeney and Soutar 2001).
During the 1990s, perceived value was a very popular marketing concept and it is nowadays seen as one of the driving forces of customer purchase decision making (Zeithaml 1988; Chang and Wildt 1994; Tarn 1999). It is therefore included as a factor that could affect the adoption of an unsought good:
H7: Advertising initiates the interest in an unsought good and could thus indirectly affect the adoption of such a good.
19
2.2.5 Service provided
Levitt (1969) introduced the concept of the augmented product, stressing that consumers are interested in the total buying experience, not just the core product. Nowadays, manufacturers selling goods use services to augment their product offerings (Tarn 1999). These services are offered in addition to the core product. Examples are after-sale services, financial services, and training services. Manufacturers are doing this to build a loyal customer base, as not offering a product augmented by service support often creates customers’ disappointment and jeopardizes long-term commitment (Asugman, Johnson and McCullough 1997). By offering additional services, the threshold to adopt an unsought good could be reduced. This leads to the following hypothesis:
2.2.6 Perceived risk
Customers perceive various risks when buying a product or service. This varies from relatively low risks (i.e., the purchase of hand soap) to high risks (i.e., insurance purchases). Functional risk, performance risk, physical risk, psychological risk, social risk, and financial risk all influence customer decision making (Tsiros and Heilman 2005). The relation between perceived risk and purchase intention has been researched by several scholars in various settings. In a study towards online apparel shopping, Park, Lennon and Stoel (2005) discovered that there is a negative relationship between perceived risk and apparel purchase intention. In a study among bank customers, results revealed that perceived risk influences customers’ intention to purchase (Page and Luding 2003). According to Monat (2009), risk perception is the most important determinant in industrial customer decision making. As the adoption of unsought goods is associated with relatively high risks, chances are that customers will not adopt the product if perceived risks are too high. I therefore specify the tenth hypothesis as follows:
H9: The number of services accompanying the unsought good positively affects the adoption of such a good.
20
2.2.7 Influence of personal sources (i.e., family or friends)
During his or her life, a person buys numerous products and services. Individuals who face difficulties with the decision-making process may prefer to ask advice from family members or friends (Furse, Puni and Stewart 1984). This is often the case when the purchase of a good is associated with a high degree of perceived risk. Mitra, Reiss and Capella (1999) postulate that the higher the perceived risk in a purchase decision, the greater the importance of personal sources in making the decision. This makes it plausible that customers, who are thinking of adopting an unsought good, first consult their personal sources before making a decision. The opinion of family members or friends can either positively or negatively influence an individual’s purchase decision. This leads to the following hypothesis:
2.2.8 Environmental factors: the industry’s reputation
The reputation of an industry can be defined as the status ascribed to the industry by outsiders (Fombrun and Shanley 1990). An industry’s reputation can be affected to some extent by the media. According to Mahon (2002), the media’s role in underscoring the reputation of the firm or industry cannot be overestimated. In some cases, customers do not have direct experience with the firm or industry. When an issue arises, these customers will rely on others to supply them with information about the reputation of the firm or industry. That “other” could be the media. It is no secret that the reputation of the pharmaceutical industry is sometimes negatively affected by items in the media. As perceptions of corporations play an important role in purchase decisions (Yoon and Kijewski 1993; Fombrun 1996), it is assumed that this goes for the perception of an industry as well. In this research, this influence is negative, leading to the following hypothesis:
H12: A negative industry reputation negatively influences the adoption of an unsought good.
21
2.3 Determinants of customer retention
According to Gerpott, Rams, and Schindler (2001), customer retention is concerned with maintaining the business relationship established between a supplier and a customer. Customer retention has often been conceptualized and operationalized as a dimension of the customer loyalty construct (Boulding et al. 1993; Zeithaml, Berry and Parasuraman 1996). It is important for companies to retain their customers, as the cost of winning a new customer is about five times greater than the cost of retaining a current customer (Rosenberg and Czepiel 1984). Moreover, according to Reichheld and Sasser (1990), companies can improve profits by almost 100% by retaining just 5% more of their customers. This paragraph discusses the factors that determine whether customers of unsought goods continue the relationship with the provider of this type of good.4 The first section discusses the relationship between customer satisfaction and customer retention. After that, the influence of both the quality and length of the interpersonal relationship between the customer and vendor on customer retention is described. The paragraph ends with the relationship between the quality of additional services and customer retention.
2.3.1 Customer satisfaction
In the last decade, scholars have done a considerable amount of research on the relationship between customer satisfaction and customer retention. Customer satisfaction is defined as a customer’s overall evaluation of the performance of an offering (Johnson and Fornell 1991). Several scholars have found that customer satisfaction is positively related to customer retention. In a study from Bolton (1998), a positive effect of overall customer satisfaction on the duration of the relationship for mobile phone customers has been found. Anderson and Sullivan (1993) drew a sample of customers of major companies in a variety of industries. Among these industries were airlines, banks, clothing retail, furniture stores, insurance, and supermarkets. They found that customer satisfaction leads to repurchase intentions. Mittal and Kamakura (2001) did a study among automotive customers and found a strong effect of customer satisfaction on repurchase behavior. Additionally, Gustaffsson, Johnson and Roos (2005) conducted research among customers of a
4
22
large telecommunications company. Their results supported a consistent effect of customer satisfaction on retention. As a lot of scholars support the positive relationship between customer satisfaction and customer retention, I specify the following hypothesis:
2.3.2 Interpersonal relationship between customer and vendor
Just as the quality and length of the interpersonal relationship between the customer and vendor could positively affect the adoption of an unsought good, these two variables could also influence a customer’s decision to maintain the business relationship with the vendor of an unsought good. Below, I discuss the literature with regard to this subject.
Quality of the interpersonal relationship between customer and vendor
Hennig-Thurau and Klee (1997) were some of the first scholars who postulated that there should be a strong and positive relationship between relationship quality and customer retention. Boles, Barksdale and Johnson (1997) also studied the relationship of these two variables. The results of their study indicate that buyers who had a higher-quality relationship with their salesperson indicated significantly higher intentions to remain a customer compared to buyers with lower quality relationships. Other researchers investigated the relationship between customer retention and a single component of relationship quality. In a study among fixed telephone users, trust in the supplier was positively associated with customer retention (Ranaweera and Prabhu 2003). Gounaris (2005) conducted a study among business-to-business customers and found that trust and commitment are two important notions that cause customers to uphold a relationship with their provider. Based on these results, I define the following hypothesis:
23
Length of the interpersonal relationship between customer and vendor
Besides the quality of the interpersonal relationship between the customer and vendor, the length of this relationship could influence customer retention as well. Bolton (1998) conducted a study among both financial service and telecommunications customers and found a positive relationship between relationship duration and customer retention. Verhoef (2003) did a study among customers of a Dutch financial service company and found that relationship duration has a positive effect on customer retention. This leads to the following hypothesis:
2.3.3 Service provided
An unsought good could be accompanied with a number of additional services. The existence of additional services does not only influence the adoption of such a good; the quality of such services could also be related to customer retention. Service quality is defined as a global evaluation of a firm’s service delivery system (Anderson, Fornell and Lehmann 1994). It incorporates various dimensions that relate to both core and augmented service offerings (Bitran and Lojo 1993). Several scholars have investigated the relationship between service quality and repurchase intentions. Bitner (1990) conducted a study among airline customers and found a positive relationship between service quality and repurchase intentions. In a study among hotel visitors, Boulding et al. (1993) also found a positive relationship between these two variables. The findings of a study from Venetis and Ghauri (2004), with the advertising sector as research setting, indicate that service quality contributes to customer retention. Based on these results, it is assumed that the quality of the
H14: The quality of the interpersonal relationship between the customer and vendor of an unsought good positively affects the intention of customers to maintain the business relationship with this vendor.
24
additional services that accompanies an unsought good is positively related to customer retention. This leads to the following hypothesis:
2.4 Conceptual model
The findings of the previous paragraphs are summarized in a conceptual model (see figure 1). This model shows that there are several factors that can affect the adoption of an unsought good. The information quality (+), type of vendor, reputation and salesmanship of the vendor (+), advertising, perceived value (+), perceived risk (-), personal sources such as family or friends, and the reputation of the industry (-) can all influence the adoption of this type of good. Both the length and quality of the interpersonal relationship between the customer and vendor (+) and the services that accompany the good (+) not only affect the adoption of an unsought good, these variables are also seen as determinants of retention. Another important determinant of retention is customer satisfaction. In the next chapter, the validity of this model is tested with the use of qualitative research. The validity of this model is tested in the next two chapters.
25
Adoption
Retention
Information quality H1 (+) Type of vendor H2 Quality of relationship between customer and vendor Length of relationship between customer and vendor H3 (+) H4 (+) Reputation of vendor Salesmanship of vendor H6 (+) H5 (+) Advertising H7 Perceived risk H10 (-) Service provided H9 (+) Perceived value H8 (+) Personal sources (family/friends) H11 Industry reputation H12 (-) Customer satisfaction H14 (+) H15 (+) H16 (+) H13 (+)26
Chapter 3 Research design: qualitative research
Primary data is either qualitative or quantitative in nature. Qualitative research provides insights and understanding of the problem setting, whereas quantitative research seeks to quantify the data and applies some form of statistical analysis (Malhotra 2007: 143). In this study, both qualitative and quantitative research is conducted. Since there is few literature regarding the subject of this thesis, first qualitative research is done to test whether the conceptual model is complete and to what extent it corresponds with the reality. Subsequently, quantitative research is conducted to statistically test the variables of the conceptual model. This chapter describes the qualitative research design. In order to gather the data, a case study is conducted. The next section gives a description of the company that is used as a case study. Next, the data collection and sample are discussed. The chapter ends with a data analysis scheme.
3.1 Case study description
For this research, sanofi-aventis serves as a case study. Case studies are an intensive examination of a few selected cases of the phenomenon of interest. These cases could be consumers, stores, firms or a variety of other units such as markets (Malhotra 2007: 42).
3.1.1 Sanofi-aventis
27
disorders, internal medicines, and vaccines. Its product range includes branded prescription medicines and Over-the-Counter (OTC) drugs.
At the end of 2008, sanofi-aventis has introduced a transitioning program, focusing on three key points:
• Increasing innovation in Research & Development; • Adapting structures to meet the challenges of the future; • Exploring external growth opportunities.
By focusing on these three areas, sanofi-aventis wants to realize its goal of becoming a diversified global healthcare leader.
3.1.2 Sanofi-aventis Netherlands B.V.
Sanofi-aventis Netherlands B.V., an affiliate of sanofi-aventis, is located in Gouda and employs approximately 350 persons. In 2008, the company had a turnover of €200 million. The primary business activities of sanofi-aventis Netherlands B.V. are the registration, marketing, sales, and distribution of prescription drugs. The goal of the company is to deliver innovative, high-quality products and services. This is also visible in the organization’s mission: ‘assuring health by innovation’.
3.2 Data collection and sample description
There are several ways to obtain qualitative data. We can use either in-depth interviews or focus groups. The use of focus groups – an interview conducted by a moderator among a small group of respondents in an unstructured and natural manner (Malhotra 2007: 145) - is not appropriate for this research, as it is assumed that people do not feel comfortable talking about their health when sitting in a group. Instead, I conducted depth interviews. An in-depth interview is an unstructured, direct, personal interview in which a single respondent is probed by a skilled interviewer to uncover underlying motivations, beliefs, attitudes, and feelings on a topic (Malhotra 2007: 158).
28
disadvantages. The quality and completeness of the results depend heavily on the interviewer’s skills. Moreover, the data obtained are difficult to analyze and results cannot be generalized (Malhotra 2007: 161). I have tried to overcome these difficulties by using a semi-structured list of interview topics. This list can be found in appendix 1. In addition, I have followed a training on how to conduct interviews. Subsequently, I have practiced the interview techniques that I have learnt from this training with colleagues of sanofi-aventis.
For this research, both patients and health care professionals have been interviewed. The next sections give more insight in both populations.
3.2.1 Patients
I have conducted in-depth interviews with patients from three different clinical trials.5 In order to find patients, a number of research nurses were contacted. It was necessary to contact these research nurses, as employees of pharmaceutical companies are not allowed to contact patients themselves. The research nurses contacted a number of patients by phone. This was quite a time-consuming task, as the majority of the patients were not interested in participating in an interview. Finally, an appointment was made with the patients who were interested in participating in an interview. The interviews took place at the patient’s house or the hospital where the patient normally has to go for check-ups. The length of the interviews varied from 45 minutes to one hour.
In total ten patients have been interviewed. Although the sample size normally varies between 20 and 30 in a qualitative study (Maykut and Morehouse 1994: 63), for this study a sample size of 10 is sufficient, since this qualitative research part serves as a pilot-study. The interviewees originate from the regions Rotterdam, Utrecht, and Breda. The sample frame includes men and women who have had a pulmonary embolism, women who have been diagnosed with benign cancer in one of their breasts, and men and women with diabetes mellitus. These groups were chosen by sanofi-aventis. More information about the three trials and ailments can be found in appendix 2. The sample can be classified as a convenience sample, as the research nurses decided who they wanted to ask for the interview. This is also why the sample is not completely representative. In one group, the male/female division is
5
29
asymmetric; the interviewees who have had a pulmonary embolism are all men. This does not correspond with the reality, as this ailment occurs among both men and women.
3.2.2 Health care professionals
In addition to the patient interviews, I have interviewed four health care professionals: a cardiologist, a neurologist, and two research nurses. Health care professionals see numerous patients a day and could therefore have useful insights regarding patient recruitment and retention. All health care professionals who have been interviewed have executed clinical trials for sanofi-aventis or work for sanofi-aventis. As these persons already have a connection with the company, it is assumed that they are more willing to cooperate in an interview than ‘unknown’ health care professionals. This explains why the sample is a convenience sample, as the list of names of potential participants came from employees of sanofi-aventis.
The length of the interviews varied from 20 to 40 minutes, somewhat shorter than the patient interviews. This is due to the fact that most health care professionals have a very tight and busy schedule.
3.3 Data analysis
30
Chapter 4 Qualitative research results
In this chapter the qualitative research results are discussed. First, the factors that affect the adoption of an unsought good are examined. The second section contains the determinants of retention. In the third section, two new additional variables are discussed. The chapter ends with a summary of the qualitative research results and the proposition of a new conceptual model.
4.1 Factors affecting the adoption of an unsought good
The results of the in-depth interviews indicate that all the variables that were mentioned in the literature review affect the adoption of an unsought good. The quality of the information that is given to the patient, the influence of the vendor, perceived value, perceived risk, the service provided, the opinion of family members and/or friends, the reputation of the pharmaceutical industry, and advertisements all influence a patient’s decision of whether or not to participate in a clinical trial. I will now describe each variable in more detail.
Transfer of information
A large majority of the patients (eight out of ten) indicated that the information they received played an important role in their decision of whether or not to participate in a clinical trial. Understandable information reduces the amount of perceived risk and makes it easier to make a well considered decision. All patients were very satisfied with the oral information they received from their physician or research nurse. However, some of the interviewees were less content with the informed consent, that is, the written information they received. According to these patients, the informed consent was too long and included lots of difficult words. In addition, every potential side effect was described in detail. This scares patients. The health care professionals share this opinion. According to them, it happens quite often that a patient, who was originally enthusiastic about participating in the clinical trial, does not want to join the trial anymore after they have read the informed consent.
The vendor
31
professionals, most patients attach importance to status and knowledge. They have experienced that the majority of the patients are more willing to participate in a clinical trial when their physician asks them instead of a research nurse. This is confirmed by the patients. Although all of them were asked for the clinical trial by their physician, more than half of the patients (seven out of ten) indicated that they would not have participated in the trial if the research nurse had asked them.
Besides the type of vendor, the salesmanship of the vendor plays an important role in the adoption process as well. The majority of the patients have experienced the conversation, in which their physician told them about the clinical trial, as very pleasant. In most cases, the physician took enough time to explain the ins and outs of the trial. He or she was able to answer all questions and gave the patients enough freedom of choice. The majority of the patients (eight out of ten) indicated that both this pleasant conversation as well as the relationship with their physician positively influenced their decision to participate in the clinical trial. The health care professionals also think that the patient-physician relationship plays an important role in a patient’s decision of whether or not to participate in a clinical trial. The physicians have noticed that the longer they know a patient, the higher the chance that this patient decides to join the clinical trial. Besides the relationship with their physician, some patients also mentioned the reputation of their physician as an influencing factor on the choice of whether or not to join the clinical trial.
Advertising
32
Although these results are not completely convincing, I do add this variable to the conceptual model. In this way, it is possible to statistically test whether it is profitable for pharmaceutical companies to make more extensive use of advertisements in the future.
Perceived value
Six out of ten patients mentioned that participating in a clinical trial to help other people was not the only reason why they decided to join the trial; it also benefits them. The interviewees say that participating in a clinical trial gives them a feeling of safety, since they have to go to the hospital more often. In addition, patients mention that participating in a clinical trial gives them the opportunity to try new medicines, possibly resulting in an improved quality of life. It is worth noting that none of the patients who have been interviewed suffered from side effects, so this probably made it easier for them to ‘forget’ the disadvantages of participating in the clinical trial.
The health care professionals also believe that self-interest is one of the reasons a patient decides to join the clinical trial. According to them, patients always want to know what is in it for them.
Service provided
Participating in a clinical trial is often accompanied with little extras, such as a reimbursement for travel expenses, a designated contact person6, patient newsletters or little gifts. Six out of ten patients knew they would get (one of) these services up front. Although the offer of these extras was not the main reason of joining the clinical trial, five patients mentioned that it did make their decision of whether or not to participate in the trial easier. The health care professionals also noticed that the presence of extra services can simplify a patient’s decision of whether or not to participate in a clinical trial. However, in their opinion, pharmaceutical companies should not overdo in the offer of extra services, as the borderline between bribery and voluntarily participation is a narrow one.
6
33 Perceived risk
The health care professionals have noticed that the amount of perceived risk that is associated with participating in a clinical trial is often one of the reasons why a patient decides not to participate in such a trial. Physicians are obliged to explain the risks in detail and this often scares patients. According to the health care professionals, it is difficult to diminish the amount of risk. On the contrary, the majority of the patients who have been interviewed (seven out of ten) do not associate participating in a clinical trial with high risks. They believe that their physician would not have asked them for the clinical trial if it would not have been safe.
Personal sources (i.e., family members and/or friends)
All patients have discussed their intended participation in the clinical trial with family members and/or friends. These personal sources greatly influence a patient’s decision of whether or not to participate in such a trial. If a patient has the feeling that one of their family members and/or friends do not support their decision, chances are high that this patient declines the offer of joining the clinical trial. This indicates that physicians and/or research nurses not only have to convince the patient of the value of the clinical trial, but also his or her family members and/or friends.
The health care professionals have also noticed that personal sources can influence a patient’s choice of whether or not to join the clinical trial. According to them, it happens quite often that patients who were originally enthusiastic about participating in the clinical trial, do not want to join the trial anymore after they have had a conversation with a family member or friend.
Industry reputation
34
participating in the clinical trial. These results are, however, somewhat biased, as patients who refused to sign the informed consent have not been interviewed. That is why I do include this variable in the conceptual model, because this makes it possible to statistically test if the negative reputation of the pharmaceutical industry influences a patient’s decision of whether or not to participate in a clinical trial.
4.2 Determinants of retention
Literature revealed that customer satisfaction, the quality and length of the interpersonal relationship between a customer and vendor, and the quality of the service provided are all determinants of retention. This also appears from the in-depth interviews. Below I will describe the results in more detail.
Customer satisfaction
All ten patients are very satisfied with the execution of the clinical trial. There are no excessive waiting times for the check-ups, the instructions for taking in the medicines are very clear, and patients feel respected. None of the patients has the intention to quit the trial. According to most patients, this is due to the fact that they do not have reason to complain.
The health care professionals believe that customer satisfaction is one of the most important determinants of retention. They mentioned that they are doing everything they can to keep the patients as satisfied as possible. In their view, satisfied patients are more motivated to continue than dissatisfied patients. That is why the professionals often tell their patients that they are doing a good job, send cards when a patient has celebrated something special, and schedule the check-up appointments at times that are convenient for the patient.
Quality and length of the relationship between customer and vendor
35 Service provided
All patients mention that they receive extra services like a reimbursement for travel expenses, patient newsletters or little gifts. The majority of the patients (eight out of ten) have positive feelings about these extras. It gives them a feeling of being appreciated and motivates them to finish the entire trial. The health care professionals also indicate that these extras can motivate patients to stay in the trial program, but emphasize that pharmaceutical companies should not overdo this. They believe that patients should stay in the trial program to help themselves and other people, not because of the extras.
4.3 Additional variables
The results of the in-depth interviews indicate that all the variables that were mentioned in chapter 2 belong to the conceptual model. There are, however, two extra variables that have not been mentioned in literature. These two variables are discussed in more detail below.
Volunteer work
After the interviews were finished, I noticed a remarkable element. Eight out of ten patients who have been interviewed do or have done volunteer work. These eight patients see participating in a clinical trial as a prolongation of their regular volunteer work. This could be the case on a larger scale as well. If so, it could make the patient recruitment process easier by positioning participating in a clinical trial as a new type of volunteer work. I therefore decided to add the following hypothesis7:
Personal sources (i.e., family members and/or friends
The results of the patient interviews reveal that personal sources such as family members and friends do not only influence a patient’s decision of whether or not to participate in a clinical trial; the support of family members and friends can also be seen as a determinant of retention. The majority of the patients (eight out of ten) mentioned that they often talk with a
7
Although adding the variable ‘volunteer work’ makes sense in this research, it should be noted that this variable might be less important for other types of unsought goods, such as life insurances and encyclopedias.
36
family member or friend about the clinical trial. The patients indicate that they are happy with the support they get from these people, as it motivates them to continue. This indicates that patients who feel the support of their family members and/or friends are more likely to complete the trial program than patients who do not feel this support. I therefore decided to add the following hypothesis:
4.4 Conceptual model
Based on the qualitative research results, a new conceptual model is proposed (see figure 2). This model shows that the information quality (+), type of vendor, reputation and salesmanship of the vendor (+), advertising, perceived value (+), perceived risk (-), the reputation of the pharmaceutical industry (-) and doing volunteer work (+) all affect the adoption of an unsought good. Both the length and quality of the interpersonal relationship between the customer and vendor (+), the services that accompany the good (+) and personal sources such as family members or friends not only affect the adoption of an unsought good; these variables are also seen as determinants of retention. Another important determinant of retention is customer satisfaction. I will test the validity of these variables statistically in the next two chapters.
37 Fig. 2: Factors affecting the adoption of an unsought good and the determinants of retention
38
Chapter 5 Research design: quantitative research
In addition to qualitative research, quantitative research is performed. The goal of doing quantitative research is to statistically test the variables of the conceptual model. This chapter gives more information on the quantitative research design. First, the data collection, sample, and questionnaire are discussed. After that, a data analysis scheme is given.
5.1 Data collection and sample description
Quantitative data can be obtained with the use of a questionnaire. A questionnaire is a formalized set of questions for obtaining information from respondents (Malhotra 2007: 299). For this research, a questionnaire was distributed among Dutch patients who currently participate or have participated in a clinical trial (from now on referred to as participants) and Dutch patients who refused to participate in a clinical trial (non-participants). More information on the questionnaire can be found in section 5.2. The sample frame includes, just as with the qualitative research part, the following groups:
1) Men and women who have had a pulmonary embolism;
2) Women who have been diagnosed with benign cancer in one of their breasts; 3) Men and women with type 2 diabetes mellitus.
39
All questionnaires were sent out via mail by an employee of the hospital. Each respondent received an envelope including a cover letter, questionnaire, return envelope, and ball point. The questionnaires were distributed in the period September – October 2009. In total 300 questionnaires were sent out; 212 to participants and 88 to non-participants. These numbers differ from each other, as it was more difficult to retrieve information on the non-participants. Eventually, 192 people returned the questionnaire. 5 surveys were not filled in properly and have been deleted for further research. In total 133 participants and 54 non-participants returned a complete questionnaire. This makes that the response rate among participants is 62.7 % and 61.3% among non-participants. The high response rate among non-participants is surprising. This group does not have a connection with the employees of the hospital, so it was therefore assumed that most of them would not be interested in filling in the questionnaire.
5.2 Questionnaire
In this section, more information on the development of the questionnaire is given. According to Malhotra (2007: 299), a questionnaire has three objectives:
1) A questionnaire translates the information needed into a set of specific questions that the respondents can and will answer;
2) A questionnaire must uplift, motivate, and encourage the respondent to become involved in the interview, to cooperate, and to complete the interview;
3) A questionnaire should minimize response error.
During the development of the questionnaire, all three objectives were kept in mind. This is why the questionnaire starts with an introduction. The purpose of the introduction is to try to get the respondent involved with the research subject. It involves a description of the research purpose and an estimation of how much time it will cost to fill in the questionnaire. In order to increase response rates, respondents received a ball point to fill in the questionnaire.
