A Marketing Perspective on Public Policy Options in an Epidemic
The ability to quickly vaccinate a population is an important capability in a world proliferating biological weapons and rapidly mutating viruses. In the 1918 world-wide influenza epidemic, 20 million people died after a formerly benign virus mutated. In recent years, there have been increasing worries about biologic agents, such as the smallpox virus, being used to attack a population. A key to managing all of these threats is the ability to rapidly vaccinate the threatened population. While considerable research effort has been devoted to understanding the most effective biological makeup of these vaccines (cf. Gay and Richie 2007) and optimizing the distribution of vaccines (Gostin et al. 2002), little research has been devoted to the issue of how business or government marketing strategies may convince a population that they should obtain a vaccine before it is too late.
Thus, this research addresses the issue of how vaccination rates may be increased through the use of public policy and marketing efforts. We approach the vaccination dilemma in terms of how patients process information through “cues to action,” strategies that are designed to activate a person’s readiness to engage in a particular health behavior (Janz, Champion, and Stretcher 2002). We illustrate the effect of five different cues to action (information only, mandate on a related group, direct mandate, an ad, and a health survey) on perceived threat and intention to seek a vaccination. Our context is the human papilloma virus (HPV) and the possibility of a female acquiring cervical cancer if she is not vaccinated against HPV.
Study 1 examined differences in two cues to action within an experimental design. One group of participants (all were university females) was presented a scenario which provided general information about the HPV vaccination (“information” cue). The second group was shown a scenario describing a hypothetical mandate requiring 6th grade girls in the state to receive the HPV vaccination (“mandate on related group” cue). After reading a scenario, each resondents answered questions about her health belief attitudes, perceived seriousness of the virus, her perceived susceptibility, and willingness and likelihood of obtaining the vaccination.
Across our sample, perceived threat of cervical cancer positively and significantly influenced intention to obtain the HPV vaccination (β = 0.425, F = 10.39, p .01). We used a multivariate data analytic approach to determine whether or not threat and intention differed across our two cues to action. An insignificant Wilks’ lambda = .989 (F(2,73) = 0.42, p = .657) for the main effect of condition suggests that there were no differences in the two cues’ impact on threat and intention. No significant differences in prior knowledge of the vaccination across conditions (F = 1.137, p = .290) were found.
Given that our analysis suggested that both the information and indirect mandate cues had the same effect, we expanded the type of cues we used in Study 2. First, a “direct mandate” cue was used; the associated hypothetical scenario stated that a law would require the college-age female to obtain the HPV vaccination. (This cue is in contrast to the earlier mandate cue, which required the vaccination for younger girls.) Second, in the “advertisement cue” condition respondents were shown a pharmaceutical ad for the HPV vaccination. In the third condition, participants were only asked to complete a general health-related survey (the “survey” cue). Each read a scenario and then was asked the same set of questions as in Study 1.
We used a GLM multivariate procedure to analyze our data. The MANCOVA resulted in a significant Wilks’ Lambda = .87 (F(6, 234) = 2.91, p .01) for the main effect of condition on both threat and intention when skepticism and prior knowledge were both controlled. Our examination of the main effect of the cues overall on each dependent variable showed that cues to action significantly influence perceived threat (F = 3.44, p .05) and intention (F = 3.88, p .01). We explored the effect for each condition separately to follow up on this main effect. The results illustrated that there are significant differences in threat and intention in each pair of contrasts, (each cue to action versus survey-only group). Specifically, both the mandate on a related group (Mthreat = 4.94, t = -2.12, p .05; Mintention = -.04stand, t = -2.06, p .05) and direct mandate conditions (Mthreat = 4.84, t = -2.74, p .01; Mintention = -.08stand, t = -2.22, p .05) scored lower on perceived threat and intention to obtain the vaccination than did respondents in the survey-only condition (Mthreat = 5.29; Mintention = .33stand). This finding suggests that reactance theory may be at work with the government mandate cues to action, actually reducing both the threat the respondent feels and her intention to obtain the vaccination. Interestingly, the advertising cue (Mthreat = 4.89, t = -2.68, p .01; Mintention = -.19stand, t = -3.30, p .01) also produced less threat and intention than the survey-only condition. The survey-only condition may be acting as a cue generating internal concerns, whereas the advertisement may be a weaker cue comparatively, given its external nature.
In this research, we find that the cues to action influence the perception of threat of a disease and intention to obtain the vaccination. Given that cues prompt across-the-board action, the question for policy makers and companies alike is which type of cue produces a greater positive effect, given potential risks and costs. Our research suggests that while the mandate-related cues in Study 2 influenced women’s perceptions of the threat of HPV and intention to obtain the vaccination, the mandate conditions did not gain more potential compliance than taking a general health survey. We suggest that the survey-only condition generates internal cues which may be far stronger than external cues. Future research examining the effectiveness of various combinations of marketing activities (e.g., indirect mandate plus traditional advertising versus direct mandate plus positive word of mouth) may uncover a hierarchy of effectiveness in marketing activities combinations.
A Marketing Perspective on Public Policy Options in an Epidemic
"The triumph of persuasion over force is the sign of a civilized society."
(Alfred North Whitehead, 1933)
Introduction
The ability to quickly vaccinate a population is an important capability in a world proliferating biological weapons and rapidly mutating viruses. In the 1918 world-wide influenza epidemic, 20 million people died after a formerly benign virus mutated. In recent years, there have been increasing worries about biologic agents, such as the smallpox virus, being used to attack a population. A key to managing all of these threats is the ability to rapidly vaccinate the threatened population. While considerable research effort has been devoted to understanding the most effective biological makeup of these vaccines (cf. Gay and Richie 2007) and optimizing the distribution of vaccines (Gostin et al. 2002), little research has been devoted to the issue of how business or government marketing strategies may convince a population that they should obtain a vaccine before it is too late.
Thus, this research looks as the effectiveness of government mandates and other mass marketing strategies as triggers of seeking preventive health behavior in an epidemic. Specifically, we address the issue of how vaccination rates may be increased through the use of public policy and marketing efforts. We approach the vaccination dilemma in terms of how patients process information through “cues to action,” strategies that are designed to activate a person’s readiness to engage in a particular health behavior (Janz, Champion, and Stretcher 2002). We illustrate the effect of four different cues to action on perceived threat and intention to seek a vaccination, giving particular attention to the effects of government mandates on health beliefs and behaviors as compared to other options from the marketing domain.
Conceptualization
Perceived susceptibility to disease, defined as an individual’s subjective perception of the risk of contracting a health condition (Janz, Champion, and Stretcher 2002) or one’s perceived vulnerability to experiencing negative consequences from an acute health threat (cf. Gerund et al. 2004), has been highlighted as a core component driving whether or not a person seeks preventive health care[1] (Kasl and Cobb 1966). Several cross-sectional studies (e.g., Ford and Norris 1995; Liau and Zimet 2000) have found significant associations between perceived susceptibility and behavior.
Proponents of stage theories of health-relevant behavior, such as the Precaution Adoption Process Model (PAPM), believe that cost-benefit theories of health behavior, such as the Health Belief Model (Rosenstock 1974), the Theory of Reasoned Action (Ajzen and Fishbein 1980), and the Theory of Planned Behavior (Ajzen 1985), are dealing mainly with the factors that govern how people who get to the stage of deciding to act, actually decide what to do (Weinstein and Sandman 2002). These authors conceptualize perceived susceptibility as one factor that frequently influences what people decide. Because people are disinclined to acknowledge personal susceptibility to harm even when they recognize the risk faced by others (Weinstein 1987), it seems that overcoming this reluctance is a major barrier to getting people to decide to act.
Perceived susceptibility has been found to be influenced by psychological factors, such as perceived similarity to a person with a disease and perceived prevalence of the disease (Gerend et al. 2004). These effects were shown to exist beyond respondents’ medical risk factors (Gerend et al. 2004) and were conceptualized as a cognitive heuristics factor influencing susceptibility; however, the operationalization of the constructconsolidated measures of media (the amount of media communications exposed to), professional guidance (information from a doctor or health care professional) and social influence (number of friends with the disease studied). While positively related, this combined measure did not have significant effects on susceptibility. Perhaps part of the reason no significant effects were found is that the construct mixed distinct elements of influence, or cues to action.
Perceived severity is defined as a person’s beliefs about the seriousness of a condition and its secondary consequences and results (Janz, Champion, and Stretcher 2002). Perceived severity reflects feelings about the seriousness of contracting an illness or of leaving it untreated, and may include evaluations of both medical and clinical consequences (for example, death, disability, and pain) and possible social consequences (such as effects of the conditions on work, family life, and social relations). The combination of susceptibility and severity has been labeled perceived threat.
Perceived threat is a sequential function of perceived severity and perceived susceptibility (Janz, Champion, and Stretcher 2002). A heightened state of severity is required before perceived susceptibility becomes a powerful predictor. Perceived susceptibility, under the condition of high perceived severity, is likely to be a stronger predictor of intention to engage in health-related behaviors than it will be a predictor of actual engagement in health-related behaviors (Janz, Champion, and Stretcher (2002). Further, cues to action may have a greater influence on behavior in situations in which perceived threat is large. Little is known about cues to action and their relative effects; clearly, more research is needed (Janz, Champion, and Stretcher 2002).
A cue to action is a stimulus or trigger that prompts an individual to initiate some behavior (Burns 1992; Rosenstock 1966). For a cue to action to stimulate a behavior, it is presumed that there is some level of readiness to act. There is an inverse relationship between readiness and the intensity of the cue to action as it relates to taking some health behavior. That is, the lower the perceived threat, the greater the intensity of the cue to action that is needed in order for the individual to take some health behavior. Cues to action are contingent on some pre-existing knowledge of the disease, which has impacted the attitudes of the individual.
Cues to action exist in many forms, from a single, fleeting bodily event (e.g., a sneeze) to large scale environmental events (e.g., a multi-channel marketing campaign) (Janz et al. 2002). Of course, cues to action that are external to the individual and which serve as channels of influence are of primary interest here. These may be classified according to dimensions rooted in the marketing literature: level of personalization, i.e., degree of differentiation across target segments; and source of information, whether marketer-initiated versus non-marketer initiated. Personal cues are highly individualized, often targeted to a segment of one, and include physician reminder postcards and phone calls, a visits from a counselor or social worker, and talks with a friend or family member. Non-personal cues are broadly targeted and may emanate from broadcast or print advertisements and public service announcements or take the form of government advisories or legislation (mandates). Both marketer-initiated and non marketer-initiated action cues span the range of personalization from narrowly targeted direct marketing pieces such as postcards, to mass market broadcast appeals.
Three classes of strategic tools for influencing public health and social issue behaviors are often mentioned in the literature: education, marketing, and law (Rothchild 1999). Although marketing and law are considering unique forms of influencing strategies, law may be sought actively by a firm as part of a comprehensive marketing strategy, as in the case of a pharmaceutical company that lobbies for enactment of legislation mandating the use of its vaccines in a particular segment of the population.
In this study, we examine the ability of various broadly targeted, marketer-initiated cues to modify the preexisting knowledge, attitudes, and behavioral intentions of a relevant target segment. This is a relatively understudied area in both the medical and marketing literatures. In the marketing literature, research incorporating cues to action have focused primarily on advertisements and reminders. For example, in their study of pharmaceutical product-use compliance, Bowman, Heilman and Seetharaman (2004) include examination of an advertising cue to action and find mixed results. The panel data used captured the advertising cue as advertising expenditure and whether the consumer recalled having seen the product advertised. Thus it was not a test of a signal advertisement or type of advertisement, but a measure of overall ad impact and recall. The results illustrated that the ad cue in some cases had an effect on product-use compliance, but the effect varied in being a positive or negative influence depending on the product category and consumer segment. No clear conclusions were reached in relation to the cue.