Seeking and Processing Influenza Vaccine Information: A Study of Health Care Workers at a Large Urban Hospital

Abstract

In response to low influenza vaccination rates among health care workers, health officials have explored determinants of uptake and developed communication interventions. Key to these efforts is how workers seek and attend to vaccine information. We applied a model of risk information seeking and processing to survey 226 physicians and nurses at a large, urban hospital (response rate = 5.7%; cooperation rate = 91.5%). Our goals were to (1) apply the model to a new setting, audience, and issue and (2) explore novel concepts (e.g., source credibility and information usefulness) and moderating relationships among variables. Consistent with previous research, information subjective norms—perceived social pressure to remain informed about vaccination—were a strong predictor of information seeking and systematic processing. Neither source credibility nor information usefulness independently increased explained variance, although they moderated the relationship between norms and communication behavior. We discuss theoretical and practical implications of these findings.

Notes

¹“Health care worker” is a broad term. CDC defines HCWs as “all medical and non-medical personnel in contact with patients,” ranging from physicians and nurses to emergency response workers, employees of nursing homes and assisted living facilities, medical students, and home health care providers (Hofmann et al., 2006, p. 143). However, influenza vaccination rates are often not stratified at the level of each individual group. CDC (2006), for example, has a category for “health care personnel” (self-reported occupation; 42% uptake) and household contacts for children 5–17 years old (10.8% uptake) and children 25–59 months old (35%). Moreover, studies that explore rates at particular facilities use different categories. Some are more general (for example, employment in a hospital or nonhospital setting), while others are more specific according to occupation (i.e., health-diagnosing professions such as physicians and health-assessment professions such as nurses) (Walker et al., 2006).

²Not all RISP model components were included in the present study. Omitted variables included perceived hazard characteristics (i.e., risk susceptibility and severity) and affective response (i.e., worry). In the RISP model, affective response mediates the relationship between perceived hazard characteristics and information sufficiency (Griffin et al., 1999). Subsequent research has also found a direct relationship between affective response and information seeking and processing (Griffin et al., 2008). Our study, however, did not focus specifically on information about vaccination (or influenza) risk, but rather on vaccination more broadly.

³ Trumbo and McComas (2003) found that perceiving industry and government health officials as highly credible (and citizen groups as comparatively less credible) was positively associated with heuristic processing; conversely, perceiving citizen advocacy groups as more credible (and industry and government officials as less credible) was positively associated with systematic processing. Trumbo and McComas suggested that the explanation may lie in the information each group tended to present. Most cancer clusters investigations yield little or no evidence of statistically higher cancer rates; even for those that do, few identify a clear cause. Heuristic processing of information from government health officials and industry representatives may reflect a general faith in science and in the scientific approach to studying potential clusters. However, citizen groups may push for an interpretation that differs from the official conclusion—mainly, that a cancer cluster does exist. Thus, perceiving these groups as credible may motivate people to think more critically about the issue and “arrive at a more critical conclusion that was not strictly based on scientific rationality or faith in science [alone]” (Trumbo & McComas, 2008, p. 72).