Elsevier

Vaccine

Volume 21, Issues 13–14, 28 March 2003, Pages 1486-1491
Vaccine

Sensitivity and specificity of patient self-report of influenza and pneumococcal polysaccharide vaccinations among elderly outpatients in diverse patient care strata

https://doi.org/10.1016/S0264-410X(02)00700-4Get rights and content

Abstract

National surveys of adult vaccination indicate moderate self-reported immunization rates in the US, with limited validity data. We compared self-report with medical record abstraction for 820 persons aged ≥66 years from inner-city health centers, Veterans Affairs (VA) outpatient clinics, rural and suburban practices. For influenza vaccine, sensitivity was 98% (95% CI: 96–99%); specificity was 38% (95% CI: 33–43%). For pneumococcal polysaccharide vaccine, sensitivity was 85% (95% CI: 82–89%) and specificity was 46% (95% CI: 42–50%). The VA had the highest sensitivity and lowest specificity for both vaccines while the converse was true in inner-city centers. High negative predictive values indicate that clinicians can confidently vaccinate based on negative patient self-report.

Introduction

Influenza and pneumonia are the seventh leading cause of death in the US, with influenza causing about 20,000 deaths annually. This figure climbs to 40,000 or more excess deaths in selected years with large epidemics [1]. Furthermore, each year there are an average of 114,000 excess influenza-related hospitalizations [1]. The elderly (>65 years) and those with chronic medical conditions are most at risk. Streptococcus pneumoniae causes an estimated 3000 cases of meningitis, 50,000 cases of bacteremia, and 500,000 cases of pneumonia annually [2], [3], and accounts for a major proportion of invasive bacterial disease in all age groups. The proportion of penicillin-resistant pneumococci has increased to 35% in some areas [2]. Even with appropriate treatment, the case-fatality rate for pneumococcal bacteremia in younger adults is 15–20%, climbing to 30–40% in the elderly [2].

It is estimated that influenza vaccine and pneumococcal polysaccharide vaccine (pneumococcal vaccine) prevent thousands of deaths each year, yet among adults >65 years in the first quarter of 2002, vaccination rates in the US were 66% for influenza and 55% for pneumococcus [4]. Influenza vaccination rates among minority populations were lower still at 48% for Hispanics and 50% for African–Americans; pneumococcal vaccination rates were 26% for Hispanics and 32% for African–Americans, explaining why immunizations are one of the six areas nationally targeted to reduce racial disparity [4].

Of course, all of these data are based on patient self-report. Large national surveys such as the Medicare Beneficiary Survey and the Behavioral Risk Factor Surveillance Survey depend upon self-report, which, especially among the elderly, is subject to recall bias (the aging have increasingly higher rates of memory loss). Several studies show good sensitivity and specificity for self-report of influenza vaccination among Veterans Affairs (VA) patients and a community managed care organization, yet we are unaware of data in rural or inner-city areas (other than VA institutions) where immunization rates may be the lowest [5], [6], [7]. The validity of self-reported vaccination status among older adults and among disadvantaged persons has been inadequately evaluated for pneumococcal vaccine; we are aware of only one published study [5].

The goal of this study was to determine the validity of patient self-report of receipt of influenza vaccine and pneumococcal vaccine in a variety of patient care settings, including settings of poverty. We selected four patient care strata to ensure access to a broad spectrum of patients and vaccination policies: (a) rural medical practices in a network in western to central Pennsylvania; (b) urban/suburban (suburban) medical practices in the same network; (c) outpatient clinics in VA health centers; and (d) inner-city neighborhood health centers in Pittsburgh, PA. A network of non-academic practices affiliated with the University of Pittsburgh Medical Center was chosen for access to suburban patients. Rural sites were included as a separate group, as it was thought that those patients might have less access to care and be more impoverished than patients in suburban practices. The VA was included because it has implemented a multi-component program to raise influenza vaccination rates. Inner-city neighborhood health centers were included due to the issues of disparity in immunization rates associated with poverty and race.

Section snippets

Materials and methods

A detailed description of the overall methods used in this study has been previously published [8].

Response rate

We sent requests for participation to 1642 persons and found 259 were ineligible or had invalid phone numbers, leaving 1383 potential respondents. When we concluded interviewing, we had 1007 completed interviews, 227 refusals and 149 persons whom we had tried repeatedly to reach by phone but were unable to contact. This equaled an interview response rate of 73% [11].

Most (919/1007=91%) of the respondents agreed to the medical record review. Reviews were conducted on 820 records with 99 records

Discussion

Self-report of immunization status is used in national surveys such as the Medicare Beneficiary Survey and the Behavioral Risk Factor Surveillance Survey. This raises a question as to the quality of self-report of vaccination. We found that sensitivity, or the ability to accurately recall having received vaccine when compared to the records of the primary care physician, was high for both influenza vaccine and pneumococcal vaccine, although specificity, the ability to accurately recall not

Conclusions

Sensitivity of self-report of influenza vaccine and of pneumococcal vaccine are high but the specificity is lower for each. Based on the relatively high negative predictive value of self-report and the excellent safety record of adult vaccines, we believe that physicians can confidently recommend vaccination to adult patients who report not having received influenza and pneumococcal vaccines.

Acknowledgements

This project was approved by the Institutional Review Board of the University of Pittsburgh, and the Human Subjects Use Subcommittee of the Institutional Review Board of the VA Healthcare System of Pittsburgh. This publication/project was funded by HS09874 from the Agency for Healthcare Research and Quality.

References (16)

There are more references available in the full text version of this article.

Cited by (230)

  • Agreement among sources of adult influenza vaccination in the age of immunization information systems

    2021, Vaccine
    Citation Excerpt :

    Thus, understanding the relationships among these sources of influenza vaccination information is essential. Previous studies of influenza vaccine self-report among older adults compared with medical charts and electronic databases as the “gold standards” have reported sensitivity of 98%–100% indicating that self-report of receiving influenza vaccine was considered to be highly reliable; whereas, self-report of not receiving influenza vaccine was not as reliable, with specificity of 38%–71% [4,5]. This level of agreement was confirmed in more recent studies comparing the EMR and self-report [6,7].

View all citing articles on Scopus
View full text