ACP American College of Physicians - Internal Medicine - Doctors for Adults

Effective Clinical Practice

SPECIAL TOPIC

Using the National Health Interview Survey: Time Trends in Influenza Vaccinations among Targeted Adults

Effective Clinical Practice, May/June 2002

John R. Pleis, MS, Jane F. Gentleman, PhD, Division of Health Interview Statistics, National Center for Health Statistics, Centers for Disease Control and Prevention, Hyattsville, Md

For author affiliations, current addresses, and contributions, see end of text.

Context. Influenza and its complications result in significant morbidity and mortality each year. Certain groups are at increased risk for influenza and influenza-related complications. They, and others who are in close contact with them, are target groups to receive a yearly influenza immunization according to recommendations from the Advisory Committee on Immunization Practices (ACIP).

Objective. To estimate the proportions of adults in selected target groups who received influenza vaccination in 1995 and 1998 and to identify characteristics associated with vaccination receipt.

Data Source. The National Health Interview Survey (NHIS), a nationally representative survey of civilian noninstitutionalized persons conducted annually by the Centers for Disease Control's National Center for Health Statistics. We used data for adults (> 18 years of age) from the 1995 and 1998 NHIS.

Outcome Measure. Proportions of persons in target groups self-reporting influenza vaccination in the 12 months before the NHIS interview.

Results. Between 1995 and 1998, influenza vaccination increased for persons aged 65 and older (58.2% to 63.3%; P<0.05) and for adults under 65 belonging to selected ACIP target groups (27.5% to 30.1%; P<0.05). Examination of 1998 data shows that regardless of age, the likelihood of influenza vaccination is strongly influenced by having health coverage or a regular source of care. For example, 66% of the elderly with private fee- for-service health care coverage were vaccinated, compared with 23% of the elderly with no insurance (adjusted odds ratio [OR], 3.9; 95% CI, 1.6 to 9.3). For persons aged 18 to 64 years belonging to an ACIP target group, the corresponding figures are 32% vs. 16% (adjusted OR, 1.8; CI, 1.4 to 2.3). The likelihood of vaccination also varied by race and ethnicity: For age 65 and older, 66% of non-Hispanic whites were vaccinated compared with 46% of non-Hispanic blacks (adjusted OR, 2; CI, 1.6 to 2.4).

Conclusions. The use of influenza vaccination among adults at high risk for influenza and influenza-related complications increased between 1995 and 1998. Younger individuals at high risk, people without insurance or a regular source of care, and nonwhites still have low vaccination rates.


Influenza, a serious upper respiratory infection, affects between 35 to 50 million people in the United States each year (1,2) and accounts for significant morbidity and mortality.(3) In 1998, influenza-related hospitalizations at community hospitals totaled 110,000.(4) Almost 2,600,000 influenza-related visits to health care providers were recorded during 1998 (National Ambulatory Medical Care Survey. Personal communication). Further, influenza-related hospitalization and intensive treatment are more common among certain high-risk groups.(5) In its 1997 and 1998 annual publications, the Advisory Committee on Immunization Practices (ACIP), the members of which are appointed by the Secretary of Health and Human Services, recommended influenza immunization for various target groups:(6-8) people aged 65 and older, people with various chronic diseases, close contacts of other target group members, health care workers, certain pregnant women, and anyone who wishes to reduce the chance of becoming infected with influenza (Table 1).

A yearly influenza vaccination is one of the most effective methods for preventing influenza and its complications.(9) Influenza vaccinations have been shown to reduce the amount of hospitalizations, physician visits, and mortality related to influenza.(10,11) Immunization rates for groups at increased risk for influenza and influenza-related complications are components of the Healthy People 2010 objectives, which include yearly influenza vaccinations for 90% of all noninstitutionalized persons 65 years of age or older and 60% of all noninstitutionalized persons 18 to 64 years of age who have a chronic condition that places them at high risk for influenza complications.(12,13)

Data from the National Health Interview Survey (NHIS), which is conducted yearly by the Centers for Disease Control and Prevention's National Center for Health Statistics (NCHS), have been used to assess progress toward achieving these objectives.(12-14) In this study, we use the public-use data files for the NHIS to examine how use of influenza vaccinations changed over time (by comparing data from 1995 to those from 1998) among adults in selected ACIP-defined target groups. We also examined the association of health care coverage, a regular source of health care, and a variety of sociodemographic factors with receipt of influenza vaccination by using 1998 NHIS data.

Methods

We compared the use of influenza vaccinations among target group adults from 1995 and 1998 using data from the NHIS. The details of the NHIS are described elsewhere.(15)

Outcome Measure

The main outcome measure—influenza vaccination status—was assessed in the NHIS interview with the following question: "During the PAST 12 MONTHS, have you had a flu vaccination? A flu vaccination is usually given in the fall and protects against influenza for the influenza season" (yes/no/refused/don't know).

For the 1995 NHIS, questions about the influenza vaccine appeared in the Year 2000 Objectives Supplement, completed by 17,317 adults and corresponding to a weighted population of about 191,000,000 adults.(16,17) The response rate to the Supplement was 81%. In 1998, the influenza questions were part of the core questionnaire completed by 32,440 "sample adults," one randomly selected from each family and corresponding to a weighted population of about 197,300,000 adults. Response rate for the household questionnaire was 90% and the response rate for the Sample Adult questionnaire was 74%.

Target Groups for Vaccination

The ACIP's definitions of target groups for recommended influenza vaccinations remained relatively unchanged from 1994-1995 to 1997-1998, except for pregnant women (6,7,18,19) (Table 1). We identified NHIS adults with selected ACIP-defined target conditions: diabetes; asthma, chronic bronchitis, emphysema, or other lung disorders(20); heart disease or any other heart condition (including angina or a heart attack) (21,22); liver disease; and weak or failing kidneys. NHIS respondents were not asked directly if they were health care workers. We identified these workers by using data from NHIS questions on occupation that had been coded into Standard Occupational Classification (SOC) categories.(23)

Covariates

Several characteristics of target group adults were considered in our analyses. These included health care coverage (categories for adults < 65 years of age differed from those for adults > 65 years of age), regular source of health care, race/ethnicity, sex, education, marital status, poverty ratio,(24) region of residence,(25) smoking status, and health status. The exact specifications used for these variables are presented in Table 2.

Statistical Analysis

We estimated the percentages of adults in the ACIP target groups who received an influenza vaccination (in the past 12 months) in 1995 and 1998. Results were stratified by age because of differences in health care coverage. Multiple logistic regressions (stratified by age) were performed to calculate the odds that target group adults received an influenza vaccination, after adjusting for the foregoing covariates. All estimates were weighted, and SUDAAN(26) statistical software (Research Triangle Park, NC) was used to calculate variance estimates that accounted for the complex survey design. As a result of the high prevalence of influenza vaccinations in the general population, the odds ratios presented should not be interpreted as risk ratios.(27) We did not convert the odds ratios to risk ratios because, while the odds ratio does not vary with changes in the values of the other covariates, the risk ratio is a function of these values.

As a measure of reliability, NCHS uses the relative standard error, which is the ratio of the standard error of the estimate to the estimate itself. NCHS considers estimates unreliable if the relative standard error is greater than 30%; such estimates are identified in the regression results below and should be interpreted with caution.(28)

Results

Vaccination: 1995 vs. 1998

Table 3 shows that influenza vaccination rates increased from 1995 to 1998 in older adults from 58.2% to 63.3% (P<0.05). Influenza vaccination rates increased for adults under age 65 in all the ACIP target groups we studied (27.5% to 30.1%; P<0.05); statistically significant increases were noted among health care workers (29.3% to 36.7%), diabetics (33.9% to 40.7%), and adults with kidney disease (26.1% to 37.4%). It should be noted, however, that the estimates for the last two groups did not meet the NCHS reliability standard, primarily because of smaller sample sizes in 1995, and should therefore be interpreted with caution. Vaccination rates also increased for nontargeted adults under age 65, from 13.1% to 15.6% (P<0.05).

Vaccination and Sociodemographic Characteristics

Table 4 shows percentages and crude and adjusted odds ratios of vaccination (using 1998 data) for persons aged 65 years and older in ACIP target groups according to selected characteristics. Target group adults who had health care coverage or a regular source of care were substantially more likely to have received a flu vaccination than those who did not. For example, 66% of the elderly with private fee-for-service insurance were vaccinated, compared with 23% of the elderly who had no health care coverage (adjusted OR, 3.9; CI, 1.6 to 9.3). Differences in the likelihood of vaccination were also observed with regard to a variety of sociodemographic characteristics: Vaccination was somewhat more likely among highly educated, economically advantaged, and non-Hispanic black and Hispanic persons as well as non- and former smokers. At age 65 and older, 66% of non-Hispanic whites were vaccinated, compared with 46% of non-Hispanic blacks (adjusted OR, 2; CI, 1.6 to 2.4).

The overall pattern of findings was similar for people in target groups, although in persons aged 18 to 64 years vaccination rates were lower by about one half (Table 5).

Finally, additional odds ratios were calculated to determine if the type of health care coverage (managed care, fee-for-service) was associated with receiving a flu vaccination for target group adults. For older adults with fee-for-service coverage, differences were seen between those with private health care coverage (in addition to Medicare) and Medicare-only health care coverage (OR, 1.4; CI, 1.1 to 1.8). No other differences were seen.

Discussion

The need for influenza vaccinations in certain groups is heavily publicized in the United States. There have been some notable successes in fostering compliance to 1997 and 1998 influenza vaccination recommendations made by ACIP.

Between 1995 and 1998, even persons who were not in any of the specific ACIP target groups identifiable by NHIS data increased influenza vaccination rates, which were, as expected, appreciably lower than those for specific target groups. By far, persons 65 years of age and older had the highest influenza vaccination rates, and those rates increased from 1995 to 1998.

In the analysis of target group members, having some type of health care coverage and a regular source of health care strongly and significantly increased the likelihood of having complied to ACIP recommendations, even more so for persons 65 years of age and older than for younger adults.

For persons with health care coverage, the type of coverage can make a difference. Among older adults with fee-for-service health care coverage, the odds of receiving a flu vaccination in the past 12 months were nearly 40% higher for those adults with private coverage plus Medicare, compared with older adults who had Medicare alone (even though influenza immunizations are covered by Medicare).

Influenza vaccinations are relatively inexpensive and usually needed only once per year. In many places, it is not necessary to obtain them at a doctor's office—they are offered at easily accessible and frequently visited locations, such as in the workplace and at grocery stores. Nevertheless, immunization rates among target group members are well under 100%, and there are certain target group members toward whom future efforts to increase rates could be directed. Rates were only in the 20%-to-30% range for adults aged 18 to 64 years belonging to ACIP target groups. Regardless of age, adults with no health care coverage, adults without a regular source of health care, non-Hispanic blacks, and Hispanic persons were substantially less likely to receive influenza vaccination than other adults.

This study has certain limitations. NHIS data are not available to completely identify the populations targeted by ACIP recommendations (Table 1). The NHIS does not cover military personnel or institutionalized adults, so their compliance to influenza vaccination recommendations could not be assessed. In particular, compliance of nursing home residents to influenza vaccination recommendations could not be determined. Also, while the NHIS asks women if they are pregnant, how long they have been so is not determined. In addition, this study covered only a few chronic diseases, and although the NHIS gathers information on frequency of health care system use (including hospitalization), NHIS data cannot be used to attribute use of the health care system to persons with a particular chronic condition. In addition, existing NHIS data cannot be used to determine if all close contacts of persons at increased risk for influenza and influenza-related complications received an influenza immunization.

Another limitation is that the status of health care coverage and other characteristics may have been different when recorded at the time of the interview than when the influenza vaccination was received (up to a year before the interview). The data in the NHIS are self-reported (either by proxy or by the sample adults), and items such as medical conditions are not verified with other sources. Finally, it is important to note that in a cross-sectional survey such as the NHIS, conclusions can be made about associations between individual characteristics, but cause-and-effect relationships can only be suggested—not proven—by the data analysis results.

This study shows that adults who receive influenza vaccinations differ from those who do not receive them. Identifying subgroups who are at increased risk for influenza and influenza-related complications, but who do not comply with ACIP recommendations to be immunized against influenza, can help public health policymakers evaluate and target programs aimed at decreasing the incidence of influenza and influenza-related complications. These and other subgroups may need to be targeted for disease-prevention education and/or designated for subgroup-specific programs and funding with the objectives of increasing compliance with influenza vaccination recommendations and reducing health disparities. Future surveys could evaluate compliance with influenza-vaccination recommendations more accurately if survey questions were developed explicitly for that purpose.

References

1. Influenza fact sheet. Bethesda, MD: Office of Communications and Public Liaison, National Institute of Allergies and Infectious Disease, National Institutes of Health. Available at: http://www.niaid.nih.gov/factsheets/influenza.htm. Accessed April 18, 2001.

2. Influenza: general information. Atlanta, GA: National Center for Infectious Diseases, Centers for Disease Control and Prevention. Available at: http://www.cdc.gov/ncidod/diseases/influenza/influenzainfo.htm. Accessed on April 18, 2001.

3. Brammer TL, Isurieta HS, Fukuda K, et al. Surveillance for influenza—United States, 1994-1995, 1995-1996, and 1996-1997 seasons. MMWR CDC Surveil Summ. 2000;49:13-28.

4. Popovic JR, Kozak LJ. National Hospital Discharge Survey: annual summary, 1998. National Center for Health Statistics. Vital Health Stat 13:148;2000.

5. Szucs T. The socio-economic burden of influenza. J Antimicrob Chemother. 1999;44:S11-5.

6. Prevention and control of influenza: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Morb Mortal Wkly Rep. 1997;46:1-25.

7. Prevention and control of influenza: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Morb Mortal Wkly Rep. 1998;47:1-26.

8. Prevention and control of influenza: Recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Morb Mortal Wkly Rep. 2000;49:1-38.

9. Subbarao K. Influenza vaccines: present and future. Adv Virus Res. 1999;54:349-73.

10. Nichol KL. Complications of influenza and benefits of vaccination. Vaccine. 1999;17:S47-52.

11. Nichol KL, Goodman M. The health and economic benefits of influenza vaccination for healthy and at-risk persons aged 65 to 74 years. Pharmacoeconomics. 1999;16:S63-71.

12. US Department of Health and Human Services. Healthy People 2010. 2nd ed. With Understanding and Improving Health and Objectives for Improving Health. Washington, DC: US Government Printing Office; 2000.

13. US Department of Health and Human Services. Tracking Healthy People 2010. Washington, DC: US Government Printing Office; 2000.

14. Singleton JA, Greby SM, Wooten KG, Walker FJ, Strikas R. Influenza, pneumococcal, and tetanus toxoid vaccination of adults—United States, 1993-1997. MMWR CDC Surveill Summ. 2000;49:39-50.

15. Gentleman JF, Pleis JR. The National Health Interview Survey: a primer. Eff Clin Pract. 2002;5:XX-XX.

16. National Center for Health Statistics. Data file documentation. National Health Interview Survey, 1998 (electronic data file and documentation). Hyattsville, MD: National Center for Health Statistics. Available at: (ftp://ftp.cdc.gov/pub/Health_Statistics/NCHS/Dataset_Documentation/NHIS/1998/srvydesc.pdf). Accessed on March 29, 2002.

17. National Center for Health Statistics. Data file documentation. National Health Interview Survey of Topics Related to the Year 2000 Health Objectives, 1995 (machine readable data file and documentation). Hyattsville, MD: National Center for Health Statistics.

18. Prevention and control of influenza: part I, vaccine recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Morb Mortal Wkly Rep. 1994;43:1-13.

19. Prevention and control of influenza: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Morb Mortal Wkly Rep. 1995;44:1-22.

20. The lungs in health and disease. Bethesda, MD: National Heart, Lung, and Blood Institute Information Office, National Institutes of Health. Available at: http://www.nhlbi.nih.gov/health/public/lung/other/lungs_hd.pdf. Accessed on April 22, 2001.

21. Facts about angina. Bethesda, MD: National Heart, Lung, and Blood Institute Information Office, National Institutes of Health. Available at: http://www.nhlbi.nih.gov/health/public/heart/other/angin.htm. Accessed March 29, 2001.

22. Facts about coronary heart disease. Bethesda, MD: National Heart, Lung, and Blood Institute Information Office, National Institutes of Health. Available at: http://www.nhlbi.nih.gov/health/public/heart/other/chdfacts.htm. Accessed on March 29, 2001.

23. Standard Occupational Classification Manual. Washington, DC: Office of Federal Statistical Policy and Standards, US Department of Commerce; 1980.

24. Dalaker J, Proctor BD. Poverty in the United States: 1999. US Census Bureau, Current Population Reports, Series P60-210. Washington, DC: US Government Printing Office; 2000.

25. Adams PF, Hendershot GE, Marano MA. Current estimates from the National Health Interview Survey, 1996. National Center for Health Statistics. Vital Health Stat 10:200;1999.

26. Shah B, Barnwell B, Bieier G. SUDAAN User's Manual, Release 7.5. Research Triangle Park, NC: Research Triangle Institute;1997.

27. Zhang J, Yu KF. What's the relative risk? A method of correcting the odds ratio in cohort studies of common outcomes. JAMA. 1998;280:1690-1.

28. Adams PF, Hendershot GE, Marano MA. Current estimates from the National Health Interview Survey, 1996. Appendix I: technical notes. National Center for Health Statistics. Vital Health Stat 10:200;1999.

Correspondence

John Pleis, MS, Division of Health Interview Statistics, National Center for Health Statistics, 6525 Belcrest Road, Room 860, Hyattsville, MD 20782; telephone: 301-458-4759; fax: 301-458-4035; e-mail: JPleis@cdc.gov.