Impact of the 1999 AAP/USPHS Joint Statement on Thimerosal in Vaccines on Infant Hepatitis B Vaccination Practices FREE

MMWR. 2001;50:94-97

1 figure omitted

On July 8,1999, the American Academy of Pediatrics (AAP) and the U.S. Public Health Service (PHS) jointly recommended reducing infant exposure to thimerosal, a commonly used vaccine preservative that contains mercury.^1- 2 Specific recommendations were made to postpone the first hepatitis B vaccine dose until 2-6 months of age for infants born to hepatitis B surface antigen (HBsAg)–negative (i.e., not hepatitis B virus [HBV]–infected) women.^1- 2 Infants born to HBsAg–positive (i.e., HBV-infected) women, or to women whose HBsAg status was unknown, were recommended to receive postexposure prophylaxis with the first dose of hepatitis B vaccine administered within 12 hours of birth.^1- 2 By mid-September 1999, when adequate supplies of preservative-free hepatitis B vaccine became available, PHS advocated a return to previous infant hepatitis B vaccination practices, including administering the first dose of hepatitis B vaccine to newborns in hospitals that had discontinued the practice.³ In 2000, preliminary assessments of the impact of these policy changes on routine hepatitis B vaccination practices were conducted by public health officials in Wisconsin, Oklahoma, Oregon, and Michigan. This report summarizes the results of these analyses, which indicate that many hospitals in Wisconsin have not reinstated policies to ensure routine administration of hepatitis B vaccine to newborns despite the availability of preservative-free hepatitis B vaccine, that the number of hepatitis B vaccine doses given to newborns in Oklahoma and Oregon has declined, and that an unvaccinated Michigan infant died from fulminant hepatitis B. Restoring routine newborn hepatitis B vaccination practices may require active advocacy by professional and government groups.

In Wisconsin in February 2000, the Division of Public Health mailed a survey to nurse managers of all Wisconsin birthing hospitals to assess the impact of the thimerosal statements on hepatitis B vaccination practices for newborns. Information was collected for the following periods: (1) before July 1999, (2) July-November 1999, and (3) March 2000. In Oklahoma and Oregon, data collected by previously established vaccination registries were used to assess the number of doses of hepatitis B vaccine administered to newborns before and after the publication of the thimerosal statements and after preservative-free hepatitis B vaccine became available. In Michigan, an infant death attributed to HBV was reported in January 2000, and an investigation by the Michigan Department of Community Health (MDCH) included a review of hospital and provider medical records and hospital vaccination policy changes in 1999.

All 110 birthing hospitals responded to the survey; 12 no longer provided obstetric services. The percentage of hospitals with written policies or standing orders for routine hepatitis B vaccination of all newborns declined from 81% before July 1999 to 10% during July-December 1999; 77% had policies or orders for routine vaccination of infants born to HBsAg-positive women during July-November 1999.

The proportion of births in hospitals where routine hepatitis B vaccination was given before discharge declined from 84% before July 1999 to 43% in March 2000. Before July 1999, 18 of 20 hospitals in southeastern Wisconsin, where 36% of HBsAg-positive pregnant women in the state resided during 1999, had written policies or standing orders to routinely provide hepatitis B vaccine to newborns. As of March 2000, five of these 18 hospitals had continued or resumed routine administration of hepatitis B vaccine to all newborns.

In Oklahoma and Oregon, the number of doses administered to newborns and young infants declined in July 1999 . In both states, the number of doses administered to newborns and young infants has not returned to pre-July 1999 levels. Among Oklahoma infants aged <1 month and Oregon infants aged <5 days, the number of hepatitis B vaccine doses administered during May-June 2000 declined 50% and 28,% respectively, compared with May-June 1999.

On December 14, 1999, a previously healthy 3-month-old infant was admitted to a hospital with diarrhea and jaundice, and acute hepatic failure attributed to HBV infection was diagnosed. The infant died on December 17, 1999. The infant had not received her first dose of hepatitis B vaccine until age 2.5 months.

The infant's mother was found to be HBsAg-positive at the first of 10 prenatal visits. However, the prenatal-care record provided to the birth hospital indicated that the mother was hepatitis-negative. Neither the provider nor the laboratory reported the mother's test results to MDCH as required by law. Before July 1999, the birth hospital had routinely administered hepatitis B vaccine series to newborns before discharge but had discontinued this practice in July 1999 because of concerns about thimerosal.

N Fasano, MA, J Blostein, MPH, Michigan Dept of Community Health. TN Saari, MD, Univ of Wisconsin, Madison; MB Hurie, MS, JP Davis, MD, Bur of Communicable Disease, Wisconsin Div of Public Health. S Dooley, ARNP, ED Rhoades, MD, Maternal and Child Health Svcs; D Blose, MA, Immunization Div, Oklahoma State Dept of Health. H Gillette, MPH, B Canavan, MPH, Immunization Program, Health Div, Oregon Dept of Human Svcs. Health Svcs Research and Evaluation Br and Program Operations Br, Div of Immunization Svcs, National Immunization Program and Hepatitis Br, Div of Viral and Rickettsial Diseases, National Center for Infectious Diseases, CDC.

The findings in this report indicate that the 1999 statements on thimerosal led to rapid changes in routine perinatal HBV infection prevention practices. Prevention of perinatal and early childhood infection by providing hepatitis B vaccine to newborns is a cornerstone of hepatitis B prevention strategies.⁴ An estimated 18,000 children aged <10 years were infected with HBV each year before universal infant hepatitis B vaccination was implemented in the United States (CDC, unpublished data, 2000). Approximately half acquired infection through perinatal transmission; the remainder acquired infection during early childhood through contact with other HBsAg-positive persons (horizontal transmission). HBV infection during infancy and childhood carries a higher risk for chronic HBV infection compared with infection during adulthood.^5- 6 Early hepatitis B vaccination is a safe and effective way to reduce the risk for both perinatal and horizontal HBV transmission and increases the likelihood of children completing the vaccine series on schedule.^7- 8

The reported case of acute liver failure from perinatal HBV infection in Michigan underscores the problems associated with discontinuing routine hepatitis B vaccination at birth without being certain that appropriate safeguards against perinatal infection are in place. Hepatitis B vaccine administered alone is 70%-95% effective in preventing perinatal HBV infection when the first dose is given within 24 hours of birth.⁴ Results from the Wisconsin survey are consistent with results from a national survey of 1000 birthing hospitals conducted during December 1999, 3 months after thimerosal-free vaccine became widely available for infants. In this national survey, the percentage of hospitals with written policies or standing orders for routine hepatitis B vaccination of newborns born to HBsAg-negative women declined from 85% before the 1999 thimerosal statement to 34% in December 1999 (S.J. Clark, University of Michigan, personal communication, 2000). Of 88 hospitals that had discontinued written policies or standing orders for routine vaccination of newborn infants, including infants born to HBsAg-positive women, 67% had not reinstated the policies or standing orders (S.J. Clark, University of Michigan, personal communication, 2000).

It is unknown whether changes in hospital policies and reductions in hepatitis B vaccination coverage of newborns are causing other missed opportunities for vaccination among infants at high risk for perinatal infection, especially among those born to unscreened and HBsAg-positive women. The impact of the public and private health-care system response to concerns about thimerosal may not be understood fully until ongoing analysis of surveillance data and birthing hospital chart reviews provide a more complete assessment of the number of infants who acquired chronic HBV infection as the result of missed vaccination opportunities. CDC is supporting such studies in several states.

AAP and PHS advocate the reintroduction of routine hepatitis B vaccination policies for all newborn infants born in hospitals in which this practice was discontinued because of concerns about thimerosal.^3,8 After administering a dose at birth, providers may complete the series with either 2 more doses of single antigen hepatitis B vaccine or with 3 doses of combination Haemophilus influenzae type b/hepatitis B vaccine according to previously recommended schedules.⁹ All birthing hospitals should have hepatitis B vaccine available for use in infants born to HBsAg-positive and unscreened women. Hospitals should continue to vaccinate all infants at birth until procedures are in place to guarantee that (1) the HBsAg status of every pregnant woman is available and reviewed at delivery, (2) appropriate passive-active immunoprophylaxis (HBIG and hepatitis B vaccine) is provided for infants of HBsAg-positive women within 12 hours of birth, and (3) appropriate active immunoprophylaxis (hepatitis B vaccine) is provided for infants of women with an unknown HBsAg status. Pregnant women who are identified as HBsAg-positive should be reported to local or state health departments to ensure that their infants, family, and household contacts receive a full hepatitis B vaccination series.

Vaccination practices are influenced substantially by recommendations of professional and government advisory groups. The 1999 joint statement and the subsequent AAP guidelines were issued as a precautionary measure and were intended to apply only to infants born to HBsAg-negative women. The inadvertent effect in many hospitals was a persisting change in policies for administering hepatitis B vaccine to infants, most importantly to infants born to HBsAg-positive and unscreened women for whom no changes in vaccination practices had been recommended. Changes in established recommendations, especially if they occur without timely communication and education of health-care providers, may result in misinterpretation and unanticipated changes in vaccination practices.

MMWR. 2001;50:162-166

2 tables, 1 figure omitted

Colorectal cancer is the second leading cause of cancer-related death in the United States.¹ An estimated 135,400 new cases and 56,700 deaths from colorectal cancer are expected during 2001.¹ Since the mid-1990s, national guidelines have recommended that persons aged ≥50 years at average risk for colorectal cancer should have screening tests regularly. To estimate rates for the use of colorectal cancer screening tests and to evaluate trends in test use, CDC analyzed data from the 1999 Behavioral Risk Factor Surveillance System (BRFSS) on the use of a home administered fecal occult blood test (FOBT) and sigmoidoscopy/colonoscopy, and then compared them with similar data from 1997. The findings in this report indicate that the proportion of the U.S. population that has been screened remains low. In 1999, 44% of BRFSS respondents reported receiving FOBT and/or sigmoidoscopy/colonoscopy within the recommended period compared with approximately 41% reporting FOBT and/or sigmoidoscopy/proctoscopy within the recommended period in 1997.² Efforts to address barriers and to promote the use of colorectal cancer screening should be intensified.

In 1999, the 50 states, District of Columbia, and Puerto Rico participated in BRFSS, an ongoing, statebased, random-digit–dialed telephone survey of the civilian, noninstitutionalized population aged ≥18 years. A total of 63,555 respondents aged ≥50 years were asked whether they ever had FOBT using a home kit, whether they ever had sigmoidoscopy or colonoscopy, and when the last test had been performed. Responses coded as "don't know/not sure" or "refused" were excluded from analyses (<2%). Aggregated and state-specific proportions, standard errors, 95% confidence intervals, and p-values were calculated using SAS and SUDAAN.

Data in this analysis were weighted to the age, sex, and race/ethnicity distribution of each state's adult population using intercensal estimates and were age standardized to the 1999 BRFSS population. The median state response rate of 56.7% (range: 38.4%-83.9%) was calculated using the cooperation rate formula (i.e., the number of completed interviews divided by the number of potential respondents [households with a resident aged ≥18 years]). The 1999 questions about the use of sigmoidoscopy were modified from the 1997 questions. In 1997, respondents were asked whether they had received sigmoidoscopy or proctoscopy. Proctoscopy is performed with a shorter instrument than sigmoidoscope and is not recommended as a colorectal cancer screening test. In 1999, "sigmoidoscopy/proctoscopy" was replaced with "sigmoidoscopy/colonoscopy." Colonoscopy evaluates the entire colon and is recommended once every 10 years in some guidelines.^3,4 For this report, "sigmoidoscopy/proctoscopy" and "sigmoidoscopy/colonoscopy" are referred to as "sigmoidoscopy" unless otherwise specified.

In 1999, 40.3% (25,263 of approximately 63,000) of respondents reported ever having FOBT, and 43.8% (26,388) of the respondents reported ever having sigmoidoscopy. For tests received within the recommended period, 20.6% (12,518) had FOBT within the year preceding the survey, 33.6% (19,535) had sigmoidoscopy within the preceding 5 years , and 44.0% (25,871) had either FOBT within the year preceding the survey or sigmoidoscopy within the preceding 5 years . In 1997, 19.6% (9832 of approximately 51,000) of the respondents had FOBT within the year preceding the survey, and 30.3% (14,678) had sigmoidoscopy within the preceding 5 years. Although these rate changes in testing use were statistically significant (p<0.05), actual increases were small. By state, the proportion of respondents who had FOBT within the preceding year ranged from 8.2% (112 of 1366) in Puerto Rico to 36.4% (187 of 500) in the District of Columbia; the proportion that had sigmoidoscopy/colonoscopy within the preceding 5 years ranged from 20.4% (275 of 1357) in Puerto Rico to 46.1% (410 of 981) in Delaware.

following state BRFSS coordinators: S Reese, MPH, Alabama; P Owen, Alaska; B Bender, MBA, Arizona; G Potts, MBA, Arkansas; B Davis, PhD, California; M Leff, MSPH, Colorado; M Adams, MPH, Connecticut; F Breukelman, Delaware; I Bullo, District of Columbia; S Hoecherl, Florida; L Martin, MS, Georgia; F Reyes-Salvail, MS, Hawaii; J Aydelotte, MA, Idaho; B Steiner, MS, Illinois; L Stemnock, Indiana; J Davila, Iowa; C Hunt, Kansas; T Sparks, Kentucky; B Bates, MSPH, Louisiana; D Maines, Maine; A Weinstein, MA, Maryland; D Brooks, MPH, Massachusetts; H McGee, MPH, Michigan; N Salem, PhD, Minnesota; D Johnson, MS, Mississippi; J Jackson-Thompson, PhD, Missouri; P Feigley, PhD, Montana; L Andelt, PhD, Nebraska; E DeJan, MPH, Nevada; L Powers, MA, New Hampshire; G Boeselager, MS, New Jersey; W Honey, MPH, New Mexico; C Baker, New York; Z Gizlice, PhD, North Carolina; L Shireley, MPH, North Dakota; P Pullen, Ohio; K Baker, MPH, Oklahoma; K Pickle, MPH, Oregon; L Mann, Pennsylvania; Y Cintron, MPH, Puerto Rico; J Hesser, PhD, Rhode Island; M Wu, MD, South Carolina; M Gildemaster, South Dakota; D Ridings, Tennessee; K Condon, MS, Texas; K Marti, Utah; C Roe, MS, Vermont; K Carswell, MPH, Virginia; K Wynkoop Simmons, PhD, Washington; F King, West Virginia; K Pearson, Wisconsin; M Futa, MA, Wyoming. Epidemiology and Health Svcs Research Br, Div of Cancer Prevention and Control, National Center for Chronic Disease Prevention and Health Promotion, CDC.

Since 1997, the proportion of the U.S. population that reported having had FOBT and sigmoidoscopy has increased slightly but remains low. Various factors may contribute to the continued underuse of these tests, including lack of knowledge by the public and health-care providers of the effectiveness of screening and low reimbursement rates for health-care providers who perform screening tests.^5,6

The findings in this report are subject to at least four limitations. First, because of the wording change in the BRFSS questionnaire from "sigmoidoscopy/proctoscopy" in 1997 to "sigmoidoscopy/colonoscopy" in 1999, comparing endoscopic procedures for these years must be interpreted with caution. Data on the use of colonoscopy were collected only in 1999; however, some tests reported as sigmoidoscopies/proctoscopies in 1997 probably were colonoscopies because some respondents may have been unable to distinguish among the three tests. It is unknown whether the reported increase from 1997 to 1999 represents a true increase in sigmoidoscopy use or previously unmeasured rates of colonoscopy use. Second, because the survey was administered over the telephone, only persons who own telephones were represented in this analysis. Third, 43.3% of the eligible respondents were contacted but did not complete the telephone interview or could not be reached for an interview. Finally, responses were self-reported and were not validated through medical record review.

For persons aged ≥50 years at average risk for colorectal cancer, recommended screening options include one or more of the following tests: annual FOBT, sigmoidoscopy every 5 years, colonoscopy every 10 years, or double-contrast barium enema every 5-10 years.^3,4,7 Despite their efficacy in reducing incidence and mortality from colorectal cancer,⁸ screening tests are underused. To draw attention to this disease, the U.S. Congress designated March as "National Colorectal Cancer Awareness Month." During March 2001, CDC and the Health Care Financing Administration launched the third annual "Screen for Life: A National Colorectal Cancer Action Campaign." Using print, television, and radio announcements and brochures and fact sheets, the campaign was designed to raise awareness of colorectal cancer and to encourage persons aged ≥50 years to discuss screening with their health-care provider and select the appropriate test(s). CDC also produced "A Call to Action: Prevention and Early Detection of Colorectal Cancer," a slide presentation for health-care providers. All material is available on the World-Wide Web, http://www.cdc.gov/cancer/screenforlife and http://www.cdc.gov/cancer/colorctl/calltoaction/slide_index.htm.

References: 8 available

MMWR. 2001;50:120-125

2 tables omitted

In the United States, the number of persons reporting disabling conditions increased from 49 million during 1991-1992 to 54 million during 1994-1995.^1-4 During 1996, direct medical costs for persons with disability were $260 billion.⁵ Surveillance of disability prevalence and associated health conditions is useful in setting policy, anticipating the service needs of health systems, assisting state programs, directing health promotion and disease prevention efforts, and monitoring national health objectives.^6-8 The U.S. Bureau of the Census and CDC analyzed data from the Survey of Income and Program Participation (SIPP) to determine national prevalence estimates of adults with disabilities and associated health conditions. This report summarizes findings of that analysis, which indicate that disability continues to be an important public health problem, even among working adults, and arthritis or rheumatism, back or spine problems, and heart trouble/hardening of the arteries remain the leading causes. Better health promotion and disease prevention may reduce the prevalence of disability-associated health conditions.

The 1996 SIPP panel was a multistage, stratified sample of the U.S. civilian, noninstitutionalized population based on the 1990 U.S. census. Panel members were interviewed 12 times in 4 years. During August-November 1999, the Adult Disability Topical Module of Wave 11 of the 1996 SIPP panel collected information about self-reported disability during personal interviews with persons in 36,700 households representative of the civilian, noninstitutionalized population aged ≥15 years. For this analysis, disability was defined as self-reported or proxy-reported difficulty with or reporting one or more of eight measures: (1) difficulty with one or more specified functional activities*; (2) difficulty with one or more activities of daily living (ADLs)*; (3) difficulty with one or more instrumental activities of daily living (IADLs)*; (4) reporting one or more selected impairments*; (5) use of assistive aids (e.g., wheelchair, crutches, cane, or walker) for >6 months; (6) limitation in the ability to work around the house; (7) limitation in the ability to work at a job or business (data for persons aged 16-67 years); and (8) receiving federal benefits on the basis of an inability to work. A subset of persons with disability also reported the main cause of their disability from a list of 30 associated health conditions. This subset, defined before the survey was conducted, comprised persons reporting difficulty with ADLs, IADLs, selected functional activities (excluding seeing, hearing, and having their speech understood by others), or limitation in the ability to work around the house or at a job or business. National estimates were calculated using sample weights representing the inverse of the probability for selection and complex adjustments for nonresponse and subsampling.⁶

The analysis focused on 53,636 adults aged ≥18 years (consistent with standard age categories used in other national surveys). In 1999, 44 million (22%) adults reported having a disability. The prevalence rate of disability was 24% among women and 20% among men. Approximately 32 million adults had difficulty with one or more functional activities such as climbing a flight of stairs (19.4 million), walking three city blocks (19 million), or lifting/carrying 10 lbs (14.2 million); approximately 16.7 million adults had a limitation in the ability to work around the house; 11 million had either selected impairments or difficulty with IADLs. Two million adults used a wheelchair, and seven million used a cane, crutches, or a walker. Of the total percentage of disabilities, 63% occurred among working adults (aged 8-64 years); of these, 27.8 million (16.5%) had a disability and 17.7 million (10.5%) had a limitation in the ability to work at a job or business. Of those adults aged ≥65 years, 16.3 million (50%) had a disability. The age-specific prevalence rate of disability was the highest among respondents aged ≥65 for all functional activities, ADLs, and IADLs.

Of all adults with disabilities, 41.2 million (93.4%) reported their main health condition associated with their disability (Table 2); 7.2 million (17.5%) had arthritis and rheumatism, 6.8 million (16.5%) had back or spine problems, and 3.2 million (7.8%) had heart trouble/hardening of the arteries. Women had higher rates in the arthritis or rheumatism and "other" associated health conditions categories than men. Men had higher rates of heart trouble/hardening of the arteries and deafness or hearing problems than women.

JM McNeil, J Binette, Bur of the Census, Economics and Statistics Administration, US Dept of Commerce. Disability and Health Br, Div of Birth Defects, Child Development, and Disability and Health, National Center for Environmental Health; Health Care and Aging Studies Br, Div of Adult and Community Health, National Center for Chronic Disease Prevention and Health Promotion, CDC.

Disability affects more than one in five adults. Rates of disability are higher among older adults who also have higher rates of chronic diseases. However, most disability occurs during the working years, which contributes to the high cost estimates of disability. Arthritis or rheumatism, back or spine problems, and heart trouble/hardening of the arteries continue to be the leading causes of disability. This report differs from a similar 1994 report by focusing on adults only and using a broader definition of disability.⁴

The strengths of SIPP include a survey design that allows nationally representative population estimates of disability. The broad definition of disability used in SIPP also provides a sensitive estimate of disability prevalence that is less likely to overlook persons with disability than other definitions (e.g., clinical or federal benefit program-based definitions). SIPP links disability with associated health conditions, providing information that usually is not available from other data sources. This information is important because many programs address disability prevention by disease or condition.

The findings in this report are subject to at least five limitations. First, despite complex statistical adjustment procedures used to address nonresponse over time, these procedures may not have completely eliminated bias that resulted from nonresponse errors, especially in subgroup analyses. Second, this report excluded persons in institutions, in the military, and aged <18 years. Third, persons with multiple disabilities may attribute the main disability to the one most disabling at the time of the interview, which may result in inconsistent survey responses. Fourth, because of questionnaire design, the main associated health condition was determined for most but not all adults with disability; 2.9 million (6.4%) persons whose only disabilities were difficulty with vision, hearing, or speech, who had selected impairments, used assistive aids, or received federal disability benefits were not asked about a main condition. Finally, the definition of disability used did not assess environmental and social barriers, discrimination as the result of disability, and effects on the workforce. These issues are addressed in the International Classification of Functioning, Disability, and Health (ICIDH-2), a unified and standard framework that describes the dimensions of disability.⁹ICIDH-2 complements the International Classification of Diseases by organizing information around three dimensions: body level (body systems and structure), person and society level (activities and participation), and the environment. Because of the dynamic quality of disability, a limitation in one dimension does not predict a limitation in another.

These estimates demonstrate the large impact of disability in working age and older adults and the relative contributions of associated health conditions, and provide information for public health policy makers and health systems. More detailed analyses relating the eight measures of disability and associated health conditions can assist disease-specific efforts in planning, health promotion and disease prevention, and surveillance of disability-related national health objectives.¹⁰ With increasing life expectancy and the aging of the population, health issues related to disability are likely to increase in importance.

References: 10 available

*Specified functional activities: ability to see words or letters in ordinary newspaper print, hear normal conversations, have speech understood by others, lift/carry 10 lbs, climb a flight of stairs without resting, and walk three city blocks. ADLs: getting around inside the home, getting in/out of a bed/chair, bathing, dressing, and toileting. IADLs: getting around outside the home, taking care of money and bills, preparing meals, doing light housework, and using the telephone. Selected impairments: learning disability, mental retardation, other developmental disability, Alzheimer disease/senility/dementia, and other mental disabilities.³