Objective. The Standards for Pediatric Immunization Practices suggest that hospitalization be viewed as an opportunity to vaccinate children. The purpose of the present study is 1) to determine the immunization status of an urban population of hospitalized preschool-aged children, 2) to study the impact of an immunization program designed to vaccinate hospitalized 0 to 2-year-old children who are underimmunized at admission, and 3) to make immunization a routine part of care for the hospitalized child.
Methods. Prospective evaluation of the immunization status of hospitalized 0 to 2-year-old residents of Philadelphia admitted to an urban children's hospital was performed. With verification of the child's immunization record through the primary care provider (PCP), needed immunizations were given and records were forwarded to notify the PCP. Educational information was provided to families and health care providers.
Main Outcome Measure. The percentage of children fully immunized on admission compared with the percentage at the time of discharge.
Results. Two thousand three hundred twenty-nine children from 0 to 2 years of age were hospitalized during the 22-month study period. Immunization records were verified in 86% (2006), requiring an average of 1.5 phone calls to the PCP. The mean patient age was 10 months. Average hospital length of stay was 4 days. On admission, 49% (980) of the 2006 study patients were fully immunized. The remaining 51% (1026) were eligible for vaccination. Immunizations were delayed greater than or equal to 2 months in 18% (355) of the children. Neither type of health care insurance nor site of primary care affected the immunization status of those evaluated at the time of admission.
Sixty-six percent (N = 674) of eligible patients received at least one vaccination before hospital discharge. Medical contraindications accounted for only 4% of the reasons eligible patients were not immunized. Of the 2006 children evaluated, the percentage of those fully vaccinated for age increased significantly from 44% on admission to 70% on discharge.
Conclusion. As a result of this program, there was a significant improvement in vaccination percentage at the time of hospital discharge in this group of urban preschool-aged children. The development of an immunization program to vaccinate hospitalized preschool children is an opportunity to immunize in the urban setting where there is a high prevalence of underimmunization.
In addition, it provides an opening for educational programs for families, nurses, and housestaff and linkage to the community PCPs.
Despite some improvement over the last 5 years, vaccine coverage for preschool-aged children falls short of goals set in the recent Childhood Immunization Initiative of 1993 of full immunizations for at least 90% of children by their second birthday.1
The most recent data from the Centers for Disease Control and Prevention (CDC) (National Health Interview Survey) reveals that as of July 1994 to June 1995, approximately 71% of 2-year-old children are fully vaccinated.1 The urban preschool-aged child may be at even greater risk for underimmunization. Vaccination coverage at 24 months of age among Latino and African-Americans living in south central and east Los Angeles was 42% and 26%, respectively, in 1992.2 Furthermore, urban children are more likely to experience substantial barriers in access to and use of primary care and these barriers may worsen if congressional initiatives to reduce spending for programs designed to improve access to healthcare are successful.3 Therefore, it seems reasonable that community-based primary care and hospital-based practices caring for urban children join to eliminate missed opportunities to immunize using every interaction with the health care system as an opportunity to vaccinate, including hospitalization.
The purpose of the current study is to determine the need for immunizations in an urban population of hospitalized preschool-aged children and to report the impact of an immunization program that was designed to:
Identify hospitalized 0 to 2-year-old children who are underimmunized and provide needed vaccination before discharge,
Communicate with PCPs about vaccines provided,
Make immunization a routine part of care for the hospitalized child, and
Provide education about the importance of eliminating missed opportunities to nursing, housestaff physicians, PCPs, and parents.
The Immunization Team Organization
In January 1994, The Children's Hospital of Philadelphia, with funding from the Philadelphia Health Department and the CDC, established an immunization team (consisting of a physician, a nurse coordinator, and an assistant) to implement a program designed to immunize hospitalized preschool aged children before discharge. The hospital, located in west Philadelphia, serves the local communities in the western and southern sections of the city with 15 000 inpatient admissions and 200 000 combined emergency department and outpatient visits each year. In this section of the city, there are approximately 47 000 children from birth to 4 years of age, 85% are African-American.
Implementation of the program required communication and planning within the hospital and with the external, PCP groups. Hospital preparation involved informing nursing, resident and attending physician groups, inpatient pharmacy (who had to set up a system to ensure that the patients were charged for the vaccine they dispensed), and the unit base clerks. External preparation required letters and phone calls to different health care providers in our community to explain the program and address their concerns, if any.
The immunization nurse coordinator organized the inpatient nursing staff on six different medical and surgical units orienting one new unit a month for the first 6 months. Patients in the intensive care units and the oncology unit were not included. The inpatient nursing staff on each unit chose representatives to serve on an immunization task force for that unit. During the start up of the program, quarterly meetings with the task force were held to provide education and updates on immunization practices and to discuss problems and progress of the inpatient program. Subsequently, the task force representatives acted as a link between the inpatient units and the immunization nurse coordinator.
Patient Identification and Vaccination
Prospective evaluation of the immunization status of all hospitalized 0 to 2-year-old residents of Philadelphia admitted to an urban children's hospital was begun in February 1994. On a daily basis, the immunization nurse coordinator and assistant identified all age appropriate children admitted to six different medical and surgical units. The PCP of each child was contacted to verify the child's utilization of the primary care site as well as to retrieve the immunization record.
After discussions with the PCP and review of the child's medical status and immunization record, the pediatric resident assigned to the patient was contacted and in the first 2 years of the program, was told what immunizations were needed if the child was delayed. Subsequently, the entire immunization record and the recommended immunization are taped to the front cover of the patient chart. Before vaccination, informed consent was obtained and educational information and teaching were provided about future immunizations by the inpatient nursing staff and/or the immunization nurse coordinator. The vaccine was distributed by the hospital pharmacy without charge under the CDC vaccine program. Immunizations were administered by the nursing staff on the day of discharge from the hospital. Once the child received the vaccine, an updated immunization record was mailed or faxed to the PCP.
The child's immunization status was determined on admission. Delays were determined based on the consensus or harmonized schedule agreed upon by the American Academy of Pediatrics, the American Academy of Family Practice and the CDC Advisory Committee on Immunization Practices.4 The harmonized schedule includes an extended acceptable period over which oral polio vaccine (OPV) and measles-mumps-rubella (MMR) vaccines may be administered. The immunization status was designated due for next vaccination if the child was eligible for any vaccination at the time of hospitalization but was less than 2 months delayed for age. The immunization status was designated delayed if the child was more than 2 months delayed for age.
Contraindications to immunization were determined during discussion with the team of physicians caring for the child while hospitalized (attending and housestaff) and the PCP. In general, children were considered medically eligible for vaccination before discharge with the diphtheria-pertussis and tetanus vaccine if they had no fever (<38°C), were recovering from illness, and appeared well. Likewise, children were eligible for OPV, MMR, and/or Haemophilus influenzae type b vaccinations if they were immunocompetent and had fevers <39°C, were recovering from illness, and appeared well before discharge.5 If the patient did not meet the above criteria and/or the PCP or inpatient attending physician requested no vaccination for other reasons not covered above, then immunizations were withheld.
Data Collection and Analysis
Demographic information on all evaluated children was collected from the inpatient record including PCP, insurance type, and length of stay in the hospital. In addition, dates and types of immunizations were recorded. Information on the success and effort required to obtain the immunization records was monitored. Finally, the number and types of vaccines given to each child and the immunization status at the time of discharge were recorded for all children receiving vaccinations during their hospitalization.
χ2 analysis was used for analysis of categorical data. P values <.05 were considered statistically significant. Finally, the cost of the program was analyzed. The cost of vaccine is based on CDC cost per dose from a price list published January 25, 1996 (DOC#2639).
During the first 22 months (February 1994 to November 1995) of the inpatient vaccination program, there were 2329 admissions of 0 to 2-year-old children to the medical and surgical units involved in the program. Of this group, 2080 (89%) children were primarily evaluated by the immunization team. The majority of the 249 children not evaluated were missed during the first 6 months of the start-up phase of the program, a time when a new medical or surgical unit was introduced to the program each month. The remainder of patients who were missed included those infants in a neonatal transitional unit (not included until year 2 of the program), children on the oncology unit or patients who were admitted and discharged from the hospital in a single weekend. Immunization records were available for review for 2006 or 96% of the 2080 children. The immunization records of 73 children (or 4% of the group of 2006) arrived after the children were discharged and were included in the analysis. The number 2006 comprises the group of children who were fully evaluated for immunization status and need for vaccination.
All children evaluated lived in the city of Philadelphia, 71% were from west and south Philadelphia. The mean age of the evaluated group was 10 months (±6.5), 57% were boys. The average length of hospitalization was 4.3 days (±4.2) (range 0 to 99). An average of 1.5 phone call attempts, (±1.1) (range 0 to 11), were required to obtain the immunization records from the PCP. On average, it took 4 hours each day to retrieve the immunization records. Only a few of the parents had immunization records with them. During the course of the program, more than 219 different primary care sites were contacted to verify immunization records.
Of those hospitalized children evaluated, 51% (1026) were eligible for vaccination during hospitalization. Of those who were eligible for immunization, 65% (671) were designated as due for next vaccination, and 35% (355) were delayed. There were no significant differences in immunization status between the sites of primary care (Table1). Eighty-two percent of the evaluated group were cared for either in privately owned practices or in hospital-based clinics.
The majority of children (55%) were insured by Medicaid managed care (Table 2). Children with private insurance were just as likely to be underimmunized as Medicaid-insured families. Furthermore, there was no difference in immunization status between Medicaid managed care insurers and Medicaid fee for service. The families who listed self-pay on admission (ie, no insurance), while 6% of the total group, were significantly more likely (P < .05; odds ratio 1.5, 95% confidence interval, 1.0 to 2.5) to be underimmunized than the insured families.
Of 1026 children eligible for vaccination, 66% (674) received at least one immunization before discharge. Eighty-five percent of those vaccinated received multiple immunizations. Over 1504 immunizations were given. Seventy-seven percent of the vaccinated group were brought up to date during their hospitalization. Of the 2006 children evaluated, the percentage of hospitalized children who were up to date or fully immunized for age increased significantly from 44% on admission to 70% on discharge (P < .0001; Figure).
Thirty-four percent of the 1026 children who could have been vaccinated while hospitalized did not receive any immunizations. Table3 outlines the reasons for not vaccinating these children. Interestingly, after the change in the way we informed the nurses and physicians of the immunizations that the hospitalized child required (see “Methods”) the number of ordered, not administered, missed opportunities decreased by 53%. Note that in 41 children (4%), immunizations were withheld because it was considered a medical contraindication.
Although all vaccines are provided free under the Federal Immunization Program, cost of the vaccines used (based on CDC cost per dose from price list DOC#2639, January 25, 1996) averaged $505 per month. Salary support for the immunization team was $6354 per month.
In 1988, Tifft and Ledermann6 found that 19% of hospitalized children in Baltimore needed vaccination and suggested that programs be developed to improve immunizations among hospitalized children. In 1994, eighteen standards for Pediatric Immunization Practices were recommended by the National Vaccine Advisory Committee and endorsed by the American Academy of Pediatrics. Among the standards is one that requests that each encounter with a health care provider, including an emergency room visit or hospitalization, is an opportunity to screen the immunization status and, if indicated, administer needed vaccines. The standard also calls for informing the child's regular health care provider about the immunizations administered.5 The purpose of the standard is to eliminate missed opportunities. Furthermore, recent data suggests that these nontraditional sites may be fertile areas to vaccinate preschool aged children.2,7 In the current study, 1026 hospitalized children (51% of those evaluated) were eligible to receive immunization(s) during hospitalization and 35% were significantly delayed on admission.
Interestingly, the site of primary care or the type of insurance had no affect on the immunization status of the children evaluated on admission. Children from private clinics were no more likely to be delayed than children from other sites (Table 1). This result differs from a recent study in New York where only 26% of children who attended private clinics were up to date for their age for diphtheria-pertussis and tetanus, OPV, and MMR compared with a city-wide coverage of 49%.8 Furthermore, in the current study, Medicaid-managed care insured children were not more likely to be adequately immunized than children with Medicaid, fee for service (Table 2).
The main obstacle to providing immunization in nontraditional settings is verification of the immunization records. Parent or caretaker memory of what immunizations have been given or missed has been shown to be inaccurate.9 An effective immunization program in the emergency department, for example, requires a method of rapid retrieval of the immunization record; a requirement that renders emergency department immunization impractical in most settings.7Conversely, hospitalization allows for time to obtain the immunization record. With an average length of stay of 4 days, 96% of records were obtained for review before discharge, the most time-consuming task of the immunization team. On average, more than one phone call attempt was required to the primary care site to obtain the record. Facsimile transmission of these records has recently become the most frequent mode of communication.
Indeed, obtaining the immunization record was extremely time consuming requiring one full-time person 4 to 5 hours a day to make the calls to the primary providers office. Two things helped to facilitate record retrieval. First, an introductory letter, phone call, and an occasional visit by the immunization team to the primary providers office was necessary to explain the program and establish a contact person at each office. Second, the Philadelphia Department of Public Health made it known to all providers that exchange of immunization records between health care providers was appropriate as a public health measure and did not require parental consent. Much of the work of the immunization team could be eliminated if an accessible computerized immunization data base existed for preschool aged children. Certainly, easy on line access to immunization records will become even more vital as health care reform encourages shorter and shorter lengths of stay in the hospital.
Although (51%) 1026 hospitalized children were eligible to receive immunization during hospitalization, only 66% received at least one immunization. As compliance with the program was evaluated, the number of immunizations that were not ordered (ordered, not written) or not administered (ordered, not administered) was unacceptably high (Table 3). Thirty-four percent of the not vaccinated children represented missed opportunities by the health care workers in these two categories. Reasons for this involved misunderstanding about contraindications to vaccinate and orders being overlooked by nursing during a transition to a computerized ordering system. It was discovered that taping the patients immunization record with the recommended vaccines to be given on the front cover of the child's inpatient chart was more effective than calling the resident and prescribing the needed vaccines. The nurses and resident physicians had a visual reminder on the patient chart that the child needed to be vaccinated and they could see and decide for themselves that the patient was indeed delayed by glancing at the immunization record. In only 41 children (4% of cases) were immunizations withheld for medical reasons. This low percentage is similar in other studies that found illness to be an unusual reason to withhold needed immunizations in both the inpatient and outpatient settings.10-12 Only 6% of parents refused immunizations for their children.
In the current study, 49% of 2006 preschool aged children were up to date on admission to the hospital. These results are similar to a report from Memphis in 1992 to 1993, in which 44% of 142 hospitalized children less than 2 years of age were up to date in their immunizations.13 Furthermore, the National Immunization Survey implemented by the CDC, performed a single survey for July 1994 to June 1995 of caretakers for children 19 to 35 months of age in the US. In Philadelphia, 67 (±7.5)% were found to be fully immunized, receiving 4 diphtheria, pertussis and tetanus; 3 oral polio; 1 MMR; and 3 Haemophilus influenzae type b vaccines. There is no data specifically for west and south Philadelphia. But in general, immunization rates for the children admitted to the hospital were lower than the National Immunization Survey immunization rates for Philadelphia, 49% compared with 67%.1
Despite the fact that immunizations were not given to 34% of children eligible to receive them, the percent of those who were fully immunized for age significantly improved from 47% on admission to 70% (P > .0001) on discharge from the hospital. In other words, 19% of the preschool-aged children evaluated were brought up to date for immunizations during their hospitalization.
Funding for this program provided salaries for a full-time nurse, an assistant, and a very small amount for a physician director. However, this type of immunization program could potentially be started with only minimal funding. Obtaining the records is a vital time-consuming aspect of the program that may require at least a dedicated person working half-time. Nursing education, especially in the first 6 months, may require a coordinator. The hospital, health maintenance organizations, or vaccine manufacturers may be willing to provide the relatively small amount of funds needed to get the program started. Once started, if the work of the immunization team can be incorporated into the routine care of hospitalization then no funding is needed.
Finally, side benefits of this program, that were not formally studied but felt to be a very positive aspect, were in the realm of communication and education. Over 219 different primary care sites were contacted to verify immunization records. The team approach to vaccinating the children of our community was felt to be a positive aspect of the program. The importance of eliminating missed opportunities was an implicit message in these communications both before and after immunizations were given. Heightened awareness of the importance of immunization and the use of nontraditional sites for immunizations was also an important concept learned by the nursing staff and housestaff. Another benefit was related to the Joint Commission on Accreditation of Healthcare Organizations which expects that hospitals support and develop community outreach programs related to health care. This program certainly fills part of that goal.
Further research is underway to study the impact of different types of in hospital educational interventions on immunization rates and subsequent visits to the primary care clinic for preschool-aged children.
This work was supported through a contract with the CDC and the Philadelphia Department of Public Health.
The authors wish to thank Flaurence Dong and Travis Quinn for their help in data collection, Tracey Sampson for preparation, Barbara Watson, MD and Evaline A. Alessandrini, MD for their careful review of the manuscript.
- Received June 11, 1995.
- Accepted February 15, 1996.
Reprint requests to (L.M.B.) Section of Infectious Diseases, 34th Street and Civic Center Blvd, Philadelphia, PA 19104.
Presented at the meeting of the Ambulatory Pediatric Association, May, 1995, San Diego, CA and at the 30th National Immunization Conference, April, 1996, Washington, DC.
- CDC =
- Centers for Disease Control and Prevention •
- PCP =
- primary care provider •
- OPV =
- oral polio vaccine •
- MMR =
- Wood D,
- Donald-Sherbourne C,
- Halfon N,
- Newacheck PW
- Hall CB
- ↵Centers for Disease Control and Prevention. Standards for Pediatric Immunization Practices. 6th printing. Atlanta, GA: Centers for Disease Control and Prevention; 1994:10
- Bell LM,
- Lopez NI,
- Pinto-Martin J,
- Casey R,
- Gill FM
- Fairbrother G,
- Friedman S,
- DuMont KA,
- Lobach KS
- Goldstein KP,
- Kviz FJ,
- Daum RS
- McConnochie KM,
- Roghmann KJ
- Szilagyi PG,
- Rodewald LE,
- Humiston SG,
- et al.
- Kum-Nji P,
- James D,
- Herrod HG
- Szilagyi PG,
- Rodewald LE,
- Humiston SG,
- et al.
- Copyright © 1997 American Academy of Pediatrics