Center for the Vulnerable Child, Childrens Hospital Oakland

747 Fifty Second Street
Oakland, California 94609
Contact: Cheryl Zlotnick, RN, DrPH
Phone: (510) 428-3783
Key words: children; immunization; antibiotic treatment for Otitis Media

OBJECTIVE

The goal of this project was to test the utility of two standard pediatric outcome measures--antibiotic treatment for Otitis Media (OM) and age-appropriate immunization status--and determine their usefulness for HCH programs that target children.

METHODS
A retrospective study design was used; medical records were abstracted on Center for the Vulnerable Child (CVC) clients treated in urgent or ambulatory care clinics between September 1996 and February 1997. The study population consisted of homeless children or children at-risk of becoming homeless admitted to the CVC who received care in the Children_s Hospital Oakland_s Ambulatory Care Services_ clinics. Children in this sample ranged from under 1 month to 13 years old. The median age was 4 years old. More than 60 percent were male and more than 80 percent were African-American. A total of 230 CVC children (unduplicated count) made visits to either urgent or ambulatory care clinics during the 6 month period. Of these 204 medical records were obtained. Unfortunately, 53 medical records were sent with only hospitalization information. As a result, only 151 records or 74 percent of records were reviewed for this study. A one-page instrument was developed to allow for the abstraction of information from medical records; it was designed to elicit information on OM and age-appropriate immunizations for children aged 2 years and under. In addition to preliminary identifiers, the instrument included requests for the medical record abstracter to document dates for up to three medical visits resulting in the diagnosis of OM. For each of these three dates, a space was provided to indicate whether any antibiotic treatment was prescribed. A final item in this section asked about permanent hearing loss.

The remainder of the instrument focused on immunizations. This evaluation used guidelines developed and approved by the Advisory Committee on Immunization Practices, American Academy of Pediatrics, and American Academy of Family Physicians. Because of anticipated comparisons to published immunization rates, the instrument_s format mirrored that used by the Centers for Disease Control and Prevention (CDC) and the 1996 United States National Immunization Survey (NIS) on children 19-35 months. Therefore, information was requested for the following immunizations for children aged 19-35 months: (1) four doses of Diphtheria, Pertussis and Tetanus (DPT); (2) three doses of Poliovirus Vaccine (OPV); (3) three doses of Haemophilus Influenza type B (Hib); and (4) one dose of any Measles containing vaccine (Measles, Mumps, and Rubella (MMR)). Data were entered by a research assistant into customized screens designed using dBase VI_. Analyses were made using SAS_. Data from all children were used to examine prevalence and treatment of OM. Dates of immunization were collected to determine if age-appropriate vaccinations were obtained. A range of time was allowed for each vaccination.

RESULTS

Of the total sample, a total of 54 (36 percent) were diagnosed with OM at least once. Nineteen children were diagnosed with OM at two different doctors visits and nine received a diagnosis of OM at three different doctors visits. Of the 54 children with one doctor_s visit for OM, all were treated with antibiotics. Eighteen of the 19 children who were diagnosed with OM twice were treated with antibiotics; and nine of nine children with three occurrences of OM were treated with antibiotics. The most common antibiotic prescribed was Amoxicillin. No children had permanent hearing loss. Immunization status for children aged 19-35 months was examined in two ways in accordance with published CDC standards. Percentages in this section were rounded to the nearst whole number to conform with the CDC format. Approximately 83 percent of children received four age-appropriate doses of DPT. Children received age-appropriate vaccinations at a rate of approximately 88 percent for Polio and 85 percent for Hib. A total of 92 percent received the MMR vaccination age-appropriately. Eighty-two percent of the children received all three doses of Hepatitis B. Among the series of vaccinations examined, 81 percent of the children in the sample received the combined 4

DPT/3 Polio/1 Measles. Among children who were born after 1992 (Hib was not widely available before this), 82 percent received the 4 DPT/3 Polio/1 Measles/3 Hib series. Children who were under 19 months of age were eliminated from these analyses.

DISCUSSION

Children admitted to the CVC programs obtained or exceeded National and California rates for the 4:3:1 series (4 DPT, 3 Polio, 1 Measles) and the 4:3:1:3 (4 DPT, 3 Polio, 1 Measles, 3 Hib) vaccination series. For the 4:3:1 series, the CVC achieved an 81 percent completion rate compared to the 78 percent National and California rates. The CVC also achieved or exceeded U.S. and California NIS rates for DPT, Measles and Hepatitis B. The U.S. NIS rates for 3 doses of DPT were 95 percent compared to 94 percent in California and 95 percent in the CVC. Four DPT doses were obtained by 81 percent in California and the U.S. compared to 83 percent in the CVC. The rate of Measles dose vaccination was 92 percent in the CVC, 90 percent in the U.S. and 91 percent in California. Conversely, the U.S. and California NIS rates for OPV vaccine were 91 percent; however, OPV was only received by 88 percent of CVC children. Also, Hib vaccination rates were 90 percent and 92 percent in California and the U.S., respectively, compared to 85 percent in CVC children.

Table 1. Age-Appropriate CVC Immunizations with Comparisons to Childhood Immunization Initiative (CII) Goals, 1992 National Health Interview Survey (NHIS) Results, and 1996 NIS Results.

Vaccine/Dose	CVC Sample (n=151)	CII 1996 Goal	1992 NHIS National Results	1996 NIS National Results	1996 NIS California Results
DPT/DT >3 doses >4 doses	95% 83%	90%	83% 59%	95% 81%	94% 81%
Poliovirus >3 doses	88%	90%	72%	91%	91%
Hib >3 doses	85%	90%	28%	92%	90%
Measles (MMR) >1 doses	92%	90%	83%	91%	90%
Hepatitis B >3 doses	82%	70%	8%	82%	82%
Combined Series 4 DPT/3 Polio/ 1 Measles 4 DPT/3 Polio/ 1 Measles /3 Hib*	81% 82%	-   -	55%	78% 77%	78% 76%

*Because of availability of this vaccine and time date of the NIS, only children whose year of birth was 1993 or later were included in this estimate.

The CVC compares favorably with California and the U.S. for series of vaccinations, but not as impressively for individual vaccinations such as OPV and Hib. One explanation for this was the difficulty obtaining sufficient supplies of vaccines. For example, requests made in September were still not available in November. Clearly, it is easier for parents to bring children to the clinic and obtain all vaccines recommended for that age (i.e., at 2 months receive DPT, OPV and Hib), rather than have children return because one or more vaccines were unavailable. The second outcome was to measure the rate at which children obtained antibiotic therapy for OM. This outcome measure was suggested by the BPHC Working Group on Homeless Health Outcomes. It was selected because OM is a common problem among children, and if left unattended, has a potentially lifelong impact on hearing. The BPHC recommends antibiotic treatment for diagnosed OM. A total of 54 children had been diagnosed with OM at least once, but several children had OM more than once. Eighty-two incidences of OM were recorded and 81 were treated with antibiotics for an almost 99 percent treatment success rate. The threshold goal for OM was that 95 percent of children would receive antibiotic treatment. At the Ambulatory Care Services clinics, parents are provided with the antibiotic, not just the prescription. Providing the antibiotic is preferable since it increases the likelihood of it being administered. No children were identified with hearing loss secondary to ear infections; however, with such a small sample size this occurrence would have been unlikely. Because there is some disagreement on whether antibiotic therapy for OM in children is indicated, this outcome measure may not be ideal for pediatric populations. Recent studies have found that in some cases, antibiotic treatment may not be any more effective than a placebo and that inappropriate and that frequent use may increase the development of drugresistant pathogens. Therefore, although the CVC achieved its threshold level for antibiotic treatment of OM, there is a question as to whether or not antibiotic treatment for each OM reflects overuse of antibiotics. Process measures are very important. For example, through use of outcome measures only there is no indication of how many contacts by case managers, nurses and doctors were made to obtain this high rate of age-appropriate immunizations and treated OM. While children who are stably housed and have a designated primary care pediatrician receive immunizations and antibiotic treatment for OM as a matter of routine care, several sub-populations of children who live in poverty, including children who are homeless or at-risk for homelessness, are at higher risk for physical problems and under immunization (Alperstein, Rappaport & Flanigan, 1988; Bassuk, 1990; Miller & Lin, 1988; Parker et al., 1991). In retrospect, many improvements could be made to the instrument. First, it should have included information on gender and ethnicity. In addition, since children have recurrent OM, the instrument should have been designed to accommodate more than three entries on the subject. Furthermore, it would be helpful to know whether each client visit was for urgent care or ambulatory care since children who were treated in urgent care may have been more likely to have incomplete immunizations. In addition, although the instrument was devised to be explicit and thereby employ little judgment on the part of the abstracter, it would have been better to employ a non-biased research assistant to perform the audit. Last, since this study population consists only of CVC clients who used ambulatory care services, the generalizability to other HCH projects is limited.

CONCLUSION

Of the two outcome measures selected for this pilot project, immunization rates were a commonly measured pediatric preventive outcome used by multiple different Federal agencies, childhood initiatives, and health maintenance organizations. It is a reasonable outcome measure that could easily be adopted into an HCH program that targets children. On the other hand, however, based on newly emerging research regarding overuse of antibiotics prescribed for OM, antibiotic therapy for OM may not be an ideal acute care outcome measure.