CHILDHOOD IMMUNIZATIONS

Tailoring Interventions with Analysis and Feedback

Childhood immunizations have resulted in the total or near total eradication of suffering associated with all nine diseases recommended for universal childhood immunization prior to 1990. [1]

A statewide study showed that, in 1995, only 61 percent of two-year-olds in the plan's home state were appropriately immunized. A statewide effort to increase immunization rates was begun in 1996. With a large volume of children in both commercial and Medicaid populations (5.3 percent of its population was under age 2), the plan chose improving immunization rates as a meaningful activity.

The plan used HEDIS specifications to identify the target population, and for its performance measures. The targeted population was all children in commercial and Medicaid products who turned two-years-old during the reporting year.

The performance measure was the number of children in the target population who had received all of the following:

• At least 4 DTP immunizations, or an initial DTP immunization followed by at least 3 DT immunizations
• At least 3 OPV immunizations
• At least one Measles immunization
• At least one Mumps immunization
• At least one Rubella immunization
• At least one H. influenza type b immunization

(After baseline measurement, HEDIS specifications changed, and the plan added two hepatitis B immunizations to its combination rate.)

In addition, the plan calculated individual immunization rates, and a combination rate including two H. influenza and one VZV (chicken pox) immunization.

The plan used the hybrid method of collecting data from both medical records and claims. It found it needed to retrieve more than 90 percent of the data from the medical record. For the commercial members, the plan used a random sample of 384 children. The entire Medicaid eligible population was included in the study.

Baseline performance, based on 1995 data, showed an immunization rate of 68.6 percent. The plan identified a benchmark rate from the 1996 edition of NCQAs Quality Compass. The best performing plan in the country had reported a rate of 92.9 percent. Based on this benchmark, and the 90 percent goal contained in the U.S. Public Health Services publication Healthy People 2000, the plan selected 90% as its goal.

FACILITATING DATA COLLECTION The plan used registered nurses to gather data from the medical record. When possible, the nurse coordinated data collection on immunizations with ambulatory medical reviews, office site visits, and other HEDIS data collection activities. The plan developed forms that could be read by a scanner. This automated the data entry process and made data extraction in the physician office more efficient.

A multidisciplinary group, guided by the literature available on immunization, identified possible barriers to immunization. This group included PCPs. The identified barriers included:

• Missed opportunities (when a child visits a health care provider, but does not receive all needed immunizations) due to:
• Failure to assess immunization status
• Failure to administer simultaneous immunizations
• Use of invalid contraindications
• The complexity of the immunization schedule
• Parental barriers, including:
• Perceived lack of risk of disease
• Fear of adverse reactions to vaccines
• Lack of awareness
• Access barriers

Prior to 1995, the plan had begun some general forms of intervention. It had sent birthday card reminders for two-year-olds to parents) and preventive health guidelines and immunization tracking forms to physicians. These had little impact on immunization rates. The plan began to address individual member needs with more focused interventions.

In addition to continuing parental and provider education, the plan participated with hospitals, vaccine manufacturers, the state health department and other plans in targeting practitioners with a tool kit. This kit included:

• An introductory letter explaining statewide immunization rates and the need for improvement
• Standards of practice
• Immunization tracking flow sheet for the medical record
• Practitioner self-assessment test
• A resource guide summarizing recommendations from recognized authorities
• A vaccine adverse event reporting form, with instructions
• Guide to contraindications
• A wall chart

The plan sent out tool kits to 370 pediatricians and family practitioners.

It followed this with monthly lists of practitioners' 20-month-old patients due for immunizations. The lists included a matching set of stickers to flag charts of these members.

IDENTIFYING A PCP One obstacle encountered was identifying the correct PCP at the time of immunization. The plan used enrollment and claims systems to create a PCP history profile for each member of the target population. This mapping program helped nurses track down the correct PCP and medical record.

The birthday card reminder system was changed to a postcard reminder system reminding parents of immunizations needed within the next month for their two-, six- and 12-month-olds. Over 4,400 cards were sent.

The first remeasurement, based on 1996 data, showed an immunization rate of 59.9 percent. Although lower than baseline, this measurement was not comparable because of the inclusion of the requirement for two hepatitis B immunizations, and the change in continuous enrollment criteria from HEDIS 2.5 to HEDIS 3.0. For these reasons, 1996 represented a new baseline for future comparisons.

INTER-RATER RELIABILITY Extensive training was needed to ensure inter-rater reliability for the nurses retrieving medical record data. They received detailed instructions on HEDIS specifications, and were given practice case studies. Also helpful-getting input from the reviewers on the design of the data collection forms.

A marked difference was observed between immunization rates for commercial members (65.9 percent) and Medicaid members (48.8 percent).

Reviewing the literature, the plan identified additional barriers for the Medicaid population:

• Possible language barriers
• Lack of general knowledge about immunizations
• Distrust of health systems
• Lack of transportation

The plan implemented a transportation program for its Medicaid members. Based on need, members were provided transportation to covered services, such as immunizations.

Mailings to providers now included provider-specific immunization rates. The plan set aside money for a physician incentive program, and immunizations were included as one measure. Performance was measured from the 1996 baseline.

Reminder postcards were expanded to include a fourth postcard at 18 months. Under a pilot program, a plan representative called the parents of underimmunized children turning two years old. Parents were reminded to check their child's immunization status and to call their doctor for an appointment.

Remeasurement, based on 1997 data, showed an immunization rate of 66.5 percent, with a 74.2 percent rate for commercial members, and a 55.8 percent rate for Medicaid members.

The plan analyzed the data by individual vaccine and product line. Significant improvement was seen in DTP, polio, and hepatitis B immunizations in both commercial and Medicaid populations. The commercial population also showed a statistically significant increase in H. influenza immunization. No significant changes were noted in the rates for MMR or for the H. influenza rate among Medicaid members.

Although the plan had made progress, it had not achieved its goals. The plan continued to identify barriers and develop strategies to further improve compliance with childhood immunizations.

In early 1998, the plan sent a survey to 109 pediatricians and family practitioners to assess whether the previous mailings to providers had been helpful (included in Appendix 36). Forty-six percent responded. Physician feedback confirmed the desire to continue receiving the mailings. They also recommended that reports of members needing immunizations not only be sent at 20 months of age, but at age 12 months as well.

FOUNDATIONAL EFFORTS The health plan found the telephone outreach program to be so successful, it expanded it to other initiatives such as diabetes care. Members liked the telephone reminder, and they took the time to ask questions. The calls occurred during the day so as not to interrupt the evening hour. If an answering machine was encountered, a message was left.

The plan removed immunizations from the capitation model. Billing for immunizations increased physician attention and improved claims data.

The plan added a Web site and e-mail program to send information and reminders about preventive health issues.

It fully implemented the telephone outreach program it had piloted.

Subsequent measurement in 1998 showed an immunization rate of 82.7 percent, and a rate of 73.8 percent in 1999. Although 1999 showed a slight drop, the rate remained higher than those reported in 1997 and before.

The plan has continued to participate in statewide initiatives, as well as identifying potentially more effective interventions for its Medicaid population.

[1] - Department of Health and Human Services, Centers for Disease Control and Prevention, "Achievements in Public Health, 1900-1999 Impact of Vaccines Universally Recommended for Children-United States, 1990-1998," Morbidity and Mortality Weekly Report, 48 no. 12 (Atlanta, GA: Department of Health and Human Services, Centers for Disease Control and Prevention, April 2, 1999), 243-248.