Evaluation of state-provided Postexposure Prophylaxis against Rabies in Florida.

ABSTRACT

Background. One key component of human rabies prevention is the use of postexposure prophylaxis (PEP). No information is routinely available to evaluate the appropriateness of this intervention.

Methods. A survey queried county health departments regarding rabies PEP administered during July through September in 1997 and in 1998.

Results. A total of 160 cases meeting the study criteria were characterized. Domestic animals accounted for most potential rabies exposures. Nearly a quarter of the reports indicated that PEP was not necessary. Health departments collected just over 25% of costs for antirabies biologics.

Conclusions. Human rabies prevention systems can be enhanced by reducing the number of PEP events through improved recovery of biting animals. Inappropriate administration of PEP and cost can be reduced by establishing a system of after-the-fact review of every episode of PEP with timely feedback to providers.

Although animal rabies is considered endemic in Florida, no human cases of rabies due to animal exposures have been reported within the state since 1947 (two human rabies cases in Florida reported in 1994 and 1996 were due to dog bites received in Haiti and Mexico, respectively). The rarity of this disease in people is considered largely to be the result of animal control efforts, vaccination of pet animals, and appropriate rabies PEP. This study was conducted to evaluate the appropriateness of PEP intervention.

The Florida Department of Health (DOH) spent more than $2.5 million during 1997 and 1998 for the components of PEP (rabies vaccines and human rabies immune globulin [HRIG]) (Table 1). The PEP regimen for a previously unvaccinated person consists of HRIG administered at 20 international its per kilogram of body weight given one time, and a series of five rabies vaccines administered over a period of approximately 1 month. In Florida, these rabies biologicals are made available by the DOH Central Pharmacy through 23 county "repositories." The rabies repository structure was developed two decades ago for quick availability of product to surrounding county health departments from central regional locations during a time in which these products were available only through government health agencies. Today, in addition to these repositories, some hospitals and private providers purchase their own biologicals.

Emergency room physicians and county health department staff generally make the decisions regarding PEP usage. Some repositories fill prescriptions by local physicians for PEP without reviewing the circumstances of the exposure (as they would also do for other drugs). To assist a varied audience with rabies control guidelines (both human and animal), DOH provided a manual to county health departments for further local distribution during October 1997. Second, third, and fourth editions were provided during June 1998, January 1999, and January 2000, respectively. (1) Additionally, an animal bite management "decision tree" was developed and made available to county health departments during April 1998 to share with emergency room staff and private practitioners as appropriate. Finally, veterinary and medical public health staff are available for consultation around the clock.

An evaluation of practices for human rabies control and prevention is challenging, since a variety of county health department personnel as well as outside agencies are involved with individual tasks. At the time of this study, data regarding animal bites and the use of PEP were not routinely collected at the state level. A comprehensive evaluation of the rabies/animal bite program might begin with ascertainment of all animal bites in a selected timeframe, followed by PEP use/appropriateness assessment (Figure). However, a pilot review with county health department environmental health and animal control staff showed that a comprehensive animal bite database including all people who received PEP is lacking in many counties. At one site, only 2 of 20 people who received PEP had a corresponding animal bite report among those reviewed at the animal control agency.

The DOH Central Pharmacy provides the Bureau of Epidemiology monthly data regarding the amount of biologicals ordered for each of the state's 23 rabies vaccine repositories. Three main limitations of this system are that it does not account for biologicals that are purchased by private providers, it does not provide information about individual cases (eg, reasons for PEP and completeness of PEP administration), and it does not address wastage rates.

Repository sites are invoiced by Central Pharmacy for issuance of biologicals. Repository sites attempt to recover the costs from private insurance, patients, and Medicaid or Medicare. Reimbursement data are unavailable at a centralized level, since rabies biologicals are not specifically coded for in the aggregate funds submitted directly to DOH Revenue Management by the county health department.

To evaluate the effectiveness of the DOH-sponsored human rabies prevention program in Florida, we attempted to answer the following questions: Are people who were given rabies PEP from the DOH Central Pharmacy receiving it appropriately (based on the algorithm provided in the rabies decision tree)? Did PEP decision-making "improve" when comparing 2 consecutive years, before and after distributing rabies control and prevention materials to the county health departments? How much of the cost for rabies PEP was recovered from insurance companies and patients by the county health departments? Can we modify the present system to improve efficiency and save money?

METHODS

Of the 67 county health departments in Florida, 22 were selected at random for this study. We asked the departments to review medical records for all people who began receiving PEP from the DOH Central Pharmacy (Figure, path A) at any time during two specific 3-month periods, July through September 1997 and July through September 1998. We expected approximately 200 people to meet the review criteria. A two-page survey instrument was provided for data collection. Medical record review included PEP pharmacy logs, animal bite reports, patient records, and billing information.

Epi Info Version 6 was used for summary statistics. Security and confidentiality were maintained during all steps of the study process.

RESULTS

A total of 199 survey instruments were submitted from the 22 counties. Of those, 160 (80%) from 15 of the counties met the study criteria (Table 2).

Sex was recorded on 152 reports (95%), of which 53% indicated male. Age was specified on 144 reports (90%); 26% were 1 to 17 years old, 24% were 18 to 33, 25% were 34 to 46, and 25% were 47 to 85. Median reported age was 33.5 years. Fourteen reports (9%) indicated that the person had been previously immunized against rabies.

Information on type of exposure was available in 151 reports (94%). Animal bite was listed for 107 (71%) of the cases. Among the 44 nonbite exposures, saliva or "potential saliva exposure" was listed for 22 (50%), scratch or possible scratch for 14 (32%), splashed with blood for 2, and touching the animal for 6. No statistical differences were observed in the type of animal exposure by sex or age of the patient.

The type of animal was provided in 152 reports (95%). Domestic animals accounted for two thirds (n = 99) of the reports (though only one third of those were listed as being owned). Dogs (including one wolf/dog hybrid) were listed most commonly (n = 63), followed by cats (36), bats (18), raccoons (18), foxes (5), bobcats (3), opossums (2), coyote (1), monkey (1), squirrel (1), and prairie dog (1). Multiple "exposures" to the same animal occurred on 10 occasions, 3 involving dogs (groups of 7, 4, and 2 people), 3 involving bats (3 groups of 2 people), 2 involving cats (groups of 4 and 2 people), I involving a bobcat (3 people), and 1 involving a raccoon (2 people). Thirty-eight of the reports indicated that the animal was tested for rabies; results were positive in 19 animals (5 bats, 4 dogs, 3 bobcats, 3 foxes, 3 raccoons, 1 cat). In 10 reports (9 dogs, 1 cat), the rabies tests were unsatisfactory, and the animal was therefore considered positive for the purpose of PEP determination. In 9 cases, the animal te sted negative. Excluding the 38 tested for rabies, the status of the animal was recorded for 75: 66 had escaped (45 of which were dogs or cats), 6 (dogs and cats) were observed, and 3 had been buried and were too decomposed for rabies testing. In the remaining 47 reports (39 of which indicated type of animal and 8 did not), management of the animal was not indicated.

Among the 122 reports in which the animal was not tested, animal control was consulted for 66 (54%) of the incidents and was not contacted in 18 (15%); the information was missing for 38 (31%). Domestic animal exposure comprised nearly half (n = 25) of the cases for which animal control involvement was not known or did not occur.

Regarding the decision to administer PEP, information was complete for 146 (91%) of the case reports. For these 146, applying the established algorithm for PEP decision-making to available data shows that decisions were appropriate for 104 (71%) and "inappropriate" for 32 (22%). For the remaining 10 (7%.), the decision to administer PEP was made outof-state, and the person was continuing therapy in Florida. Among the 32 cases in which PEP was considered unnecessary, no rabies exposure occurred in 15 (reported as "possible exposure" or "touched animal"), the animal tested negative for rabies in 7, the dog or cat was available for observation in 6, the animal was at low risk for rabies in 2 (squirrel or opossum), and in 2 it was possible to euthanize and test the wild pet for rabies in lieu of giving PEP to the person. Multiple exposures to the same animal and for which PEP was inappropriate occurred during 4 occasions: 7 people with no exposure to a dog suspected of being rabid (shipping problems caused the b rain tissue to be unsatisfactory for testing); 1 person whose wife was exposed to a raccoon, though he was not; 2 people who touched a bobcat; and 4 people exposed to a cat that tested negative for rabies. The decision-makers for these 32 cases were private physicians (n = 7), county health department staff (n = 6), and veterinarians (n 4); in the other 15 reports, this entry was left blank.

Excluding cases for which the decision to begin PEP was made out of state or the decision information was incomplete, dates of exposure and treatment were available for 93 (70%) of the remaining 132 case reports. Overall, 29% of the individuals received PEP within 24 hours of exposure (27% of the "PEP appropriate" group and 33% of the "PEP inappropriate" group). An additional 30% received PEP within 2 to 5 days after exposure; 20% began receiving PEP 6 to 10 days after exposure; and 21% first received PEP 11 to 60 days after exposure.

Eighty-five reports were identified during the 1998 study period compared with 75 during the 1997 period. Excluding incomplete and follow-up cases from other states, the proportion of inappropriate decisions to administer PEP was greater in the 1998 study period than in the 1997 period (31% versus 16%), but the difference was not statistically significant (P= .094).

Cost of PEP treatment was documented for 129 (80%) of the reports. Reported PEP costs ranged from $84 to $1,461 per patient and included biologicals and, for county health departments, cost for administering the biologicals. Unknown costs include health department investigation and consultation. In all, county health departments billed $97,225 for PEP, of which $25,647 (26%) was documented as being reimbursed. Type of payer was documented for 118 (74%) of the reports. Patients were billed in 54 cases, insurance in 51, Medicaid or Medicare in 9, and "indigent program" in 4. In several cases, the billed amount did not reflect the actual cost of treatment (eg, patients were billed on a sliding scale, or the county health department did not charge for some or all of the biologicals). For 27 (21%) of the 129 bills, the bill was paid in full (19 by insurance companies and 8 by patients). Reasons for nonrecovery of costs were not requested in this survey; however, comments indicated that the county health departmen t had not yet sent Out the bill; the insurance, Medicaid/Medicare, or patient refused to pay; the county health department sends out three letters, after which no further action is taken and/or the private provider told the individual that this is a free service from the state.

DISCUSSION

During the 1990s, national decision-making guidelines for human rabies prevention were developed by the Advisory Committee on Immunization Practices and published by the Centers for Disease Control and Prevention. (2,3) Clearly, if a person is exposed to a known rabid animal, antirabies treatment is indicated. Decision-making becomes more difficult to the degree that exposure is less well identified than a bite and when the animal is unavailable for testing or observation as appropriate. Given the grave outcome of rabies disease in people and pressure from the victim or victim's family, there is an incentive to make conservative decisions and provide rabies PEP in uncertain situations. Although nonbite exposure assessments are time-consuming, they are necessary to help elucidate the probability that infectious saliva or nerve tissues entered a wound or mucous membrane.

It appears that emergency room physicians frequently render decisions for rabies PEP. Animal bites are physically and emotionally traumatic, but antirabies therapy is an urgency not an emergency. If the animal is available for testing or observation as appropriate, the victim does not need to begin rabies PEP immediately. Nearly one third of the people in this review received rabies PEP within 24 hours of their exposure, and two thirds did not. A stronger network of administering physicians working with county health departments and/or animal control could have resulted in fewer antirabies treatments simply by waiting for the results of the animal rabies test or for confinement and observation of the animal. (4) Education of the decision-maker (physician and/or county health department staff) is critical to reducing the number of inappropriately administered courses of PEP. Conversely, several people received PEP more than 10 days after exposure (reasons for the delay were not ascertained).

Domestic animals accounted for two thirds of the exposures. A large proportion of these were unavailable for confinement and observation (animals escaped or were not pursued). Although the specific data were insufficient to determine the circumstances of these incidents, certainly removal of stray dogs and cats, maintaining pets on leashes, and eliminating the feeding of strays would reduce the number of potential rabies exposures. In some counties, obstacles include lack of animal control services and community resistance to controlling feral cats.

In Florida, rabies control program activities have historically been the responsibility of the health department. Through the decades--depending on county resources--some of these activities have been shifted to other organizations/agencies that were better equipped to meet specific burdens. Additionally, in some county health departments, rabies control activities are spread among environmental health, epidemiology, nursing, and immunization staff. As a result, a variety of programs exist, from complete organizational oversight and maintenance of rabies prevention, to complete delegation of responsibilities to non-DOH agencies. Evaluating such diverse program structures is challenging. In this review, retrieving documentation about circumstances of the exposure, administration of antirabies treatment, and billing was more demanding than anticipated. In some cases, local records were not detailed or complete enough to include in the analyses.

A DOH "decision tree" developed during 1998 (distributed to Florida county health departments to further disseminate) concisely outlined an algorithm for clinicians making rabies prophylaxis decisions. (*) Applying the rabies PEP management algorithm to the information provided in this review revealed that PEP decisions were inappropriate in nearly one fourth of the cases for which decision data were complete. Half of these reports did not indicate who the decision-maker was. For the others, private physicians, county health department staff and veterinarians accounted for the decisions. In one cluster of 7 people, an error in shipping a dog's brain to the laboratory caused a situation in which the county was compelled to provide PEP despite no confirmed history of rabies exposure. In a 1998 survey of rabies repository sites, most county health department staff believed that PEP was generally given appropriately, with a few notable exceptions? Liability issues and recent human rabies cases with "cryptic bat e xposures" were thought to skew decision-making toward administering PEP, perhaps even when not indicated. This study design was not able to ascertain how frequently PEP was not given when circumstances indicated it was appropriate.

No improvement of appropriately administered PEP was detected when comparing data from the two study periods. We cannot be sure that educational training session information and other materials on rabies were distributed to or reviewed by rabies PEP decision-makers.

A statement in the 1947 Florida State Board of Health Annual Reports (6) laments that the issuance of antirabies treatments to 1,434 people that year for $6,094.50 was costly. Indeed, that sum did not even "include vaccine bought by practicing physicians." (6) In this review, for the 129 cases in which financial data were available, county health departments billed nearly $100,000. Considering average distribution of rabies biologicals alone, we expected billing to be closer to $140,000 for 129 people. Reasons for the difference may be that patients considered indigent were billed on a sliding scale, patients were preimmunized and required no HRIG and fewer vaccines, or patients were receiving a vaccine in their series while they were vacationing in Florida. Among those who received PEP when not indicated, the billed amount was nearly $20,000 (estimated $75,000 statewide).

In a 1998 survey of rabies repositories, staff estimated that the majority of the cost for PEP was reimbursed by insurance companies or by the patient. (5) In this review, slightly more than one quarter of the billed amount was documented as being reimbursed to the county health departments. The assumption that provision of rabies biologicals is a free service from the state prevailed, as well as a historical context that insurance or Medicaid/Medicare will not reimburse for antirabies treatment. In fact, the DOH Central Pharmacy will bill Medicaid on behalf of the county health departments, and many insurance companies will reimburse when pursued. Finally, manufacturers of rabies vaccine have indigent programs whereby a physician can request complimentary biologicals if the patient meets certain criteria. Through the years, the health department has provided prophylaxis or treatment for selected infectious diseases. While rabies PEP has been available from the state through the repository system, private phy sicians have also had access to and have been providing the biologicals. Biologicals are readily available using overnight shipping, which did not exist when the repository system was established 20 years ago. Several county health departments noted that certain hospitals in their area are stocking and administering the PEP series or beginning the series and referring the patient to the county health department to continue treatment. One county refers all potential rabies exposure victims to their local hospitals.

This latter practice is comparable to that of the Alabama Department of Health, which has elected not to participate in providing rabies PEP (oral communication, William Johnston, DVM, September 1999). Private providers acquire rabies biologicals directly from the manufacturers, and the department is involved only as needed for consultation. In another model, the New York State Department of Health will release PEP free of charge to physicians who complete a PEP report form after review of the exposure circumstances ((oral communication, Millie Eidson, DVM, September 1999).

While it is important to maintain county health department flexibility, it is evident that the current rabies prevention system can be modified. Since many of the exposures in this review were attributed to domestic animals, strong local animal control activities should be supported, including removal of strays and adherence to leash laws (for cats as well as dogs), animal bite prevention education, and preimmunization of persons at high risk for rabies exposure. Additionally, capturing and observing at-large domestic animals would reduce the number of people required to receive PEP. Emergency room physicians and new county health department staff should be given the tools for making appropriate decisions regarding PEP administration, including the 1999 ACIP guidelines and the decision tree. Contact with animal control staff should be compulsory to assess the status of the biting animal.

Evaluation of human rabies prevention activities would be considerably improved if the data elements were collected and maintained in a centralized area at the county health department. Standards for evaluation can be developed with county health department participation. As a result of this study, county health departments are required to document when DOH-provided rabies biologicals are administered. In conjunction with DOH Central Pharmacy staff, developing procedural guidelines for billing and record keeping to assure reimbursement of rabies costs is recommended. Implementation of these recommendations will improve the amount of appropriate PEP offered and can be expected to save in excess of $220,000 annually in pharmaceutical costs from Florida's rabies prevention program.

Acknowledgments. We thank the county health departments and animal control staff, whose contributions made this study possible.

From the Florida Department of Health, Tallahassee.

Reprint requests to Lisa Conti, DVM, MPH, 4052 Bald Cypress Way, BIN A-12, Tallahassee, FL 32399-1720.

(*) Adapted from Istre GR, Emerson JK, Hopkins RS: Critical Decisions in Trauma. St. Louis, CV Mosby Co, 1984, pp 484-487

References

(1.) Rabies Prevention and Control in Florida. Tallahassee, Florida Department of Health, 1999

(2.) Centers for Disease Control and Prevention: Rabies Prevention--United States, 1991. Recommendations of the Immunization Practices Advisory Committee (ACIP). MMWR 1991;40:R-3:1-19

(3.) Centers for Disease Control and Prevention: Human Rabies Prevention -- United States, 1999. Recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR 1999; 48:R-1:1-21

(4.) Helmick CC: The epidemiology of human rabies postexposure prophylaxis, 1980-1981. JAMA 1983;250:1990-1996

(5.) Telephone Survey of Emergency Drug Distribution Centers for Human Rabies Biologics in Florida--Understanding the System. Survey results presented to Bureau of Epidemiology and Central Pharmacy Director, July 28, 1998

(6.) Wilson SW (Florida State Health Officer): 48th Annual Report, Florida State Board of Health. Jacksonville, 1947, p4

TABLE 1

Department of Health (DOH) Central Pharmacy Distribution Activity,
Florida, 1997-1998

                                   Mean No.               Possible No.
                   Total Vials     of Vials                 of People
                   Distributed    Distributed  Mean Cost    Given PEP
Biologicals       1/97-12/98 (*)   per Month   per Month  per Month (+)

Human rabies          5,249           228      $27,9232
 immune globulin
 (HRIG) (2 mL)
HRIG (10 mL)          1,180           51        $25,525
Rabies vaccine       12,363           537       $62,027     107 (**)
                                               $115,484

PEP = Postexposure prophylaxis.

(*)July 1997 data not included per DOH Central Pharmacy.

(+)537 doeses of vaccine/month divided by 5 (vaccines per person for
PEP).

(**)Approximately 1,300 people given PEP/year at a cost of $1.4 million
in biologicals, assuming no increase in stocked vaccine.
TABLE 2

Number of Postexposure Prophylaxis (PEP) Cases Expected and Received in
Florida

                No. of  Expected No.
County          Cases   of Cases (*)

Alachua (+)       16         8
Baker              0         1
Citrus             6         4
Collier (+)       22         8
Dade (+)           3        72
Dixie              4         0
Franklin           0         0
Gadsden            0         2
Gilchrist          0         0
Glades             0         0
Hamilton           0         1
Highlands (+)      3         3
Indian River       4         4
Jackson            3         2
Leon (+)          23         9
Okaloosa           2         6
Palm Beach (+)    27        35
Pinellas (+)      33        31
St. Lucie          1         6
Sarasota (+)      11        12
Taylor             2         1
Wakulla            0         1
Total            160       206

(*)Based on county population and random vaccine distribution within the
state.

(+)Rabies repository sites.

RELATED ARTICLE: KEY POINTS

* Emergency department physicians frequently render decisions for rabies postexposure prophylaxis (PEP). They should be well trained and have ready access to public health staff for rabies consultation.

* Human rabies prevention systems can be enhanced by reducing the number of PEP events through improved recovery of biting animals for rabies testing or observation.

* Inappropriate administration of PEP and overall costs can be reduced by establishing a system of after-the-fact review of every episode of PEP with timely feedback to providers.